Source

emacs / src / coding.c

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
6133
6134
6135
6136
6137
6138
6139
6140
6141
6142
6143
6144
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
6164
6165
6166
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
6210
6211
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
6261
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
6629
6630
6631
6632
6633
6634
6635
6636
6637
6638
6639
6640
6641
6642
6643
6644
6645
6646
6647
6648
6649
6650
6651
6652
6653
6654
6655
6656
6657
6658
6659
6660
6661
6662
6663
6664
6665
6666
6667
6668
6669
6670
6671
6672
6673
6674
6675
6676
6677
6678
6679
6680
6681
6682
6683
6684
6685
6686
6687
6688
6689
6690
6691
6692
6693
6694
6695
6696
6697
6698
6699
6700
6701
6702
6703
6704
6705
6706
6707
6708
6709
6710
6711
6712
6713
6714
6715
6716
6717
6718
6719
6720
6721
6722
6723
6724
6725
6726
6727
6728
6729
6730
6731
6732
6733
6734
6735
6736
6737
6738
6739
6740
6741
6742
6743
6744
6745
6746
6747
6748
6749
6750
6751
6752
6753
6754
6755
6756
6757
6758
6759
6760
6761
6762
6763
6764
6765
6766
6767
6768
6769
6770
6771
6772
6773
6774
6775
6776
6777
6778
6779
6780
6781
6782
6783
6784
6785
6786
6787
6788
6789
6790
6791
6792
6793
6794
6795
6796
6797
6798
6799
6800
6801
6802
6803
6804
6805
6806
6807
6808
6809
6810
6811
6812
6813
6814
6815
6816
6817
6818
6819
6820
6821
6822
6823
6824
6825
6826
6827
6828
6829
6830
6831
6832
6833
6834
6835
6836
6837
6838
6839
6840
6841
6842
6843
6844
6845
6846
6847
6848
6849
6850
6851
6852
6853
6854
6855
6856
6857
6858
6859
6860
6861
6862
6863
6864
6865
6866
6867
6868
6869
6870
6871
6872
6873
6874
6875
6876
6877
6878
6879
6880
6881
6882
6883
6884
6885
6886
6887
6888
6889
6890
6891
6892
6893
6894
6895
6896
6897
6898
6899
6900
6901
6902
6903
6904
6905
6906
6907
6908
6909
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
6931
6932
6933
6934
6935
6936
6937
6938
6939
6940
6941
6942
6943
6944
6945
6946
6947
6948
6949
6950
6951
6952
6953
6954
6955
6956
6957
6958
6959
6960
6961
6962
6963
6964
6965
6966
6967
6968
6969
6970
6971
6972
6973
6974
6975
6976
6977
6978
6979
6980
6981
6982
6983
6984
6985
6986
6987
6988
6989
6990
6991
6992
6993
6994
6995
6996
6997
6998
6999
7000
7001
7002
7003
7004
7005
7006
7007
7008
7009
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
7026
7027
7028
7029
7030
7031
7032
7033
7034
7035
7036
7037
7038
7039
7040
7041
7042
7043
7044
7045
7046
7047
7048
7049
7050
7051
7052
7053
7054
7055
7056
7057
7058
7059
7060
7061
7062
7063
7064
7065
7066
7067
7068
7069
7070
7071
7072
7073
7074
7075
7076
7077
7078
7079
7080
7081
7082
7083
7084
7085
7086
7087
7088
7089
7090
7091
7092
7093
7094
7095
7096
7097
7098
7099
7100
7101
7102
7103
7104
7105
7106
7107
7108
7109
7110
7111
7112
7113
7114
7115
7116
7117
7118
7119
7120
7121
7122
7123
7124
7125
7126
7127
7128
7129
7130
7131
7132
7133
7134
7135
7136
7137
7138
7139
7140
7141
7142
7143
7144
7145
7146
7147
7148
7149
7150
7151
7152
7153
7154
7155
7156
7157
7158
7159
7160
7161
7162
7163
7164
7165
7166
7167
7168
7169
7170
7171
7172
7173
7174
7175
7176
7177
7178
7179
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
7224
7225
7226
7227
7228
7229
7230
7231
7232
7233
7234
7235
7236
7237
7238
7239
7240
7241
7242
7243
7244
7245
7246
7247
7248
7249
7250
7251
7252
7253
7254
7255
7256
7257
7258
7259
7260
7261
7262
7263
7264
7265
7266
7267
7268
7269
7270
7271
7272
7273
7274
7275
7276
7277
7278
7279
7280
7281
7282
7283
7284
7285
7286
7287
7288
7289
7290
7291
7292
7293
7294
7295
7296
7297
7298
7299
7300
7301
7302
7303
7304
7305
7306
7307
7308
7309
7310
7311
7312
7313
7314
7315
7316
7317
7318
7319
7320
7321
7322
7323
7324
7325
7326
7327
7328
7329
7330
7331
7332
7333
7334
7335
7336
7337
7338
7339
7340
7341
7342
7343
7344
7345
7346
7347
7348
7349
7350
7351
7352
7353
7354
7355
7356
7357
7358
7359
7360
7361
7362
7363
7364
7365
7366
7367
7368
7369
7370
7371
7372
7373
7374
7375
7376
7377
7378
7379
7380
7381
7382
7383
7384
7385
7386
7387
7388
7389
7390
7391
7392
7393
7394
7395
7396
7397
7398
7399
7400
7401
7402
7403
7404
7405
7406
7407
7408
7409
7410
7411
7412
7413
7414
7415
7416
7417
7418
7419
7420
7421
7422
7423
7424
7425
7426
7427
7428
7429
7430
7431
7432
7433
7434
7435
7436
7437
7438
7439
7440
7441
7442
7443
7444
7445
7446
7447
7448
7449
7450
7451
7452
7453
7454
7455
7456
7457
7458
7459
7460
7461
7462
7463
7464
7465
7466
7467
7468
7469
7470
7471
7472
7473
7474
7475
7476
7477
7478
7479
7480
7481
7482
7483
7484
7485
7486
7487
7488
7489
7490
7491
7492
7493
7494
7495
7496
7497
7498
7499
7500
7501
7502
7503
7504
7505
7506
7507
7508
7509
7510
7511
7512
7513
7514
7515
7516
7517
7518
7519
7520
7521
7522
7523
7524
7525
7526
7527
7528
7529
7530
7531
7532
7533
7534
7535
7536
7537
7538
7539
7540
7541
7542
7543
7544
7545
7546
7547
7548
7549
7550
7551
7552
7553
7554
7555
7556
7557
7558
7559
7560
7561
7562
7563
7564
7565
7566
7567
7568
7569
7570
7571
7572
7573
7574
7575
7576
7577
7578
7579
7580
7581
7582
7583
7584
7585
7586
7587
7588
7589
7590
7591
7592
7593
7594
7595
7596
7597
7598
7599
7600
7601
7602
7603
7604
7605
7606
7607
7608
7609
7610
7611
7612
7613
7614
7615
7616
7617
7618
7619
7620
7621
7622
7623
7624
7625
7626
7627
7628
7629
7630
7631
7632
7633
7634
7635
7636
7637
7638
7639
7640
7641
7642
7643
7644
7645
7646
7647
7648
7649
7650
7651
7652
7653
7654
7655
7656
7657
7658
7659
7660
7661
7662
7663
7664
7665
7666
7667
7668
7669
7670
7671
7672
7673
7674
7675
7676
7677
7678
7679
7680
7681
7682
7683
7684
7685
7686
7687
7688
7689
7690
7691
7692
7693
7694
7695
7696
7697
7698
7699
7700
7701
7702
7703
7704
7705
7706
7707
7708
7709
7710
7711
7712
7713
7714
7715
7716
7717
7718
7719
7720
7721
7722
7723
7724
7725
7726
7727
7728
7729
7730
7731
7732
7733
7734
7735
7736
7737
7738
7739
7740
7741
7742
7743
7744
7745
7746
7747
7748
7749
7750
7751
7752
7753
7754
7755
7756
7757
7758
7759
7760
7761
7762
7763
7764
7765
7766
7767
7768
7769
7770
7771
7772
7773
7774
7775
7776
7777
7778
7779
7780
7781
7782
7783
7784
7785
7786
7787
7788
7789
7790
7791
7792
7793
7794
7795
7796
7797
7798
7799
7800
7801
7802
7803
7804
7805
7806
7807
7808
7809
7810
7811
7812
7813
7814
7815
7816
7817
7818
7819
7820
7821
7822
7823
7824
7825
7826
7827
7828
7829
7830
7831
7832
7833
7834
7835
7836
7837
7838
7839
7840
7841
7842
7843
7844
7845
7846
7847
7848
7849
7850
7851
7852
7853
7854
7855
7856
7857
7858
7859
7860
7861
7862
7863
7864
7865
7866
7867
7868
7869
7870
7871
7872
7873
7874
7875
7876
7877
7878
7879
7880
7881
7882
7883
7884
7885
7886
7887
7888
7889
7890
7891
7892
7893
7894
7895
7896
7897
7898
7899
7900
7901
7902
7903
7904
7905
7906
7907
7908
7909
7910
7911
7912
7913
7914
7915
7916
7917
7918
7919
7920
7921
7922
7923
7924
7925
7926
7927
7928
7929
7930
7931
7932
7933
7934
7935
7936
7937
7938
7939
7940
7941
7942
7943
7944
7945
7946
7947
7948
7949
7950
7951
7952
7953
7954
7955
7956
7957
7958
7959
7960
7961
7962
7963
7964
7965
7966
7967
7968
7969
7970
7971
7972
7973
7974
7975
7976
7977
7978
7979
7980
7981
7982
7983
7984
7985
7986
7987
7988
7989
7990
7991
7992
7993
7994
7995
7996
7997
7998
7999
8000
8001
8002
8003
8004
8005
8006
8007
8008
8009
8010
8011
8012
8013
8014
8015
8016
8017
8018
8019
8020
8021
8022
8023
8024
8025
8026
8027
8028
8029
8030
8031
8032
8033
8034
8035
8036
8037
8038
8039
8040
8041
8042
8043
8044
8045
8046
8047
8048
8049
8050
8051
8052
8053
8054
8055
8056
8057
8058
8059
8060
8061
8062
8063
8064
8065
8066
8067
8068
8069
8070
8071
8072
8073
8074
8075
8076
8077
8078
8079
8080
8081
8082
8083
8084
8085
8086
8087
8088
8089
8090
8091
8092
8093
8094
8095
8096
8097
8098
8099
8100
8101
8102
8103
8104
8105
8106
8107
8108
8109
8110
8111
8112
8113
8114
8115
8116
8117
8118
8119
8120
8121
8122
8123
8124
8125
8126
8127
8128
8129
8130
8131
8132
8133
8134
8135
8136
8137
8138
8139
8140
8141
8142
8143
8144
8145
8146
8147
8148
/* Coding system handler (conversion, detection, and etc).
   Copyright (C) 2001, 2002, 2003, 2004, 2005,
                 2006, 2007 Free Software Foundation, Inc.
   Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
     2005, 2006, 2007
     National Institute of Advanced Industrial Science and Technology (AIST)
     Registration Number H14PRO021

This file is part of GNU Emacs.

GNU Emacs is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3, or (at your option)
any later version.

GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with GNU Emacs; see the file COPYING.  If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
Boston, MA 02110-1301, USA.  */

/*** TABLE OF CONTENTS ***

  0. General comments
  1. Preamble
  2. Emacs' internal format (emacs-mule) handlers
  3. ISO2022 handlers
  4. Shift-JIS and BIG5 handlers
  5. CCL handlers
  6. End-of-line handlers
  7. C library functions
  8. Emacs Lisp library functions
  9. Post-amble

*/

/*** 0. General comments ***/


/*** GENERAL NOTE on CODING SYSTEMS ***

  A coding system is an encoding mechanism for one or more character
  sets.  Here's a list of coding systems which Emacs can handle.  When
  we say "decode", it means converting some other coding system to
  Emacs' internal format (emacs-mule), and when we say "encode",
  it means converting the coding system emacs-mule to some other
  coding system.

  0. Emacs' internal format (emacs-mule)

  Emacs itself holds a multi-lingual character in buffers and strings
  in a special format.  Details are described in section 2.

  1. ISO2022

  The most famous coding system for multiple character sets.  X's
  Compound Text, various EUCs (Extended Unix Code), and coding
  systems used in Internet communication such as ISO-2022-JP are
  all variants of ISO2022.  Details are described in section 3.

  2. SJIS (or Shift-JIS or MS-Kanji-Code)

  A coding system to encode character sets: ASCII, JISX0201, and
  JISX0208.  Widely used for PC's in Japan.  Details are described in
  section 4.

  3. BIG5

  A coding system to encode the character sets ASCII and Big5.  Widely
  used for Chinese (mainly in Taiwan and Hong Kong).  Details are
  described in section 4.  In this file, when we write "BIG5"
  (all uppercase), we mean the coding system, and when we write
  "Big5" (capitalized), we mean the character set.

  4. Raw text

  A coding system for text containing random 8-bit code.  Emacs does
  no code conversion on such text except for end-of-line format.

  5. Other

  If a user wants to read/write text encoded in a coding system not
  listed above, he can supply a decoder and an encoder for it as CCL
  (Code Conversion Language) programs.  Emacs executes the CCL program
  while reading/writing.

  Emacs represents a coding system by a Lisp symbol that has a property
  `coding-system'.  But, before actually using the coding system, the
  information about it is set in a structure of type `struct
  coding_system' for rapid processing.  See section 6 for more details.

*/

/*** GENERAL NOTES on END-OF-LINE FORMAT ***

  How end-of-line of text is encoded depends on the operating system.
  For instance, Unix's format is just one byte of `line-feed' code,
  whereas DOS's format is two-byte sequence of `carriage-return' and
  `line-feed' codes.  MacOS's format is usually one byte of
  `carriage-return'.

  Since text character encoding and end-of-line encoding are
  independent, any coding system described above can have any
  end-of-line format.  So Emacs has information about end-of-line
  format in each coding-system.  See section 6 for more details.

*/

/*** GENERAL NOTES on `detect_coding_XXX ()' functions ***

  These functions check if a text between SRC and SRC_END is encoded
  in the coding system category XXX.  Each returns an integer value in
  which appropriate flag bits for the category XXX are set.  The flag
  bits are defined in macros CODING_CATEGORY_MASK_XXX.  Below is the
  template for these functions.  If MULTIBYTEP is nonzero, 8-bit codes
  of the range 0x80..0x9F are in multibyte form.  */
#if 0
int
detect_coding_emacs_mule (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  ...
}
#endif

/*** GENERAL NOTES on `decode_coding_XXX ()' functions ***

  These functions decode SRC_BYTES length of unibyte text at SOURCE
  encoded in CODING to Emacs' internal format.  The resulting
  multibyte text goes to a place pointed to by DESTINATION, the length
  of which should not exceed DST_BYTES.

  These functions set the information about original and decoded texts
  in the members `produced', `produced_char', `consumed', and
  `consumed_char' of the structure *CODING.  They also set the member
  `result' to one of CODING_FINISH_XXX indicating how the decoding
  finished.

  DST_BYTES zero means that the source area and destination area are
  overlapped, which means that we can produce a decoded text until it
  reaches the head of the not-yet-decoded source text.

  Below is a template for these functions.  */
#if 0
static void
decode_coding_XXX (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  ...
}
#endif

/*** GENERAL NOTES on `encode_coding_XXX ()' functions ***

  These functions encode SRC_BYTES length text at SOURCE from Emacs'
  internal multibyte format to CODING.  The resulting unibyte text
  goes to a place pointed to by DESTINATION, the length of which
  should not exceed DST_BYTES.

  These functions set the information about original and encoded texts
  in the members `produced', `produced_char', `consumed', and
  `consumed_char' of the structure *CODING.  They also set the member
  `result' to one of CODING_FINISH_XXX indicating how the encoding
  finished.

  DST_BYTES zero means that the source area and destination area are
  overlapped, which means that we can produce encoded text until it
  reaches at the head of the not-yet-encoded source text.

  Below is a template for these functions.  */
#if 0
static void
encode_coding_XXX (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     unsigned char *source, *destination;
     int src_bytes, dst_bytes;
{
  ...
}
#endif

/*** COMMONLY USED MACROS ***/

/* The following two macros ONE_MORE_BYTE and TWO_MORE_BYTES safely
   get one, two, and three bytes from the source text respectively.
   If there are not enough bytes in the source, they jump to
   `label_end_of_loop'.  The caller should set variables `coding',
   `src' and `src_end' to appropriate pointer in advance.  These
   macros are called from decoding routines `decode_coding_XXX', thus
   it is assumed that the source text is unibyte.  */

#define ONE_MORE_BYTE(c1)					\
  do {								\
    if (src >= src_end)						\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_SRC;	\
	goto label_end_of_loop;					\
      }								\
    c1 = *src++;						\
  } while (0)

#define TWO_MORE_BYTES(c1, c2)					\
  do {								\
    if (src + 1 >= src_end)					\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_SRC;	\
	goto label_end_of_loop;					\
      }								\
    c1 = *src++;						\
    c2 = *src++;						\
  } while (0)


/* Like ONE_MORE_BYTE, but 8-bit bytes of data at SRC are in multibyte
   form if MULTIBYTEP is nonzero.  In addition, if SRC is not less
   than SRC_END, return with RET.  */

#define ONE_MORE_BYTE_CHECK_MULTIBYTE(c1, multibytep, ret)	\
  do {								\
    if (src >= src_end)						\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_SRC;	\
	return ret;						\
      }								\
    c1 = *src++;						\
    if (multibytep && c1 == LEADING_CODE_8_BIT_CONTROL)		\
      c1 = *src++ - 0x20;					\
  } while (0)

/* Set C to the next character at the source text pointed by `src'.
   If there are not enough characters in the source, jump to
   `label_end_of_loop'.  The caller should set variables `coding'
   `src', `src_end', and `translation_table' to appropriate pointers
   in advance.  This macro is used in encoding routines
   `encode_coding_XXX', thus it assumes that the source text is in
   multibyte form except for 8-bit characters.  8-bit characters are
   in multibyte form if coding->src_multibyte is nonzero, else they
   are represented by a single byte.  */

#define ONE_MORE_CHAR(c)					\
  do {								\
    int len = src_end - src;					\
    int bytes;							\
    if (len <= 0)						\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_SRC;	\
	goto label_end_of_loop;					\
      }								\
    if (coding->src_multibyte					\
	|| UNIBYTE_STR_AS_MULTIBYTE_P (src, len, bytes))	\
      c = STRING_CHAR_AND_LENGTH (src, len, bytes);		\
    else							\
      c = *src, bytes = 1;					\
    if (!NILP (translation_table))				\
      c = translate_char (translation_table, c, -1, 0, 0);	\
    src += bytes;						\
  } while (0)


/* Produce a multibyte form of character C to `dst'.  Jump to
   `label_end_of_loop' if there's not enough space at `dst'.

   If we are now in the middle of a composition sequence, the decoded
   character may be ALTCHAR (for the current composition).  In that
   case, the character goes to coding->cmp_data->data instead of
   `dst'.

   This macro is used in decoding routines.  */

#define EMIT_CHAR(c)							\
  do {									\
    if (! COMPOSING_P (coding)						\
	|| coding->composing == COMPOSITION_RELATIVE			\
	|| coding->composing == COMPOSITION_WITH_RULE)			\
      {									\
	int bytes = CHAR_BYTES (c);					\
	if ((dst + bytes) > (dst_bytes ? dst_end : src))		\
	  {								\
	    coding->result = CODING_FINISH_INSUFFICIENT_DST;		\
	    goto label_end_of_loop;					\
	  }								\
	dst += CHAR_STRING (c, dst);					\
	coding->produced_char++;					\
      }									\
    									\
    if (COMPOSING_P (coding)						\
	&& coding->composing != COMPOSITION_RELATIVE)			\
      {									\
	CODING_ADD_COMPOSITION_COMPONENT (coding, c);			\
	coding->composition_rule_follows				\
	  = coding->composing != COMPOSITION_WITH_ALTCHARS;		\
      }									\
  } while (0)


#define EMIT_ONE_BYTE(c)					\
  do {								\
    if (dst >= (dst_bytes ? dst_end : src))			\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_DST;	\
	goto label_end_of_loop;					\
      }								\
    *dst++ = c;							\
  } while (0)

#define EMIT_TWO_BYTES(c1, c2)					\
  do {								\
    if (dst + 2 > (dst_bytes ? dst_end : src))			\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_DST;	\
	goto label_end_of_loop;					\
      }								\
    *dst++ = c1, *dst++ = c2;					\
  } while (0)

#define EMIT_BYTES(from, to)					\
  do {								\
    if (dst + (to - from) > (dst_bytes ? dst_end : src))	\
      {								\
	coding->result = CODING_FINISH_INSUFFICIENT_DST;	\
	goto label_end_of_loop;					\
      }								\
    while (from < to)						\
      *dst++ = *from++;						\
  } while (0)


/*** 1. Preamble ***/

#ifdef emacs
#include <config.h>
#endif

#include <stdio.h>

#ifdef emacs

#include "lisp.h"
#include "buffer.h"
#include "charset.h"
#include "composite.h"
#include "ccl.h"
#include "coding.h"
#include "window.h"
#include "intervals.h"

#else  /* not emacs */

#include "mulelib.h"

#endif /* not emacs */

Lisp_Object Qcoding_system, Qeol_type;
Lisp_Object Qbuffer_file_coding_system;
Lisp_Object Qpost_read_conversion, Qpre_write_conversion;
Lisp_Object Qno_conversion, Qundecided;
Lisp_Object Qcoding_system_history;
Lisp_Object Qsafe_chars;
Lisp_Object Qvalid_codes;
Lisp_Object Qascii_incompatible;

extern Lisp_Object Qinsert_file_contents, Qwrite_region;
Lisp_Object Qcall_process, Qcall_process_region;
Lisp_Object Qstart_process, Qopen_network_stream;
Lisp_Object Qtarget_idx;

/* If a symbol has this property, evaluate the value to define the
   symbol as a coding system.  */
Lisp_Object Qcoding_system_define_form;

Lisp_Object Vselect_safe_coding_system_function;

int coding_system_require_warning;

/* Mnemonic string for each format of end-of-line.  */
Lisp_Object eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac;
/* Mnemonic string to indicate format of end-of-line is not yet
   decided.  */
Lisp_Object eol_mnemonic_undecided;

/* Format of end-of-line decided by system.  This is CODING_EOL_LF on
   Unix, CODING_EOL_CRLF on DOS/Windows, and CODING_EOL_CR on Mac.
   This has an effect only for external encoding (i.e. for output to
   file and process), not for in-buffer or Lisp string encoding.  */
int system_eol_type;

#ifdef emacs

/* Information about which coding system is safe for which chars.
   The value has the form (GENERIC-LIST . NON-GENERIC-ALIST).

   GENERIC-LIST is a list of generic coding systems which can encode
   any characters.

   NON-GENERIC-ALIST is an alist of non generic coding systems vs the
   corresponding char table that contains safe chars.  */
Lisp_Object Vcoding_system_safe_chars;

Lisp_Object Vcoding_system_list, Vcoding_system_alist;

Lisp_Object Qcoding_system_p, Qcoding_system_error;

/* Coding system emacs-mule and raw-text are for converting only
   end-of-line format.  */
Lisp_Object Qemacs_mule, Qraw_text;

Lisp_Object Qutf_8;

/* Coding-systems are handed between Emacs Lisp programs and C internal
   routines by the following three variables.  */
/* Coding-system for reading files and receiving data from process.  */
Lisp_Object Vcoding_system_for_read;
/* Coding-system for writing files and sending data to process.  */
Lisp_Object Vcoding_system_for_write;
/* Coding-system actually used in the latest I/O.  */
Lisp_Object Vlast_coding_system_used;

/* A vector of length 256 which contains information about special
   Latin codes (especially for dealing with Microsoft codes).  */
Lisp_Object Vlatin_extra_code_table;

/* Flag to inhibit code conversion of end-of-line format.  */
int inhibit_eol_conversion;

/* Flag to inhibit ISO2022 escape sequence detection.  */
int inhibit_iso_escape_detection;

/* Flag to make buffer-file-coding-system inherit from process-coding.  */
int inherit_process_coding_system;

/* Coding system to be used to encode text for terminal display.  */
struct coding_system terminal_coding;

/* Coding system to be used to encode text for terminal display when
   terminal coding system is nil.  */
struct coding_system safe_terminal_coding;

/* Coding system of what is sent from terminal keyboard.  */
struct coding_system keyboard_coding;

/* Default coding system to be used to write a file.  */
struct coding_system default_buffer_file_coding;

Lisp_Object Vfile_coding_system_alist;
Lisp_Object Vprocess_coding_system_alist;
Lisp_Object Vnetwork_coding_system_alist;

Lisp_Object Vlocale_coding_system;

#endif /* emacs */

Lisp_Object Qcoding_category, Qcoding_category_index;

/* List of symbols `coding-category-xxx' ordered by priority.  */
Lisp_Object Vcoding_category_list;

/* Table of coding categories (Lisp symbols).  */
Lisp_Object Vcoding_category_table;

/* Table of names of symbol for each coding-category.  */
char *coding_category_name[CODING_CATEGORY_IDX_MAX] = {
  "coding-category-emacs-mule",
  "coding-category-sjis",
  "coding-category-iso-7",
  "coding-category-iso-7-tight",
  "coding-category-iso-8-1",
  "coding-category-iso-8-2",
  "coding-category-iso-7-else",
  "coding-category-iso-8-else",
  "coding-category-ccl",
  "coding-category-big5",
  "coding-category-utf-8",
  "coding-category-utf-16-be",
  "coding-category-utf-16-le",
  "coding-category-raw-text",
  "coding-category-binary"
};

/* Table of pointers to coding systems corresponding to each coding
   categories.  */
struct coding_system *coding_system_table[CODING_CATEGORY_IDX_MAX];

/* Table of coding category masks.  Nth element is a mask for a coding
   category of which priority is Nth.  */
static
int coding_priorities[CODING_CATEGORY_IDX_MAX];

/* Flag to tell if we look up translation table on character code
   conversion.  */
Lisp_Object Venable_character_translation;
/* Standard translation table to look up on decoding (reading).  */
Lisp_Object Vstandard_translation_table_for_decode;
/* Standard translation table to look up on encoding (writing).  */
Lisp_Object Vstandard_translation_table_for_encode;

Lisp_Object Qtranslation_table;
Lisp_Object Qtranslation_table_id;
Lisp_Object Qtranslation_table_for_decode;
Lisp_Object Qtranslation_table_for_encode;

/* Alist of charsets vs revision number.  */
Lisp_Object Vcharset_revision_alist;

/* Default coding systems used for process I/O.  */
Lisp_Object Vdefault_process_coding_system;

/* Char table for translating Quail and self-inserting input.  */
Lisp_Object Vtranslation_table_for_input;

/* Global flag to tell that we can't call post-read-conversion and
   pre-write-conversion functions.  Usually the value is zero, but it
   is set to 1 temporarily while such functions are running.  This is
   to avoid infinite recursive call.  */
static int inhibit_pre_post_conversion;

Lisp_Object Qchar_coding_system;

/* Return `safe-chars' property of CODING_SYSTEM (symbol).  Don't check
   its validity.  */

Lisp_Object
coding_safe_chars (coding_system)
     Lisp_Object coding_system;
{
  Lisp_Object coding_spec, plist, safe_chars;

  coding_spec = Fget (coding_system, Qcoding_system);
  plist = XVECTOR (coding_spec)->contents[3];
  safe_chars = Fplist_get (XVECTOR (coding_spec)->contents[3], Qsafe_chars);
  return (CHAR_TABLE_P (safe_chars) ? safe_chars : Qt);
}

#define CODING_SAFE_CHAR_P(safe_chars, c) \
  (EQ (safe_chars, Qt) || !NILP (CHAR_TABLE_REF (safe_chars, c)))


/*** 2. Emacs internal format (emacs-mule) handlers ***/

/* Emacs' internal format for representation of multiple character
   sets is a kind of multi-byte encoding, i.e. characters are
   represented by variable-length sequences of one-byte codes.

   ASCII characters and control characters (e.g. `tab', `newline') are
   represented by one-byte sequences which are their ASCII codes, in
   the range 0x00 through 0x7F.

   8-bit characters of the range 0x80..0x9F are represented by
   two-byte sequences of LEADING_CODE_8_BIT_CONTROL and (their 8-bit
   code + 0x20).

   8-bit characters of the range 0xA0..0xFF are represented by
   one-byte sequences which are their 8-bit code.

   The other characters are represented by a sequence of `base
   leading-code', optional `extended leading-code', and one or two
   `position-code's.  The length of the sequence is determined by the
   base leading-code.  Leading-code takes the range 0x81 through 0x9D,
   whereas extended leading-code and position-code take the range 0xA0
   through 0xFF.  See `charset.h' for more details about leading-code
   and position-code.

   --- CODE RANGE of Emacs' internal format ---
   character set	range
   -------------	-----
   ascii		0x00..0x7F
   eight-bit-control	LEADING_CODE_8_BIT_CONTROL + 0xA0..0xBF
   eight-bit-graphic	0xA0..0xBF
   ELSE			0x81..0x9D + [0xA0..0xFF]+
   ---------------------------------------------

   As this is the internal character representation, the format is
   usually not used externally (i.e. in a file or in a data sent to a
   process).  But, it is possible to have a text externally in this
   format (i.e. by encoding by the coding system `emacs-mule').

   In that case, a sequence of one-byte codes has a slightly different
   form.

   Firstly, all characters in eight-bit-control are represented by
   one-byte sequences which are their 8-bit code.

   Next, character composition data are represented by the byte
   sequence of the form: 0x80 METHOD BYTES CHARS COMPONENT ...,
   where,
	METHOD is 0xF0 plus one of composition method (enum
	composition_method),

	BYTES is 0xA0 plus the byte length of these composition data,

	CHARS is 0xA0 plus the number of characters composed by these
	data,

	COMPONENTs are characters of multibyte form or composition
	rules encoded by two-byte of ASCII codes.

   In addition, for backward compatibility, the following formats are
   also recognized as composition data on decoding.

   0x80 MSEQ ...
   0x80 0xFF MSEQ RULE MSEQ RULE ... MSEQ

   Here,
	MSEQ is a multibyte form but in these special format:
	  ASCII: 0xA0 ASCII_CODE+0x80,
	  other: LEADING_CODE+0x20 FOLLOWING-BYTE ...,
	RULE is a one byte code of the range 0xA0..0xF0 that
	represents a composition rule.
  */

enum emacs_code_class_type emacs_code_class[256];

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in Emacs' internal format.  If it is,
   return CODING_CATEGORY_MASK_EMACS_MULE, else return 0.  */

static int
detect_coding_emacs_mule (src, src_end, multibytep)
      unsigned char *src, *src_end;
      int multibytep;
{
  unsigned char c;
  int composing = 0;
  /* Dummy for ONE_MORE_BYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;

  while (1)
    {
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep,
				     CODING_CATEGORY_MASK_EMACS_MULE);
      if (composing)
	{
	  if (c < 0xA0)
	    composing = 0;
	  else if (c == 0xA0)
	    {
	      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, 0);
	      c &= 0x7F;
	    }
	  else
	    c -= 0x20;
	}

      if (c < 0x20)
	{
	  if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO)
	    return 0;
	}
      else if (c >= 0x80 && c < 0xA0)
	{
	  if (c == 0x80)
	    /* Old leading code for a composite character.  */
	    composing = 1;
	  else
	    {
	      unsigned char *src_base = src - 1;
	      int bytes;

	      if (!UNIBYTE_STR_AS_MULTIBYTE_P (src_base, src_end - src_base,
					       bytes))
		return 0;
	      src = src_base + bytes;
	    }
	}
    }
}


/* Record the starting position START and METHOD of one composition.  */

#define CODING_ADD_COMPOSITION_START(coding, start, method)	\
  do {								\
    struct composition_data *cmp_data = coding->cmp_data;	\
    int *data = cmp_data->data + cmp_data->used;		\
    coding->cmp_data_start = cmp_data->used;			\
    data[0] = -1;						\
    data[1] = cmp_data->char_offset + start;			\
    data[3] = (int) method;					\
    cmp_data->used += 4;					\
  } while (0)

/* Record the ending position END of the current composition.  */

#define CODING_ADD_COMPOSITION_END(coding, end)			\
  do {								\
    struct composition_data *cmp_data = coding->cmp_data;	\
    int *data = cmp_data->data + coding->cmp_data_start;	\
    data[0] = cmp_data->used - coding->cmp_data_start;		\
    data[2] = cmp_data->char_offset + end;			\
  } while (0)

/* Record one COMPONENT (alternate character or composition rule).  */

#define CODING_ADD_COMPOSITION_COMPONENT(coding, component)		\
  do {									\
    coding->cmp_data->data[coding->cmp_data->used++] = component;	\
    if (coding->cmp_data->used - coding->cmp_data_start			\
	== COMPOSITION_DATA_MAX_BUNCH_LENGTH)				\
      {									\
	CODING_ADD_COMPOSITION_END (coding, coding->produced_char);	\
	coding->composing = COMPOSITION_NO;				\
      }									\
  } while (0)


/* Get one byte from a data pointed by SRC and increment SRC.  If SRC
   is not less than SRC_END, return -1 without incrementing Src.  */

#define SAFE_ONE_MORE_BYTE() (src >= src_end ? -1 : *src++)


/* Decode a character represented as a component of composition
   sequence of Emacs 20 style at SRC.  Set C to that character, store
   its multibyte form sequence at P, and set P to the end of that
   sequence.  If no valid character is found, set C to -1.  */

#define DECODE_EMACS_MULE_COMPOSITION_CHAR(c, p)		\
  do {								\
    int bytes;							\
								\
    c = SAFE_ONE_MORE_BYTE ();					\
    if (c < 0)							\
      break;							\
    if (CHAR_HEAD_P (c))					\
      c = -1;							\
    else if (c == 0xA0)						\
      {								\
	c = SAFE_ONE_MORE_BYTE ();				\
	if (c < 0xA0)						\
	  c = -1;						\
	else							\
	  {							\
	    c -= 0x80;						\
	    *p++ = c;						\
	  }							\
      }								\
    else if (BASE_LEADING_CODE_P (c - 0x20))			\
      {								\
	unsigned char *p0 = p;					\
								\
	c -= 0x20;						\
	*p++ = c;						\
	bytes = BYTES_BY_CHAR_HEAD (c);				\
	while (--bytes)						\
	  {							\
	    c = SAFE_ONE_MORE_BYTE ();				\
	    if (c < 0)						\
	      break;						\
	    *p++ = c;						\
	  }							\
	if (UNIBYTE_STR_AS_MULTIBYTE_P (p0, p - p0, bytes)	\
	    || (coding->flags /* We are recovering a file.  */	\
		&& p0[0] == LEADING_CODE_8_BIT_CONTROL		\
		&& ! CHAR_HEAD_P (p0[1])))			\
	  c = STRING_CHAR (p0, bytes);				\
	else							\
	  c = -1;						\
      }								\
    else							\
      c = -1;							\
  } while (0)


/* Decode a composition rule represented as a component of composition
   sequence of Emacs 20 style at SRC.  Set C to the rule.  If not
   valid rule is found, set C to -1.  */

#define DECODE_EMACS_MULE_COMPOSITION_RULE(c)		\
  do {							\
    c = SAFE_ONE_MORE_BYTE ();				\
    c -= 0xA0;						\
    if (c < 0 || c >= 81)				\
      c = -1;						\
    else						\
      {							\
	gref = c / 9, nref = c % 9;			\
	c = COMPOSITION_ENCODE_RULE (gref, nref);	\
      }							\
  } while (0)


/* Decode composition sequence encoded by `emacs-mule' at the source
   pointed by SRC.  SRC_END is the end of source.  Store information
   of the composition in CODING->cmp_data.

   For backward compatibility, decode also a composition sequence of
   Emacs 20 style.  In that case, the composition sequence contains
   characters that should be extracted into a buffer or string.  Store
   those characters at *DESTINATION in multibyte form.

   If we encounter an invalid byte sequence, return 0.
   If we encounter an insufficient source or destination, or
   insufficient space in CODING->cmp_data, return 1.
   Otherwise, return consumed bytes in the source.

*/
static INLINE int
decode_composition_emacs_mule (coding, src, src_end,
			       destination, dst_end, dst_bytes)
     struct coding_system *coding;
     const unsigned char *src, *src_end;
     unsigned char **destination, *dst_end;
     int dst_bytes;
{
  unsigned char *dst = *destination;
  int method, data_len, nchars;
  const unsigned char *src_base = src++;
  /* Store components of composition.  */
  int component[COMPOSITION_DATA_MAX_BUNCH_LENGTH];
  int ncomponent;
  /* Store multibyte form of characters to be composed.  This is for
     Emacs 20 style composition sequence.  */
  unsigned char buf[MAX_COMPOSITION_COMPONENTS * MAX_MULTIBYTE_LENGTH];
  unsigned char *bufp = buf;
  int c, i, gref, nref;

  if (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH
      >= COMPOSITION_DATA_SIZE)
    {
      coding->result = CODING_FINISH_INSUFFICIENT_CMP;
      return -1;
    }

  ONE_MORE_BYTE (c);
  if (c - 0xF0 >= COMPOSITION_RELATIVE
	   && c - 0xF0 <= COMPOSITION_WITH_RULE_ALTCHARS)
    {
      int with_rule;

      method = c - 0xF0;
      with_rule = (method == COMPOSITION_WITH_RULE
		   || method == COMPOSITION_WITH_RULE_ALTCHARS);
      ONE_MORE_BYTE (c);
      data_len = c - 0xA0;
      if (data_len < 4
	  || src_base + data_len > src_end)
	return 0;
      ONE_MORE_BYTE (c);
      nchars = c - 0xA0;
      if (c < 1)
	return 0;
      for (ncomponent = 0; src < src_base + data_len; ncomponent++)
	{
	  /* If it is longer than this, it can't be valid.  */
	  if (ncomponent >= COMPOSITION_DATA_MAX_BUNCH_LENGTH)
	    return 0;

	  if (ncomponent % 2 && with_rule)
	    {
	      ONE_MORE_BYTE (gref);
	      gref -= 32;
	      ONE_MORE_BYTE (nref);
	      nref -= 32;
	      c = COMPOSITION_ENCODE_RULE (gref, nref);
	    }
	  else
	    {
	      int bytes;
	      if (UNIBYTE_STR_AS_MULTIBYTE_P (src, src_end - src, bytes)
		  || (coding->flags /* We are recovering a file.  */
		      && src[0] == LEADING_CODE_8_BIT_CONTROL
		      && ! CHAR_HEAD_P (src[1])))
		c = STRING_CHAR (src, bytes);
	      else
		c = *src, bytes = 1;
	      src += bytes;
	    }
	  component[ncomponent] = c;
	}
    }
  else if (c >= 0x80)
    {
      /* This may be an old Emacs 20 style format.  See the comment at
	 the section 2 of this file.  */
      while (src < src_end && !CHAR_HEAD_P (*src)) src++;
      if (src == src_end
	  && !(coding->mode & CODING_MODE_LAST_BLOCK))
	goto label_end_of_loop;

      src_end = src;
      src = src_base + 1;
      if (c < 0xC0)
	{
	  method = COMPOSITION_RELATIVE;
	  for (ncomponent = 0; ncomponent < MAX_COMPOSITION_COMPONENTS;)
	    {
	      DECODE_EMACS_MULE_COMPOSITION_CHAR (c, bufp);
	      if (c < 0)
		break;
	      component[ncomponent++] = c;
	    }
	  if (ncomponent < 2)
	    return 0;
	  nchars = ncomponent;
	}
      else if (c == 0xFF)
	{
	  method = COMPOSITION_WITH_RULE;
	  src++;
	  DECODE_EMACS_MULE_COMPOSITION_CHAR (c, bufp);
	  if (c < 0)
	    return 0;
	  component[0] = c;
	  for (ncomponent = 1;
	       ncomponent < MAX_COMPOSITION_COMPONENTS * 2 - 1;)
	    {
	      DECODE_EMACS_MULE_COMPOSITION_RULE (c);
	      if (c < 0)
		break;
	      component[ncomponent++] = c;
	      DECODE_EMACS_MULE_COMPOSITION_CHAR (c, bufp);
	      if (c < 0)
		break;
	      component[ncomponent++] = c;
	    }
	  if (ncomponent < 3)
	    return 0;
	  nchars = (ncomponent + 1) / 2;
	}
      else
	return 0;
    }
  else
    return 0;

  if (buf == bufp || dst + (bufp - buf) <= (dst_bytes ? dst_end : src))
    {
      CODING_ADD_COMPOSITION_START (coding, coding->produced_char, method);
      for (i = 0; i < ncomponent; i++)
	CODING_ADD_COMPOSITION_COMPONENT (coding, component[i]);
      CODING_ADD_COMPOSITION_END (coding, coding->produced_char + nchars);
      if (buf < bufp)
	{
	  unsigned char *p = buf;
	  EMIT_BYTES (p, bufp);
	  *destination += bufp - buf;
	  coding->produced_char += nchars;
	}
      return (src - src_base);
    }
 label_end_of_loop:
  return -1;
}

/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".  */

static void
decode_coding_emacs_mule (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  const unsigned char *src = source;
  const unsigned char *src_end = source + src_bytes;
  unsigned char *dst = destination;
  unsigned char *dst_end = destination + dst_bytes;
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source code, or
     when there's not enough destination area to produce a
     character.  */
  const unsigned char *src_base;

  coding->produced_char = 0;
  while ((src_base = src) < src_end)
    {
      unsigned char tmp[MAX_MULTIBYTE_LENGTH];
      const unsigned char *p;
      int bytes;

      if (*src == '\r')
	{
	  int c = *src++;

	  if (coding->eol_type == CODING_EOL_CR)
	    c = '\n';
	  else if (coding->eol_type == CODING_EOL_CRLF)
	    {
	      ONE_MORE_BYTE (c);
	      if (c != '\n')
		{
		  src--;
		  c = '\r';
		}
	    }
	  *dst++ = c;
	  coding->produced_char++;
	  continue;
	}
      else if (*src == '\n')
	{
	  if ((coding->eol_type == CODING_EOL_CR
	       || coding->eol_type == CODING_EOL_CRLF)
	      && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
	    {
	      coding->result = CODING_FINISH_INCONSISTENT_EOL;
	      goto label_end_of_loop;
	    }
	  *dst++ = *src++;
	  coding->produced_char++;
	  continue;
	}
      else if (*src == 0x80 && coding->cmp_data)
	{
	  /* Start of composition data.  */
	  int consumed  = decode_composition_emacs_mule (coding, src, src_end,
							 &dst, dst_end,
							 dst_bytes);
	  if (consumed < 0)
	    goto label_end_of_loop;
	  else if (consumed > 0)
	    {
	      src += consumed;
	      continue;
	    }
	  bytes = CHAR_STRING (*src, tmp);
	  p = tmp;
	  src++;
	}
      else if (UNIBYTE_STR_AS_MULTIBYTE_P (src, src_end - src, bytes)
	       || (coding->flags /* We are recovering a file.  */
		   && src[0] == LEADING_CODE_8_BIT_CONTROL
		   && ! CHAR_HEAD_P (src[1])))
	{
	  p = src;
	  src += bytes;
	}
      else
	{
	  int i, c;

	  bytes = BYTES_BY_CHAR_HEAD (*src);
	  src++;
	  for (i = 1; i < bytes; i++)
	    {
	      ONE_MORE_BYTE (c);
	      if (CHAR_HEAD_P (c))
		break;
	    }
	  if (i < bytes)
	    {
	      bytes = CHAR_STRING (*src_base, tmp);
	      p = tmp;
	      src = src_base + 1;
	    }
	  else
	    {
	      p = src_base;
	    }
	}
      if (dst + bytes >= (dst_bytes ? dst_end : src))
	{
	  coding->result = CODING_FINISH_INSUFFICIENT_DST;
	  break;
	}
      while (bytes--) *dst++ = *p++;
      coding->produced_char++;
    }
 label_end_of_loop:
  coding->consumed = coding->consumed_char = src_base - source;
  coding->produced = dst - destination;
}


/* Encode composition data stored at DATA into a special byte sequence
   starting by 0x80.  Update CODING->cmp_data_start and maybe
   CODING->cmp_data for the next call.  */

#define ENCODE_COMPOSITION_EMACS_MULE(coding, data)			\
  do {									\
    unsigned char buf[1024], *p0 = buf, *p;				\
    int len = data[0];							\
    int i;								\
    									\
    buf[0] = 0x80;							\
    buf[1] = 0xF0 + data[3];	/* METHOD */				\
    buf[3] = 0xA0 + (data[2] - data[1]); /* COMPOSED-CHARS */		\
    p = buf + 4;							\
    if (data[3] == COMPOSITION_WITH_RULE				\
	|| data[3] == COMPOSITION_WITH_RULE_ALTCHARS)			\
      {									\
	p += CHAR_STRING (data[4], p);					\
	for (i = 5; i < len; i += 2)					\
	  {								\
	    int gref, nref;						\
	     COMPOSITION_DECODE_RULE (data[i], gref, nref);		\
	    *p++ = 0x20 + gref;						\
	    *p++ = 0x20 + nref;						\
	    p += CHAR_STRING (data[i + 1], p);				\
	  }								\
      }									\
    else								\
      {									\
	for (i = 4; i < len; i++)					\
	  p += CHAR_STRING (data[i], p);				\
      }									\
    buf[2] = 0xA0 + (p - buf);	/* COMPONENTS-BYTES */			\
    									\
    if (dst + (p - buf) + 4 > (dst_bytes ? dst_end : src))		\
      {									\
	coding->result = CODING_FINISH_INSUFFICIENT_DST;		\
	goto label_end_of_loop;						\
      }									\
    while (p0 < p)							\
      *dst++ = *p0++;							\
    coding->cmp_data_start += data[0];					\
    if (coding->cmp_data_start == coding->cmp_data->used		\
	&& coding->cmp_data->next)					\
      {									\
	coding->cmp_data = coding->cmp_data->next;			\
	coding->cmp_data_start = 0;					\
      }									\
  } while (0)


static void encode_eol P_ ((struct coding_system *, const unsigned char *,
			    unsigned char *, int, int));

static void
encode_coding_emacs_mule (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  const unsigned char *src = source;
  const unsigned char *src_end = source + src_bytes;
  unsigned char *dst = destination;
  unsigned char *dst_end = destination + dst_bytes;
  const unsigned char *src_base;
  int c;
  int char_offset;
  int *data;

  Lisp_Object translation_table;

  translation_table = Qnil;

  /* Optimization for the case that there's no composition.  */
  if (!coding->cmp_data || coding->cmp_data->used == 0)
    {
      encode_eol (coding, source, destination, src_bytes, dst_bytes);
      return;
    }

  char_offset = coding->cmp_data->char_offset;
  data = coding->cmp_data->data + coding->cmp_data_start;
  while (1)
    {
      src_base = src;

      /* If SRC starts a composition, encode the information about the
	 composition in advance.  */
      if (coding->cmp_data_start < coding->cmp_data->used
	  && char_offset + coding->consumed_char == data[1])
	{
	  ENCODE_COMPOSITION_EMACS_MULE (coding, data);
	  char_offset = coding->cmp_data->char_offset;
	  data = coding->cmp_data->data + coding->cmp_data_start;
	}

      ONE_MORE_CHAR (c);
      if (c == '\n' && (coding->eol_type == CODING_EOL_CRLF
			|| coding->eol_type == CODING_EOL_CR))
	{
	  if (coding->eol_type == CODING_EOL_CRLF)
	    EMIT_TWO_BYTES ('\r', c);
	  else
	    EMIT_ONE_BYTE ('\r');
	}
      else if (SINGLE_BYTE_CHAR_P (c))
	{
	  if (coding->flags && ! ASCII_BYTE_P (c))
	    {
	      /* As we are auto saving, retain the multibyte form for
		 8-bit chars.  */
	      unsigned char buf[MAX_MULTIBYTE_LENGTH];
	      int bytes = CHAR_STRING (c, buf);

	      if (bytes == 1)
		EMIT_ONE_BYTE (buf[0]);
	      else
		EMIT_TWO_BYTES (buf[0], buf[1]);
	    }
	  else
	    EMIT_ONE_BYTE (c);
	}
      else
	EMIT_BYTES (src_base, src);
      coding->consumed_char++;
    }
 label_end_of_loop:
  coding->consumed = src_base - source;
  coding->produced = coding->produced_char = dst - destination;
  return;
}


/*** 3. ISO2022 handlers ***/

/* The following note describes the coding system ISO2022 briefly.
   Since the intention of this note is to help understand the
   functions in this file, some parts are NOT ACCURATE or are OVERLY
   SIMPLIFIED.  For thorough understanding, please refer to the
   original document of ISO2022.  This is equivalent to the standard
   ECMA-35, obtainable from <URL:http://www.ecma.ch/> (*).

   ISO2022 provides many mechanisms to encode several character sets
   in 7-bit and 8-bit environments.  For 7-bit environments, all text
   is encoded using bytes less than 128.  This may make the encoded
   text a little bit longer, but the text passes more easily through
   several types of gateway, some of which strip off the MSB (Most
   Significant Bit).

   There are two kinds of character sets: control character sets and
   graphic character sets.  The former contain control characters such
   as `newline' and `escape' to provide control functions (control
   functions are also provided by escape sequences).  The latter
   contain graphic characters such as 'A' and '-'.  Emacs recognizes
   two control character sets and many graphic character sets.

   Graphic character sets are classified into one of the following
   four classes, according to the number of bytes (DIMENSION) and
   number of characters in one dimension (CHARS) of the set:
   - DIMENSION1_CHARS94
   - DIMENSION1_CHARS96
   - DIMENSION2_CHARS94
   - DIMENSION2_CHARS96

   In addition, each character set is assigned an identification tag,
   unique for each set, called the "final character" (denoted as <F>
   hereafter).  The <F> of each character set is decided by ECMA(*)
   when it is registered in ISO.  The code range of <F> is 0x30..0x7F
   (0x30..0x3F are for private use only).

   Note (*): ECMA = European Computer Manufacturers Association

   Here are examples of graphic character sets [NAME(<F>)]:
	o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ...
	o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ...
	o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ...
	o DIMENSION2_CHARS96 -- none for the moment

   A code area (1 byte=8 bits) is divided into 4 areas, C0, GL, C1, and GR.
	C0 [0x00..0x1F] -- control character plane 0
	GL [0x20..0x7F] -- graphic character plane 0
	C1 [0x80..0x9F] -- control character plane 1
	GR [0xA0..0xFF] -- graphic character plane 1

   A control character set is directly designated and invoked to C0 or
   C1 by an escape sequence.  The most common case is that:
   - ISO646's  control character set is designated/invoked to C0, and
   - ISO6429's control character set is designated/invoked to C1,
   and usually these designations/invocations are omitted in encoded
   text.  In a 7-bit environment, only C0 can be used, and a control
   character for C1 is encoded by an appropriate escape sequence to
   fit into the environment.  All control characters for C1 are
   defined to have corresponding escape sequences.

   A graphic character set is at first designated to one of four
   graphic registers (G0 through G3), then these graphic registers are
   invoked to GL or GR.  These designations and invocations can be
   done independently.  The most common case is that G0 is invoked to
   GL, G1 is invoked to GR, and ASCII is designated to G0.  Usually
   these invocations and designations are omitted in encoded text.
   In a 7-bit environment, only GL can be used.

   When a graphic character set of CHARS94 is invoked to GL, codes
   0x20 and 0x7F of the GL area work as control characters SPACE and
   DEL respectively, and codes 0xA0 and 0xFF of the GR area should not
   be used.

   There are two ways of invocation: locking-shift and single-shift.
   With locking-shift, the invocation lasts until the next different
   invocation, whereas with single-shift, the invocation affects the
   following character only and doesn't affect the locking-shift
   state.  Invocations are done by the following control characters or
   escape sequences:

   ----------------------------------------------------------------------
   abbrev  function	             cntrl escape seq	description
   ----------------------------------------------------------------------
   SI/LS0  (shift-in)		     0x0F  none		invoke G0 into GL
   SO/LS1  (shift-out)		     0x0E  none		invoke G1 into GL
   LS2     (locking-shift-2)	     none  ESC 'n'	invoke G2 into GL
   LS3     (locking-shift-3)	     none  ESC 'o'	invoke G3 into GL
   LS1R    (locking-shift-1 right)   none  ESC '~'      invoke G1 into GR (*)
   LS2R    (locking-shift-2 right)   none  ESC '}'      invoke G2 into GR (*)
   LS3R    (locking-shift 3 right)   none  ESC '|'      invoke G3 into GR (*)
   SS2     (single-shift-2)	     0x8E  ESC 'N'	invoke G2 for one char
   SS3     (single-shift-3)	     0x8F  ESC 'O'	invoke G3 for one char
   ----------------------------------------------------------------------
   (*) These are not used by any known coding system.

   Control characters for these functions are defined by macros
   ISO_CODE_XXX in `coding.h'.

   Designations are done by the following escape sequences:
   ----------------------------------------------------------------------
   escape sequence	description
   ----------------------------------------------------------------------
   ESC '(' <F>		designate DIMENSION1_CHARS94<F> to G0
   ESC ')' <F>		designate DIMENSION1_CHARS94<F> to G1
   ESC '*' <F>		designate DIMENSION1_CHARS94<F> to G2
   ESC '+' <F>		designate DIMENSION1_CHARS94<F> to G3
   ESC ',' <F>		designate DIMENSION1_CHARS96<F> to G0 (*)
   ESC '-' <F>		designate DIMENSION1_CHARS96<F> to G1
   ESC '.' <F>		designate DIMENSION1_CHARS96<F> to G2
   ESC '/' <F>		designate DIMENSION1_CHARS96<F> to G3
   ESC '$' '(' <F>	designate DIMENSION2_CHARS94<F> to G0 (**)
   ESC '$' ')' <F>	designate DIMENSION2_CHARS94<F> to G1
   ESC '$' '*' <F>	designate DIMENSION2_CHARS94<F> to G2
   ESC '$' '+' <F>	designate DIMENSION2_CHARS94<F> to G3
   ESC '$' ',' <F>	designate DIMENSION2_CHARS96<F> to G0 (*)
   ESC '$' '-' <F>	designate DIMENSION2_CHARS96<F> to G1
   ESC '$' '.' <F>	designate DIMENSION2_CHARS96<F> to G2
   ESC '$' '/' <F>	designate DIMENSION2_CHARS96<F> to G3
   ----------------------------------------------------------------------

   In this list, "DIMENSION1_CHARS94<F>" means a graphic character set
   of dimension 1, chars 94, and final character <F>, etc...

   Note (*): Although these designations are not allowed in ISO2022,
   Emacs accepts them on decoding, and produces them on encoding
   CHARS96 character sets in a coding system which is characterized as
   7-bit environment, non-locking-shift, and non-single-shift.

   Note (**): If <F> is '@', 'A', or 'B', the intermediate character
   '(' can be omitted.  We refer to this as "short-form" hereafter.

   Now you may notice that there are a lot of ways of encoding the
   same multilingual text in ISO2022.  Actually, there exist many
   coding systems such as Compound Text (used in X11's inter client
   communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR
   (used in Korean Internet), EUC (Extended UNIX Code, used in Asian
   localized platforms), and all of these are variants of ISO2022.

   In addition to the above, Emacs handles two more kinds of escape
   sequences: ISO6429's direction specification and Emacs' private
   sequence for specifying character composition.

   ISO6429's direction specification takes the following form:
	o CSI ']'      -- end of the current direction
	o CSI '0' ']'  -- end of the current direction
	o CSI '1' ']'  -- start of left-to-right text
	o CSI '2' ']'  -- start of right-to-left text
   The control character CSI (0x9B: control sequence introducer) is
   abbreviated to the escape sequence ESC '[' in a 7-bit environment.

   Character composition specification takes the following form:
	o ESC '0' -- start relative composition
	o ESC '1' -- end composition
	o ESC '2' -- start rule-base composition (*)
	o ESC '3' -- start relative composition with alternate chars  (**)
	o ESC '4' -- start rule-base composition with alternate chars  (**)
  Since these are not standard escape sequences of any ISO standard,
  the use of them with these meanings is restricted to Emacs only.

  (*) This form is used only in Emacs 20.5 and older versions,
  but the newer versions can safely decode it.
  (**) This form is used only in Emacs 21.1 and newer versions,
  and the older versions can't decode it.

  Here's a list of example usages of these composition escape
  sequences (categorized by `enum composition_method').

  COMPOSITION_RELATIVE:
	ESC 0 CHAR [ CHAR ] ESC 1
  COMPOSITION_WITH_RULE:
	ESC 2 CHAR [ RULE CHAR ] ESC 1
  COMPOSITION_WITH_ALTCHARS:
	ESC 3 ALTCHAR [ ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1
  COMPOSITION_WITH_RULE_ALTCHARS:
	ESC 4 ALTCHAR [ RULE ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1 */

enum iso_code_class_type iso_code_class[256];

#define CHARSET_OK(idx, charset, c)					\
  (coding_system_table[idx]						\
   && (charset == CHARSET_ASCII						\
       || (safe_chars = coding_safe_chars (coding_system_table[idx]->symbol), \
	   CODING_SAFE_CHAR_P (safe_chars, c)))				\
   && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding_system_table[idx],	\
					      charset)			\
       != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))

#define SHIFT_OUT_OK(idx) \
  (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0)

#define COMPOSITION_OK(idx)	\
  (coding_system_table[idx]->composing != COMPOSITION_DISABLED)

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in ISO2022.  If it is, return an
   integer in which appropriate flag bits any of:
	CODING_CATEGORY_MASK_ISO_7
	CODING_CATEGORY_MASK_ISO_7_TIGHT
	CODING_CATEGORY_MASK_ISO_8_1
	CODING_CATEGORY_MASK_ISO_8_2
	CODING_CATEGORY_MASK_ISO_7_ELSE
	CODING_CATEGORY_MASK_ISO_8_ELSE
   are set.  If a code which should never appear in ISO2022 is found,
   returns 0.  */

static int
detect_coding_iso2022 (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  int mask = CODING_CATEGORY_MASK_ISO;
  int mask_found = 0;
  int reg[4], shift_out = 0, single_shifting = 0;
  int c, c1, charset;
  /* Dummy for ONE_MORE_BYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;
  Lisp_Object safe_chars;

  reg[0] = CHARSET_ASCII, reg[1] = reg[2] = reg[3] = -1;
  while (mask)
    {
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, mask & mask_found);
    retry:
      switch (c)
	{
	case ISO_CODE_ESC:
	  if (inhibit_iso_escape_detection)
	    break;
	  single_shifting = 0;
	  ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, mask & mask_found);
	  if (c >= '(' && c <= '/')
	    {
	      /* Designation sequence for a charset of dimension 1.  */
	      ONE_MORE_BYTE_CHECK_MULTIBYTE (c1, multibytep, mask & mask_found);
	      if (c1 < ' ' || c1 >= 0x80
		  || (charset = iso_charset_table[0][c >= ','][c1]) < 0)
		/* Invalid designation sequence.  Just ignore.  */
		break;
	      reg[(c - '(') % 4] = charset;
	    }
	  else if (c == '$')
	    {
	      /* Designation sequence for a charset of dimension 2.  */
	      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, mask & mask_found);
	      if (c >= '@' && c <= 'B')
		/* Designation for JISX0208.1978, GB2312, or JISX0208.  */
		reg[0] = charset = iso_charset_table[1][0][c];
	      else if (c >= '(' && c <= '/')
		{
		  ONE_MORE_BYTE_CHECK_MULTIBYTE (c1, multibytep,
						 mask & mask_found);
		  if (c1 < ' ' || c1 >= 0x80
		      || (charset = iso_charset_table[1][c >= ','][c1]) < 0)
		    /* Invalid designation sequence.  Just ignore.  */
		    break;
		  reg[(c - '(') % 4] = charset;
		}
	      else
		/* Invalid designation sequence.  Just ignore.  */
		break;
	    }
	  else if (c == 'N' || c == 'O')
	    {
	      /* ESC <Fe> for SS2 or SS3.  */
	      mask &= CODING_CATEGORY_MASK_ISO_7_ELSE;
	      break;
	    }
	  else if (c >= '0' && c <= '4')
	    {
	      /* ESC <Fp> for start/end composition.  */
	      if (COMPOSITION_OK (CODING_CATEGORY_IDX_ISO_7))
		mask_found |= CODING_CATEGORY_MASK_ISO_7;
	      else
		mask &= ~CODING_CATEGORY_MASK_ISO_7;
	      if (COMPOSITION_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT))
		mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT;
	      else
		mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT;
	      if (COMPOSITION_OK (CODING_CATEGORY_IDX_ISO_8_1))
		mask_found |= CODING_CATEGORY_MASK_ISO_8_1;
	      else
		mask &= ~CODING_CATEGORY_MASK_ISO_8_1;
	      if (COMPOSITION_OK (CODING_CATEGORY_IDX_ISO_8_2))
		mask_found |= CODING_CATEGORY_MASK_ISO_8_2;
	      else
		mask &= ~CODING_CATEGORY_MASK_ISO_8_2;
	      if (COMPOSITION_OK (CODING_CATEGORY_IDX_ISO_7_ELSE))
		mask_found |= CODING_CATEGORY_MASK_ISO_7_ELSE;
	      else
		mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE;
	      if (COMPOSITION_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))
		mask_found |= CODING_CATEGORY_MASK_ISO_8_ELSE;
	      else
		mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE;
	      break;
	    }
	  else
	    /* Invalid escape sequence.  Just ignore.  */
	    break;

	  /* We found a valid designation sequence for CHARSET.  */
	  mask &= ~CODING_CATEGORY_MASK_ISO_8BIT;
	  c = MAKE_CHAR (charset, 0, 0);
	  if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset, c))
	    mask_found |= CODING_CATEGORY_MASK_ISO_7;
	  else
	    mask &= ~CODING_CATEGORY_MASK_ISO_7;
	  if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset, c))
	    mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT;
	  else
	    mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT;
	  if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset, c))
	    mask_found |= CODING_CATEGORY_MASK_ISO_7_ELSE;
	  else
	    mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE;
	  if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset, c))
	    mask_found |= CODING_CATEGORY_MASK_ISO_8_ELSE;
	  else
	    mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE;
	  break;

	case ISO_CODE_SO:
	  if (inhibit_iso_escape_detection)
	    break;
	  single_shifting = 0;
	  if (shift_out == 0
	      && (reg[1] >= 0
		  || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE)
		  || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE)))
	    {
	      /* Locking shift out.  */
	      mask &= ~CODING_CATEGORY_MASK_ISO_7BIT;
	      mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT;
	    }
	  break;

	case ISO_CODE_SI:
	  if (inhibit_iso_escape_detection)
	    break;
	  single_shifting = 0;
	  if (shift_out == 1)
	    {
	      /* Locking shift in.  */
	      mask &= ~CODING_CATEGORY_MASK_ISO_7BIT;
	      mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT;
	    }
	  break;

	case ISO_CODE_CSI:
	  single_shifting = 0;
	case ISO_CODE_SS2:
	case ISO_CODE_SS3:
	  {
	    int newmask = CODING_CATEGORY_MASK_ISO_8_ELSE;

	    if (inhibit_iso_escape_detection)
	      break;
	    if (c != ISO_CODE_CSI)
	      {
		if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags
		    & CODING_FLAG_ISO_SINGLE_SHIFT)
		  newmask |= CODING_CATEGORY_MASK_ISO_8_1;
		if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags
		    & CODING_FLAG_ISO_SINGLE_SHIFT)
		  newmask |= CODING_CATEGORY_MASK_ISO_8_2;
		single_shifting = 1;
	      }
	    if (VECTORP (Vlatin_extra_code_table)
		&& !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c]))
	      {
		if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags
		    & CODING_FLAG_ISO_LATIN_EXTRA)
		  newmask |= CODING_CATEGORY_MASK_ISO_8_1;
		if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags
		    & CODING_FLAG_ISO_LATIN_EXTRA)
		  newmask |= CODING_CATEGORY_MASK_ISO_8_2;
	      }
	    mask &= newmask;
	    mask_found |= newmask;
	  }
	  break;

	default:
	  if (c < 0x80)
	    {
	      single_shifting = 0;
	      break;
	    }
	  else if (c < 0xA0)
	    {
	      single_shifting = 0;
	      if (VECTORP (Vlatin_extra_code_table)
		  && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c]))
		{
		  int newmask = 0;

		  if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags
		      & CODING_FLAG_ISO_LATIN_EXTRA)
		    newmask |= CODING_CATEGORY_MASK_ISO_8_1;
		  if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags
		      & CODING_FLAG_ISO_LATIN_EXTRA)
		    newmask |= CODING_CATEGORY_MASK_ISO_8_2;
		  mask &= newmask;
		  mask_found |= newmask;
		}
	      else
		return 0;
	    }
	  else
	    {
	      mask &= ~(CODING_CATEGORY_MASK_ISO_7BIT
			| CODING_CATEGORY_MASK_ISO_7_ELSE);
	      mask_found |= CODING_CATEGORY_MASK_ISO_8_1;
	      /* Check the length of succeeding codes of the range
                 0xA0..0FF.  If the byte length is odd, we exclude
                 CODING_CATEGORY_MASK_ISO_8_2.  We can check this only
                 when we are not single shifting.  */
	      if (!single_shifting
		  && mask & CODING_CATEGORY_MASK_ISO_8_2)
		{
		  int i = 1;

		  c = -1;
		  while (src < src_end)
		    {
		      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep,
						     mask & mask_found);
		      if (c < 0xA0)
			break;
		      i++;
		    }

		  if (i & 1 && src < src_end)
		    mask &= ~CODING_CATEGORY_MASK_ISO_8_2;
		  else
		    mask_found |= CODING_CATEGORY_MASK_ISO_8_2;
		  if (c >= 0)
		    /* This means that we have read one extra byte.  */
		    goto retry;
		}
	    }
	  break;
	}
    }
  return (mask & mask_found);
}

/* Decode a character of which charset is CHARSET, the 1st position
   code is C1, the 2nd position code is C2, and return the decoded
   character code.  If the variable `translation_table' is non-nil,
   returned the translated code.  */

#define DECODE_ISO_CHARACTER(charset, c1, c2)	\
  (NILP (translation_table)			\
   ? MAKE_CHAR (charset, c1, c2)		\
   : translate_char (translation_table, -1, charset, c1, c2))

/* Set designation state into CODING.  */
#define DECODE_DESIGNATION(reg, dimension, chars, final_char)		   \
  do {									   \
    int charset, c;							   \
    									   \
    if (final_char < '0' || final_char >= 128)				   \
      goto label_invalid_code;						   \
    charset = ISO_CHARSET_TABLE (make_number (dimension),		   \
				 make_number (chars),			   \
				 make_number (final_char));		   \
    c = MAKE_CHAR (charset, 0, 0);					   \
    if (charset >= 0							   \
	&& (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg \
	    || CODING_SAFE_CHAR_P (safe_chars, c)))			   \
      {									   \
	if (coding->spec.iso2022.last_invalid_designation_register == 0	   \
	    && reg == 0							   \
	    && charset == CHARSET_ASCII)				   \
	  {								   \
	    /* We should insert this designation sequence as is so	   \
               that it is surely written back to a file.  */		   \
	    coding->spec.iso2022.last_invalid_designation_register = -1;   \
	    goto label_invalid_code;					   \
	  }								   \
	coding->spec.iso2022.last_invalid_designation_register = -1;	   \
        if ((coding->mode & CODING_MODE_DIRECTION)			   \
	    && CHARSET_REVERSE_CHARSET (charset) >= 0)			   \
          charset = CHARSET_REVERSE_CHARSET (charset);			   \
        CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset;		   \
      }									   \
    else								   \
      {									   \
	coding->spec.iso2022.last_invalid_designation_register = reg;	   \
	goto label_invalid_code;					   \
      }									   \
  } while (0)

/* Allocate a memory block for storing information about compositions.
   The block is chained to the already allocated blocks.  */

void
coding_allocate_composition_data (coding, char_offset)
     struct coding_system *coding;
     int char_offset;
{
  struct composition_data *cmp_data
    = (struct composition_data *) xmalloc (sizeof *cmp_data);

  cmp_data->char_offset = char_offset;
  cmp_data->used = 0;
  cmp_data->prev = coding->cmp_data;
  cmp_data->next = NULL;
  if (coding->cmp_data)
    coding->cmp_data->next = cmp_data;
  coding->cmp_data = cmp_data;
  coding->cmp_data_start = 0;
  coding->composing = COMPOSITION_NO;
}

/* Handle composition start sequence ESC 0, ESC 2, ESC 3, or ESC 4.
   ESC 0 : relative composition : ESC 0 CHAR ... ESC 1
   ESC 2 : rulebase composition : ESC 2 CHAR RULE CHAR RULE ... CHAR ESC 1
   ESC 3 : altchar composition :  ESC 3 ALT ... ESC 0 CHAR ... ESC 1
   ESC 4 : alt&rule composition : ESC 4 ALT RULE .. ALT ESC 0 CHAR ... ESC 1
  */

#define DECODE_COMPOSITION_START(c1)					   \
  do {									   \
    if (coding->composing == COMPOSITION_DISABLED)			   \
      {									   \
      	*dst++ = ISO_CODE_ESC;						   \
	*dst++ = c1 & 0x7f;						   \
	coding->produced_char += 2;					   \
      }									   \
    else if (!COMPOSING_P (coding))					   \
      {									   \
	/* This is surely the start of a composition.  We must be sure	   \
           that coding->cmp_data has enough space to store the		   \
           information about the composition.  If not, terminate the	   \
           current decoding loop, allocate one more memory block for	   \
           coding->cmp_data in the caller, then start the decoding	   \
           loop again.  We can't allocate memory here directly because	   \
           it may cause buffer/string relocation.  */			   \
	if (!coding->cmp_data						   \
	    || (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH \
		>= COMPOSITION_DATA_SIZE))				   \
	  {								   \
	    coding->result = CODING_FINISH_INSUFFICIENT_CMP;		   \
	    goto label_end_of_loop;					   \
	  }								   \
	coding->composing = (c1 == '0' ? COMPOSITION_RELATIVE		   \
			     : c1 == '2' ? COMPOSITION_WITH_RULE	   \
			     : c1 == '3' ? COMPOSITION_WITH_ALTCHARS	   \
			     : COMPOSITION_WITH_RULE_ALTCHARS);		   \
	CODING_ADD_COMPOSITION_START (coding, coding->produced_char,	   \
				      coding->composing);		   \
	coding->composition_rule_follows = 0;				   \
      }									   \
    else								   \
      {									   \
	/* We are already handling a composition.  If the method is	   \
           the following two, the codes following the current escape	   \
           sequence are actual characters stored in a buffer.  */	   \
	if (coding->composing == COMPOSITION_WITH_ALTCHARS		   \
	    || coding->composing == COMPOSITION_WITH_RULE_ALTCHARS)	   \
	  {								   \
	    coding->composing = COMPOSITION_RELATIVE;			   \
	    coding->composition_rule_follows = 0;			   \
	  }								   \
      }									   \
  } while (0)

/* Handle composition end sequence ESC 1.  */

#define DECODE_COMPOSITION_END(c1)					\
  do {									\
    if (! COMPOSING_P (coding))						\
      {									\
	*dst++ = ISO_CODE_ESC;						\
	*dst++ = c1;							\
	coding->produced_char += 2;					\
      }									\
    else								\
      {									\
	CODING_ADD_COMPOSITION_END (coding, coding->produced_char);	\
	coding->composing = COMPOSITION_NO;				\
      }									\
  } while (0)

/* Decode a composition rule from the byte C1 (and maybe one more byte
   from SRC) and store one encoded composition rule in
   coding->cmp_data.  */

#define DECODE_COMPOSITION_RULE(c1)					\
  do {									\
    int rule = 0;							\
    (c1) -= 32;								\
    if (c1 < 81)		/* old format (before ver.21) */	\
      {									\
	int gref = (c1) / 9;						\
	int nref = (c1) % 9;						\
	if (gref == 4) gref = 10;					\
	if (nref == 4) nref = 10;					\
	rule = COMPOSITION_ENCODE_RULE (gref, nref);			\
      }									\
    else if (c1 < 93)		/* new format (after ver.21) */		\
      {									\
	ONE_MORE_BYTE (c2);						\
	rule = COMPOSITION_ENCODE_RULE (c1 - 81, c2 - 32);		\
      }									\
    CODING_ADD_COMPOSITION_COMPONENT (coding, rule);			\
    coding->composition_rule_follows = 0;				\
  } while (0)


/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".  */

static void
decode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  const unsigned char *src = source;
  const unsigned char *src_end = source + src_bytes;
  unsigned char *dst = destination;
  unsigned char *dst_end = destination + dst_bytes;
  /* Charsets invoked to graphic plane 0 and 1 respectively.  */
  int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
  int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source code
     (within macro ONE_MORE_BYTE), or when there's not enough
     destination area to produce a character (within macro
     EMIT_CHAR).  */
  const unsigned char *src_base;
  int c, charset;
  Lisp_Object translation_table;
  Lisp_Object safe_chars;

  safe_chars = coding_safe_chars (coding->symbol);

  if (NILP (Venable_character_translation))
    translation_table = Qnil;
  else
    {
      translation_table = coding->translation_table_for_decode;
      if (NILP (translation_table))
	translation_table = Vstandard_translation_table_for_decode;
    }

  coding->result = CODING_FINISH_NORMAL;

  while (1)
    {
      int c1, c2 = 0;

      src_base = src;
      ONE_MORE_BYTE (c1);

      /* We produce no character or one character.  */
      switch (iso_code_class [c1])
	{
	case ISO_0x20_or_0x7F:
	  if (COMPOSING_P (coding) && coding->composition_rule_follows)
	    {
	      DECODE_COMPOSITION_RULE (c1);
	      continue;
	    }
	  if (charset0 < 0 || CHARSET_CHARS (charset0) == 94)
	    {
	      /* This is SPACE or DEL.  */
	      charset = CHARSET_ASCII;
	      break;
	    }
	  /* This is a graphic character, we fall down ...  */

	case ISO_graphic_plane_0:
	  if (COMPOSING_P (coding) && coding->composition_rule_follows)
	    {
	      DECODE_COMPOSITION_RULE (c1);
	      continue;
	    }
	  charset = charset0;
	  break;

	case ISO_0xA0_or_0xFF:
	  if (charset1 < 0 || CHARSET_CHARS (charset1) == 94
	      || coding->flags & CODING_FLAG_ISO_SEVEN_BITS)
	    goto label_invalid_code;
	  /* This is a graphic character, we fall down ... */

	case ISO_graphic_plane_1:
	  if (charset1 < 0)
	    goto label_invalid_code;
	  charset = charset1;
	  break;

	case ISO_control_0:
	  if (COMPOSING_P (coding))
	    DECODE_COMPOSITION_END ('1');

	  /* All ISO2022 control characters in this class have the
             same representation in Emacs internal format.  */
	  if (c1 == '\n'
	      && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
	      && (coding->eol_type == CODING_EOL_CR
		  || coding->eol_type == CODING_EOL_CRLF))
	    {
	      coding->result = CODING_FINISH_INCONSISTENT_EOL;
	      goto label_end_of_loop;
	    }
	  charset = CHARSET_ASCII;
	  break;

	case ISO_control_1:
	  if (COMPOSING_P (coding))
	    DECODE_COMPOSITION_END ('1');
	  goto label_invalid_code;

	case ISO_carriage_return:
	  if (COMPOSING_P (coding))
	    DECODE_COMPOSITION_END ('1');

	  if (coding->eol_type == CODING_EOL_CR)
	    c1 = '\n';
	  else if (coding->eol_type == CODING_EOL_CRLF)
	    {
	      ONE_MORE_BYTE (c1);
	      if (c1 != ISO_CODE_LF)
		{
		  src--;
		  c1 = '\r';
		}
	    }
	  charset = CHARSET_ASCII;
	  break;

	case ISO_shift_out:
	  if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)
	      || CODING_SPEC_ISO_DESIGNATION (coding, 1) < 0)
	    goto label_invalid_code;
	  CODING_SPEC_ISO_INVOCATION (coding, 0) = 1;
	  charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
	  continue;

	case ISO_shift_in:
	  if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT))
	    goto label_invalid_code;
	  CODING_SPEC_ISO_INVOCATION (coding, 0) = 0;
	  charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
	  continue;

	case ISO_single_shift_2_7:
	case ISO_single_shift_2:
	  if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT))
	    goto label_invalid_code;
	  /* SS2 is handled as an escape sequence of ESC 'N' */
	  c1 = 'N';
	  goto label_escape_sequence;

	case ISO_single_shift_3:
	  if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT))
	    goto label_invalid_code;
	  /* SS2 is handled as an escape sequence of ESC 'O' */
	  c1 = 'O';
	  goto label_escape_sequence;

	case ISO_control_sequence_introducer:
	  /* CSI is handled as an escape sequence of ESC '[' ...  */
	  c1 = '[';
	  goto label_escape_sequence;

	case ISO_escape:
	  ONE_MORE_BYTE (c1);
	label_escape_sequence:
	  /* Escape sequences handled by Emacs are invocation,
	     designation, direction specification, and character
	     composition specification.  */
	  switch (c1)
	    {
	    case '&':		/* revision of following character set */
	      ONE_MORE_BYTE (c1);
	      if (!(c1 >= '@' && c1 <= '~'))
		goto label_invalid_code;
	      ONE_MORE_BYTE (c1);
	      if (c1 != ISO_CODE_ESC)
		goto label_invalid_code;
	      ONE_MORE_BYTE (c1);
	      goto label_escape_sequence;

	    case '$':		/* designation of 2-byte character set */
	      if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION))
		goto label_invalid_code;
	      ONE_MORE_BYTE (c1);
	      if (c1 >= '@' && c1 <= 'B')
		{	/* designation of JISX0208.1978, GB2312.1980,
			   or JISX0208.1980 */
		  DECODE_DESIGNATION (0, 2, 94, c1);
		}
	      else if (c1 >= 0x28 && c1 <= 0x2B)
		{	/* designation of DIMENSION2_CHARS94 character set */
		  ONE_MORE_BYTE (c2);
		  DECODE_DESIGNATION (c1 - 0x28, 2, 94, c2);
		}
	      else if (c1 >= 0x2C && c1 <= 0x2F)
		{	/* designation of DIMENSION2_CHARS96 character set */
		  ONE_MORE_BYTE (c2);
		  DECODE_DESIGNATION (c1 - 0x2C, 2, 96, c2);
		}
	      else
		goto label_invalid_code;
	      /* We must update these variables now.  */
	      charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
	      charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
	      continue;

	    case 'n':		/* invocation of locking-shift-2 */
	      if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)
		  || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0)
		goto label_invalid_code;
	      CODING_SPEC_ISO_INVOCATION (coding, 0) = 2;
	      charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
	      continue;

	    case 'o':		/* invocation of locking-shift-3 */
	      if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)
		  || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0)
		goto label_invalid_code;
	      CODING_SPEC_ISO_INVOCATION (coding, 0) = 3;
	      charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
	      continue;

	    case 'N':		/* invocation of single-shift-2 */
	      if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)
		  || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0)
		goto label_invalid_code;
	      charset = CODING_SPEC_ISO_DESIGNATION (coding, 2);
	      ONE_MORE_BYTE (c1);
	      if (c1 < 0x20 || (c1 >= 0x80 && c1 < 0xA0))
		goto label_invalid_code;
	      break;

	    case 'O':		/* invocation of single-shift-3 */
	      if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)
		  || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0)
		goto label_invalid_code;
	      charset = CODING_SPEC_ISO_DESIGNATION (coding, 3);
	      ONE_MORE_BYTE (c1);
	      if (c1 < 0x20 || (c1 >= 0x80 && c1 < 0xA0))
		goto label_invalid_code;
	      break;

	    case '0': case '2':	case '3': case '4': /* start composition */
	      DECODE_COMPOSITION_START (c1);
	      continue;

	    case '1':		/* end composition */
	      DECODE_COMPOSITION_END (c1);
	      continue;

	    case '[':		/* specification of direction */
	      if (coding->flags & CODING_FLAG_ISO_NO_DIRECTION)
		goto label_invalid_code;
	      /* For the moment, nested direction is not supported.
		 So, `coding->mode & CODING_MODE_DIRECTION' zero means
		 left-to-right, and nonzero means right-to-left.  */
	      ONE_MORE_BYTE (c1);
	      switch (c1)
		{
		case ']':	/* end of the current direction */
		  coding->mode &= ~CODING_MODE_DIRECTION;

		case '0':	/* end of the current direction */
		case '1':	/* start of left-to-right direction */
		  ONE_MORE_BYTE (c1);
		  if (c1 == ']')
		    coding->mode &= ~CODING_MODE_DIRECTION;
		  else
		    goto label_invalid_code;
		  break;

		case '2':	/* start of right-to-left direction */
		  ONE_MORE_BYTE (c1);
		  if (c1 == ']')
		    coding->mode |= CODING_MODE_DIRECTION;
		  else
		    goto label_invalid_code;
		  break;

		default:
		  goto label_invalid_code;
		}
	      continue;

	    case '%':
	      if (COMPOSING_P (coding))
		DECODE_COMPOSITION_END ('1');
	      ONE_MORE_BYTE (c1);
	      if (c1 == '/')
		{
		  /* CTEXT extended segment:
		     ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
		     We keep these bytes as is for the moment.
		     They may be decoded by post-read-conversion.  */
		  int dim, M, L;
		  int size, required;
		  int produced_chars;

		  ONE_MORE_BYTE (dim);
		  ONE_MORE_BYTE (M);
		  ONE_MORE_BYTE (L);
		  size = ((M - 128) * 128) + (L - 128);
		  required = 8 + size * 2;
		  if (dst + required > (dst_bytes ? dst_end : src))
		    goto label_end_of_loop;
		  *dst++ = ISO_CODE_ESC;
		  *dst++ = '%';
		  *dst++ = '/';
		  *dst++ = dim;
		  produced_chars = 4;
		  dst += CHAR_STRING (M, dst), produced_chars++;
		  dst += CHAR_STRING (L, dst), produced_chars++;
		  while (size-- > 0)
		    {
		      ONE_MORE_BYTE (c1);
		      dst += CHAR_STRING (c1, dst), produced_chars++;
		    }
		  coding->produced_char += produced_chars;
		}
	      else if (c1 == 'G')
		{
		  unsigned char *d = dst;
		  int produced_chars;

		  /* XFree86 extension for embedding UTF-8 in CTEXT:
		     ESC % G --UTF-8-BYTES-- ESC % @
		     We keep these bytes as is for the moment.
		     They may be decoded by post-read-conversion.  */
		  if (d + 6 > (dst_bytes ? dst_end : src))
		    goto label_end_of_loop;
		  *d++ = ISO_CODE_ESC;
		  *d++ = '%';
		  *d++ = 'G';
		  produced_chars = 3;
		  while (d + 1 < (dst_bytes ? dst_end : src))
		    {
		      ONE_MORE_BYTE (c1);
		      if (c1 == ISO_CODE_ESC
			  && src + 1 < src_end
			  && src[0] == '%'
			  && src[1] == '@')
			{
			  src += 2;
			  break;
			}
		      d += CHAR_STRING (c1, d), produced_chars++;
		    }
		  if (d + 3 > (dst_bytes ? dst_end : src))
		    goto label_end_of_loop;
		  *d++ = ISO_CODE_ESC;
		  *d++ = '%';
		  *d++ = '@';
		  dst = d;
		  coding->produced_char += produced_chars + 3;
		}
	      else
		goto label_invalid_code;
	      continue;

	    default:
	      if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION))
		goto label_invalid_code;
	      if (c1 >= 0x28 && c1 <= 0x2B)
		{	/* designation of DIMENSION1_CHARS94 character set */
		  ONE_MORE_BYTE (c2);
		  DECODE_DESIGNATION (c1 - 0x28, 1, 94, c2);
		}
	      else if (c1 >= 0x2C && c1 <= 0x2F)
		{	/* designation of DIMENSION1_CHARS96 character set */
		  ONE_MORE_BYTE (c2);
		  DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2);
		}
	      else
		goto label_invalid_code;
	      /* We must update these variables now.  */
	      charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
	      charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
	      continue;
	    }
	}

      /* Now we know CHARSET and 1st position code C1 of a character.
         Produce a multibyte sequence for that character while getting
         2nd position code C2 if necessary.  */
      if (CHARSET_DIMENSION (charset) == 2)
	{
	  ONE_MORE_BYTE (c2);
	  if (c1 < 0x80 ? c2 < 0x20 || c2 >= 0x80 : c2 < 0xA0)
	    /* C2 is not in a valid range.  */
	    goto label_invalid_code;
	}
      c = DECODE_ISO_CHARACTER (charset, c1, c2);
      EMIT_CHAR (c);
      continue;

    label_invalid_code:
      coding->errors++;
      if (COMPOSING_P (coding))
	DECODE_COMPOSITION_END ('1');
      src = src_base;
      c = *src++;
      if (! NILP (translation_table))
	c = translate_char (translation_table, c, 0, 0, 0);
      EMIT_CHAR (c);
    }

 label_end_of_loop:
  coding->consumed = coding->consumed_char = src_base - source;
  coding->produced = dst - destination;
  return;
}


/* ISO2022 encoding stuff.  */

/*
   It is not enough to say just "ISO2022" on encoding, we have to
   specify more details.  In Emacs, each ISO2022 coding system
   variant has the following specifications:
	1. Initial designation to G0 through G3.
	2. Allows short-form designation?
	3. ASCII should be designated to G0 before control characters?
	4. ASCII should be designated to G0 at end of line?
	5. 7-bit environment or 8-bit environment?
	6. Use locking-shift?
	7. Use Single-shift?
   And the following two are only for Japanese:
	8. Use ASCII in place of JIS0201-1976-Roman?
	9. Use JISX0208-1983 in place of JISX0208-1978?
   These specifications are encoded in `coding->flags' as flag bits
   defined by macros CODING_FLAG_ISO_XXX.  See `coding.h' for more
   details.
*/

/* Produce codes (escape sequence) for designating CHARSET to graphic
   register REG at DST, and increment DST.  If <final-char> of CHARSET is
   '@', 'A', or 'B' and the coding system CODING allows, produce
   designation sequence of short-form.  */

#define ENCODE_DESIGNATION(charset, reg, coding)			\
  do {									\
    unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset);	\
    char *intermediate_char_94 = "()*+";				\
    char *intermediate_char_96 = ",-./";				\
    int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset);	\
    									\
    if (revision < 255)							\
      {									\
	*dst++ = ISO_CODE_ESC;						\
	*dst++ = '&';							\
	*dst++ = '@' + revision;					\
      }									\
    *dst++ = ISO_CODE_ESC;						\
    if (CHARSET_DIMENSION (charset) == 1)				\
      {									\
	if (CHARSET_CHARS (charset) == 94)				\
	  *dst++ = (unsigned char) (intermediate_char_94[reg]);		\
	else								\
	  *dst++ = (unsigned char) (intermediate_char_96[reg]);		\
      }									\
    else								\
      {									\
	*dst++ = '$';							\
	if (CHARSET_CHARS (charset) == 94)				\
	  {								\
	    if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM)		\
		|| reg != 0						\
		|| final_char < '@' || final_char > 'B')		\
	      *dst++ = (unsigned char) (intermediate_char_94[reg]);	\
	  }								\
	else								\
	  *dst++ = (unsigned char) (intermediate_char_96[reg]);		\
      }									\
    *dst++ = final_char;						\
    CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset;		\
  } while (0)

/* The following two macros produce codes (control character or escape
   sequence) for ISO2022 single-shift functions (single-shift-2 and
   single-shift-3).  */

#define ENCODE_SINGLE_SHIFT_2				\
  do {							\
    if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS)	\
      *dst++ = ISO_CODE_ESC, *dst++ = 'N';		\
    else						\
      *dst++ = ISO_CODE_SS2;				\
    CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1;	\
  } while (0)

#define ENCODE_SINGLE_SHIFT_3				\
  do {							\
    if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS)	\
      *dst++ = ISO_CODE_ESC, *dst++ = 'O';		\
    else						\
      *dst++ = ISO_CODE_SS3;				\
    CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1;	\
  } while (0)

/* The following four macros produce codes (control character or
   escape sequence) for ISO2022 locking-shift functions (shift-in,
   shift-out, locking-shift-2, and locking-shift-3).  */

#define ENCODE_SHIFT_IN				\
  do {						\
    *dst++ = ISO_CODE_SI;			\
    CODING_SPEC_ISO_INVOCATION (coding, 0) = 0;	\
  } while (0)

#define ENCODE_SHIFT_OUT			\
  do {						\
    *dst++ = ISO_CODE_SO;			\
    CODING_SPEC_ISO_INVOCATION (coding, 0) = 1;	\
  } while (0)

#define ENCODE_LOCKING_SHIFT_2			\
  do {						\
    *dst++ = ISO_CODE_ESC, *dst++ = 'n';	\
    CODING_SPEC_ISO_INVOCATION (coding, 0) = 2;	\
  } while (0)

#define ENCODE_LOCKING_SHIFT_3			\
  do {						\
    *dst++ = ISO_CODE_ESC, *dst++ = 'o';	\
    CODING_SPEC_ISO_INVOCATION (coding, 0) = 3;	\
  } while (0)

/* Produce codes for a DIMENSION1 character whose character set is
   CHARSET and whose position-code is C1.  Designation and invocation
   sequences are also produced in advance if necessary.  */

#define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1)			\
  do {									\
    if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding))			\
      {									\
	if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS)			\
	  *dst++ = c1 & 0x7F;						\
	else								\
	  *dst++ = c1 | 0x80;						\
	CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0;			\
	break;								\
      }									\
    else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0))	\
      {									\
	*dst++ = c1 & 0x7F;						\
	break;								\
      }									\
    else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1))	\
      {									\
	*dst++ = c1 | 0x80;						\
	break;								\
      }									\
    else								\
      /* Since CHARSET is not yet invoked to any graphic planes, we	\
	 must invoke it, or, at first, designate it to some graphic	\
	 register.  Then repeat the loop to actually produce the	\
	 character.  */							\
      dst = encode_invocation_designation (charset, coding, dst);	\
  } while (1)

/* Produce codes for a DIMENSION2 character whose character set is
   CHARSET and whose position-codes are C1 and C2.  Designation and
   invocation codes are also produced in advance if necessary.  */

#define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2)		\
  do {									\
    if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding))			\
      {									\
	if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS)			\
	  *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F;			\
	else								\
	  *dst++ = c1 | 0x80, *dst++ = c2 | 0x80;			\
	CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0;			\
	break;								\
      }									\
    else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0))	\
      {									\
	*dst++ = c1 & 0x7F, *dst++= c2 & 0x7F;				\
	break;								\
      }									\
    else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1))	\
      {									\
	*dst++ = c1 | 0x80, *dst++= c2 | 0x80;				\
	break;								\
      }									\
    else								\
      /* Since CHARSET is not yet invoked to any graphic planes, we	\
	 must invoke it, or, at first, designate it to some graphic	\
	 register.  Then repeat the loop to actually produce the	\
	 character.  */							\
      dst = encode_invocation_designation (charset, coding, dst);	\
  } while (1)

#define ENCODE_ISO_CHARACTER(c)					\
  do {								\
    int charset, c1, c2;					\
    								\
    SPLIT_CHAR (c, charset, c1, c2);				\
    if (CHARSET_DEFINED_P (charset))				\
      {								\
	if (CHARSET_DIMENSION (charset) == 1)			\
	  {							\
	    if (charset == CHARSET_ASCII			\
		&& coding->flags & CODING_FLAG_ISO_USE_ROMAN)	\
	      charset = charset_latin_jisx0201;			\
	    ENCODE_ISO_CHARACTER_DIMENSION1 (charset, c1);	\
	  }							\
	else							\
	  {							\
	    if (charset == charset_jisx0208			\
		&& coding->flags & CODING_FLAG_ISO_USE_OLDJIS)	\
	      charset = charset_jisx0208_1978;			\
	    ENCODE_ISO_CHARACTER_DIMENSION2 (charset, c1, c2);	\
	  }							\
      }								\
    else							\
      {								\
	*dst++ = c1;						\
	if (c2 >= 0)						\
	  *dst++ = c2;						\
      }								\
  } while (0)


/* Instead of encoding character C, produce one or two `?'s.  */

#define ENCODE_UNSAFE_CHARACTER(c)				\
  do {								\
    ENCODE_ISO_CHARACTER (CODING_REPLACEMENT_CHARACTER);	\
    if (CHARSET_WIDTH (CHAR_CHARSET (c)) > 1)			\
      ENCODE_ISO_CHARACTER (CODING_REPLACEMENT_CHARACTER);	\
  } while (0)


/* Produce designation and invocation codes at a place pointed by DST
   to use CHARSET.  The element `spec.iso2022' of *CODING is updated.
   Return new DST.  */

unsigned char *
encode_invocation_designation (charset, coding, dst)
     int charset;
     struct coding_system *coding;
     unsigned char *dst;
{
  int reg;			/* graphic register number */

  /* At first, check designations.  */
  for (reg = 0; reg < 4; reg++)
    if (charset == CODING_SPEC_ISO_DESIGNATION (coding, reg))
      break;

  if (reg >= 4)
    {
      /* CHARSET is not yet designated to any graphic registers.  */
      /* At first check the requested designation.  */
      reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset);
      if (reg == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)
	/* Since CHARSET requests no special designation, designate it
	   to graphic register 0.  */
	reg = 0;

      ENCODE_DESIGNATION (charset, reg, coding);
    }

  if (CODING_SPEC_ISO_INVOCATION (coding, 0) != reg
      && CODING_SPEC_ISO_INVOCATION (coding, 1) != reg)
    {
      /* Since the graphic register REG is not invoked to any graphic
	 planes, invoke it to graphic plane 0.  */
      switch (reg)
	{
	case 0:			/* graphic register 0 */
	  ENCODE_SHIFT_IN;
	  break;

	case 1:			/* graphic register 1 */
	  ENCODE_SHIFT_OUT;
	  break;

	case 2:			/* graphic register 2 */
	  if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)
	    ENCODE_SINGLE_SHIFT_2;
	  else
	    ENCODE_LOCKING_SHIFT_2;
	  break;

	case 3:			/* graphic register 3 */
	  if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)
	    ENCODE_SINGLE_SHIFT_3;
	  else
	    ENCODE_LOCKING_SHIFT_3;
	  break;
	}
    }

  return dst;
}

/* Produce 2-byte codes for encoded composition rule RULE.  */

#define ENCODE_COMPOSITION_RULE(rule)		\
  do {						\
    int gref, nref;				\
    COMPOSITION_DECODE_RULE (rule, gref, nref);	\
    *dst++ = 32 + 81 + gref;			\
    *dst++ = 32 + nref;				\
  } while (0)

/* Produce codes for indicating the start of a composition sequence
   (ESC 0, ESC 3, or ESC 4).  DATA points to an array of integers
   which specify information about the composition.  See the comment
   in coding.h for the format of DATA.  */

#define ENCODE_COMPOSITION_START(coding, data)				\
  do {									\
    coding->composing = data[3];					\
    *dst++ = ISO_CODE_ESC;						\
    if (coding->composing == COMPOSITION_RELATIVE)			\
      *dst++ = '0';							\
    else								\
      {									\
	*dst++ = (coding->composing == COMPOSITION_WITH_ALTCHARS	\
		  ? '3' : '4');						\
	coding->cmp_data_index = coding->cmp_data_start + 4;		\
	coding->composition_rule_follows = 0;				\
      }									\
  } while (0)

/* Produce codes for indicating the end of the current composition.  */

#define ENCODE_COMPOSITION_END(coding, data)			\
  do {								\
    *dst++ = ISO_CODE_ESC;					\
    *dst++ = '1';						\
    coding->cmp_data_start += data[0];				\
    coding->composing = COMPOSITION_NO;				\
    if (coding->cmp_data_start == coding->cmp_data->used	\
	&& coding->cmp_data->next)				\
      {								\
	coding->cmp_data = coding->cmp_data->next;		\
	coding->cmp_data_start = 0;				\
      }								\
  } while (0)

/* Produce composition start sequence ESC 0.  Here, this sequence
   doesn't mean the start of a new composition but means that we have
   just produced components (alternate chars and composition rules) of
   the composition and the actual text follows in SRC.  */

#define ENCODE_COMPOSITION_FAKE_START(coding)	\
  do {						\
    *dst++ = ISO_CODE_ESC;			\
    *dst++ = '0';				\
    coding->composing = COMPOSITION_RELATIVE;	\
  } while (0)

/* The following three macros produce codes for indicating direction
   of text.  */
#define ENCODE_CONTROL_SEQUENCE_INTRODUCER		\
  do {							\
    if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS)	\
      *dst++ = ISO_CODE_ESC, *dst++ = '[';		\
    else						\
      *dst++ = ISO_CODE_CSI;				\
  } while (0)

#define ENCODE_DIRECTION_R2L	\
  ENCODE_CONTROL_SEQUENCE_INTRODUCER (dst), *dst++ = '2', *dst++ = ']'

#define ENCODE_DIRECTION_L2R	\
  ENCODE_CONTROL_SEQUENCE_INTRODUCER (dst), *dst++ = '0', *dst++ = ']'

/* Produce codes for designation and invocation to reset the graphic
   planes and registers to initial state.  */
#define ENCODE_RESET_PLANE_AND_REGISTER					    \
  do {									    \
    int reg;								    \
    if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0)			    \
      ENCODE_SHIFT_IN;							    \
    for (reg = 0; reg < 4; reg++)					    \
      if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0	    \
	  && (CODING_SPEC_ISO_DESIGNATION (coding, reg)			    \
	      != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg)))	    \
	ENCODE_DESIGNATION						    \
	  (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \
  } while (0)

/* Produce designation sequences of charsets in the line started from
   SRC to a place pointed by DST, and return updated DST.

   If the current block ends before any end-of-line, we may fail to
   find all the necessary designations.  */

static unsigned char *
encode_designation_at_bol (coding, translation_table, src, src_end, dst)
     struct coding_system *coding;
     Lisp_Object translation_table;
     const unsigned char *src, *src_end;
     unsigned char *dst;
{
  int charset, c, found = 0, reg;
  /* Table of charsets to be designated to each graphic register.  */
  int r[4];

  for (reg = 0; reg < 4; reg++)
    r[reg] = -1;

  while (found < 4)
    {
      ONE_MORE_CHAR (c);
      if (c == '\n')
	break;

      charset = CHAR_CHARSET (c);
      reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset);
      if (reg != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION && r[reg] < 0)
	{
	  found++;
	  r[reg] = charset;
	}
    }

 label_end_of_loop:
  if (found)
    {
      for (reg = 0; reg < 4; reg++)
	if (r[reg] >= 0
	    && CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg])
	  ENCODE_DESIGNATION (r[reg], reg, coding);
    }

  return dst;
}

/* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions".  */

static void
encode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  const unsigned char *src = source;
  const unsigned char *src_end = source + src_bytes;
  unsigned char *dst = destination;
  unsigned char *dst_end = destination + dst_bytes;
  /* Since the maximum bytes produced by each loop is 20, we subtract 19
     from DST_END to assure overflow checking is necessary only at the
     head of loop.  */
  unsigned char *adjusted_dst_end = dst_end - 19;
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source text to
     analyze multi-byte codes (within macro ONE_MORE_CHAR), or when
     there's not enough destination area to produce encoded codes
     (within macro EMIT_BYTES).  */
  const unsigned char *src_base;
  int c;
  Lisp_Object translation_table;
  Lisp_Object safe_chars;

  if (coding->flags & CODING_FLAG_ISO_SAFE)
    coding->mode |= CODING_MODE_INHIBIT_UNENCODABLE_CHAR;

  safe_chars = coding_safe_chars (coding->symbol);

  if (NILP (Venable_character_translation))
    translation_table = Qnil;
  else
    {
      translation_table = coding->translation_table_for_encode;
      if (NILP (translation_table))
	translation_table = Vstandard_translation_table_for_encode;
    }

  coding->consumed_char = 0;
  coding->errors = 0;
  while (1)
    {
      src_base = src;

      if (dst >= (dst_bytes ? adjusted_dst_end : (src - 19)))
	{
	  coding->result = CODING_FINISH_INSUFFICIENT_DST;
	  break;
	}

      if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL
	  && CODING_SPEC_ISO_BOL (coding))
	{
	  /* We have to produce designation sequences if any now.  */
	  dst = encode_designation_at_bol (coding, translation_table,
					   src, src_end, dst);
	  CODING_SPEC_ISO_BOL (coding) = 0;
	}

      /* Check composition start and end.  */
      if (coding->composing != COMPOSITION_DISABLED
	  && coding->cmp_data_start < coding->cmp_data->used)
	{
	  struct composition_data *cmp_data = coding->cmp_data;
	  int *data = cmp_data->data + coding->cmp_data_start;
	  int this_pos = cmp_data->char_offset + coding->consumed_char;

	  if (coding->composing == COMPOSITION_RELATIVE)
	    {
	      if (this_pos == data[2])
		{
		  ENCODE_COMPOSITION_END (coding, data);
		  cmp_data = coding->cmp_data;
		  data = cmp_data->data + coding->cmp_data_start;
		}
	    }
	  else if (COMPOSING_P (coding))
	    {
	      /* COMPOSITION_WITH_ALTCHARS or COMPOSITION_WITH_RULE_ALTCHAR  */
	      if (coding->cmp_data_index == coding->cmp_data_start + data[0])
		/* We have consumed components of the composition.
                   What follows in SRC is the composition's base
                   text.  */
		ENCODE_COMPOSITION_FAKE_START (coding);
	      else
		{
		  int c = cmp_data->data[coding->cmp_data_index++];
		  if (coding->composition_rule_follows)
		    {
		      ENCODE_COMPOSITION_RULE (c);
		      coding->composition_rule_follows = 0;
		    }
		  else
		    {
		      if (coding->mode & CODING_MODE_INHIBIT_UNENCODABLE_CHAR
			  && ! CODING_SAFE_CHAR_P (safe_chars, c))
			ENCODE_UNSAFE_CHARACTER (c);
		      else
			ENCODE_ISO_CHARACTER (c);
		      if (coding->composing == COMPOSITION_WITH_RULE_ALTCHARS)
			coding->composition_rule_follows = 1;
		    }
		  continue;
		}
	    }
	  if (!COMPOSING_P (coding))
	    {
	      if (this_pos == data[1])
		{
		  ENCODE_COMPOSITION_START (coding, data);
		  continue;
		}
	    }
	}

      ONE_MORE_CHAR (c);

      /* Now encode the character C.  */
      if (c < 0x20 || c == 0x7F)
	{
	  if (c == '\r')
	    {
	      if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))
		{
		  if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL)
		    ENCODE_RESET_PLANE_AND_REGISTER;
		  *dst++ = c;
		  continue;
		}
	      /* fall down to treat '\r' as '\n' ...  */
	      c = '\n';
	    }
	  if (c == '\n')
	    {
	      if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL)
		ENCODE_RESET_PLANE_AND_REGISTER;
	      if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL)
		bcopy (coding->spec.iso2022.initial_designation,
		       coding->spec.iso2022.current_designation,
		       sizeof coding->spec.iso2022.initial_designation);
	      if (coding->eol_type == CODING_EOL_LF
		  || coding->eol_type == CODING_EOL_UNDECIDED)
		*dst++ = ISO_CODE_LF;
	      else if (coding->eol_type == CODING_EOL_CRLF)
		*dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF;
	      else
		*dst++ = ISO_CODE_CR;
	      CODING_SPEC_ISO_BOL (coding) = 1;
	    }
	  else
	    {
	      if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL)
		ENCODE_RESET_PLANE_AND_REGISTER;
	      *dst++ = c;
	    }
	}
      else if (ASCII_BYTE_P (c))
	ENCODE_ISO_CHARACTER (c);
      else if (SINGLE_BYTE_CHAR_P (c))
	{
	  *dst++ = c;
	  coding->errors++;
	}
      else if (coding->mode & CODING_MODE_INHIBIT_UNENCODABLE_CHAR
	       && ! CODING_SAFE_CHAR_P (safe_chars, c))
	ENCODE_UNSAFE_CHARACTER (c);
      else
	ENCODE_ISO_CHARACTER (c);

      coding->consumed_char++;
    }

 label_end_of_loop:
  coding->consumed = src_base - source;
  coding->produced = coding->produced_char = dst - destination;
}


/*** 4. SJIS and BIG5 handlers ***/

/* Although SJIS and BIG5 are not ISO coding systems, they are used
   quite widely.  So, for the moment, Emacs supports them in the bare
   C code.  But, in the future, they may be supported only by CCL.  */

/* SJIS is a coding system encoding three character sets: ASCII, right
   half of JISX0201-Kana, and JISX0208.  An ASCII character is encoded
   as is.  A character of charset katakana-jisx0201 is encoded by
   "position-code + 0x80".  A character of charset japanese-jisx0208
   is encoded in 2-byte but two position-codes are divided and shifted
   so that it fits in the range below.

   --- CODE RANGE of SJIS ---
   (character set)	(range)
   ASCII		0x00 .. 0x7F
   KATAKANA-JISX0201	0xA1 .. 0xDF
   JISX0208 (1st byte)	0x81 .. 0x9F and 0xE0 .. 0xEF
	    (2nd byte)	0x40 .. 0x7E and 0x80 .. 0xFC
   -------------------------------

*/

/* BIG5 is a coding system encoding two character sets: ASCII and
   Big5.  An ASCII character is encoded as is.  Big5 is a two-byte
   character set and is encoded in two bytes.

   --- CODE RANGE of BIG5 ---
   (character set)	(range)
   ASCII		0x00 .. 0x7F
   Big5 (1st byte)	0xA1 .. 0xFE
	(2nd byte)	0x40 .. 0x7E and 0xA1 .. 0xFE
   --------------------------

   Since the number of characters in Big5 is larger than maximum
   characters in Emacs' charset (96x96), it can't be handled as one
   charset.  So, in Emacs, Big5 is divided into two: `charset-big5-1'
   and `charset-big5-2'.  Both are DIMENSION2 and CHARS94.  The former
   contains frequently used characters and the latter contains less
   frequently used characters.  */

/* Macros to decode or encode a character of Big5 in BIG5.  B1 and B2
   are the 1st and 2nd position-codes of Big5 in BIG5 coding system.
   C1 and C2 are the 1st and 2nd position-codes of Emacs' internal
   format.  CHARSET is `charset_big5_1' or `charset_big5_2'.  */

/* Number of Big5 characters which have the same code in 1st byte.  */
#define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40)

#define DECODE_BIG5(b1, b2, charset, c1, c2)			     	\
  do {								     	\
    unsigned int temp						     	\
      = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62);	\
    if (b1 < 0xC9)						     	\
      charset = charset_big5_1;					     	\
    else							     	\
      {								     	\
	charset = charset_big5_2;				     	\
	temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW;			     	\
      }								     	\
    c1 = temp / (0xFF - 0xA1) + 0x21;				     	\
    c2 = temp % (0xFF - 0xA1) + 0x21;				     	\
  } while (0)

#define ENCODE_BIG5(charset, c1, c2, b1, b2)			  	\
  do {								  	\
    unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21);	\
    if (charset == charset_big5_2)				  	\
      temp += BIG5_SAME_ROW * (0xC9 - 0xA1);			  	\
    b1 = temp / BIG5_SAME_ROW + 0xA1;				  	\
    b2 = temp % BIG5_SAME_ROW;					  	\
    b2 += b2 < 0x3F ? 0x40 : 0x62;				  	\
  } while (0)

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in SJIS.  If it is, return
   CODING_CATEGORY_MASK_SJIS, else return 0.  */

static int
detect_coding_sjis (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  int c;
  /* Dummy for ONE_MORE_BYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;

  while (1)
    {
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, CODING_CATEGORY_MASK_SJIS);
      if (c < 0x80)
	continue;
      if (c == 0x80 || c == 0xA0 || c > 0xEF)
	return 0;
      if (c <= 0x9F || c >= 0xE0)
	{
	  ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, 0);
	  if (c < 0x40 || c == 0x7F || c > 0xFC)
	    return 0;
	}
    }
}

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in BIG5.  If it is, return
   CODING_CATEGORY_MASK_BIG5, else return 0.  */

static int
detect_coding_big5 (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  int c;
  /* Dummy for ONE_MORE_BYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;

  while (1)
    {
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, CODING_CATEGORY_MASK_BIG5);
      if (c < 0x80)
	continue;
      if (c < 0xA1 || c > 0xFE)
	return 0;
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, 0);
      if (c < 0x40 || (c > 0x7F && c < 0xA1) || c > 0xFE)
	return 0;
    }
}

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in UTF-8.  If it is, return
   CODING_CATEGORY_MASK_UTF_8, else return 0.  */

#define UTF_8_1_OCTET_P(c)         ((c) < 0x80)
#define UTF_8_EXTRA_OCTET_P(c)     (((c) & 0xC0) == 0x80)
#define UTF_8_2_OCTET_LEADING_P(c) (((c) & 0xE0) == 0xC0)
#define UTF_8_3_OCTET_LEADING_P(c) (((c) & 0xF0) == 0xE0)
#define UTF_8_4_OCTET_LEADING_P(c) (((c) & 0xF8) == 0xF0)
#define UTF_8_5_OCTET_LEADING_P(c) (((c) & 0xFC) == 0xF8)
#define UTF_8_6_OCTET_LEADING_P(c) (((c) & 0xFE) == 0xFC)

static int
detect_coding_utf_8 (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  unsigned char c;
  int seq_maybe_bytes;
  /* Dummy for ONE_MORE_BYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;

  while (1)
    {
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, CODING_CATEGORY_MASK_UTF_8);
      if (UTF_8_1_OCTET_P (c))
	continue;
      else if (UTF_8_2_OCTET_LEADING_P (c))
	seq_maybe_bytes = 1;
      else if (UTF_8_3_OCTET_LEADING_P (c))
	seq_maybe_bytes = 2;
      else if (UTF_8_4_OCTET_LEADING_P (c))
	seq_maybe_bytes = 3;
      else if (UTF_8_5_OCTET_LEADING_P (c))
	seq_maybe_bytes = 4;
      else if (UTF_8_6_OCTET_LEADING_P (c))
	seq_maybe_bytes = 5;
      else
	return 0;

      do
	{
	  ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, 0);
	  if (!UTF_8_EXTRA_OCTET_P (c))
	    return 0;
	  seq_maybe_bytes--;
	}
      while (seq_maybe_bytes > 0);
    }
}

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in UTF-16 Big Endian (endian == 1) or
   Little Endian (otherwise).  If it is, return
   CODING_CATEGORY_MASK_UTF_16_BE or CODING_CATEGORY_MASK_UTF_16_LE,
   else return 0.  */

#define UTF_16_INVALID_P(val)	\
  (((val) == 0xFFFE)		\
   || ((val) == 0xFFFF))

#define UTF_16_HIGH_SURROGATE_P(val) \
  (((val) & 0xD800) == 0xD800)

#define UTF_16_LOW_SURROGATE_P(val) \
  (((val) & 0xDC00) == 0xDC00)

static int
detect_coding_utf_16 (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  unsigned char c1, c2;
  /* Dummy for ONE_MORE_BYTE_CHECK_MULTIBYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;

  ONE_MORE_BYTE_CHECK_MULTIBYTE (c1, multibytep, 0);
  ONE_MORE_BYTE_CHECK_MULTIBYTE (c2, multibytep, 0);

  if ((c1 == 0xFF) && (c2 == 0xFE))
    return CODING_CATEGORY_MASK_UTF_16_LE;
  else if ((c1 == 0xFE) && (c2 == 0xFF))
    return CODING_CATEGORY_MASK_UTF_16_BE;
  return 0;
}

/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".
   If SJIS_P is 1, decode SJIS text, else decode BIG5 test.  */

static void
decode_coding_sjis_big5 (coding, source, destination,
			 src_bytes, dst_bytes, sjis_p)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char  *destination;
     int src_bytes, dst_bytes;
     int sjis_p;
{
  const unsigned char *src = source;
  const unsigned char *src_end = source + src_bytes;
  unsigned char *dst = destination;
  unsigned char *dst_end = destination + dst_bytes;
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source code
     (within macro ONE_MORE_BYTE), or when there's not enough
     destination area to produce a character (within macro
     EMIT_CHAR).  */
  const unsigned char *src_base;
  Lisp_Object translation_table;

  if (NILP (Venable_character_translation))
    translation_table = Qnil;
  else
    {
      translation_table = coding->translation_table_for_decode;
      if (NILP (translation_table))
	translation_table = Vstandard_translation_table_for_decode;
    }

  coding->produced_char = 0;
  while (1)
    {
      int c, charset, c1, c2 = 0;

      src_base = src;
      ONE_MORE_BYTE (c1);

      if (c1 < 0x80)
	{
	  charset = CHARSET_ASCII;
	  if (c1 < 0x20)
	    {
	      if (c1 == '\r')
		{
		  if (coding->eol_type == CODING_EOL_CRLF)
		    {
		      ONE_MORE_BYTE (c2);
		      if (c2 == '\n')
			c1 = c2;
		      else
			/* To process C2 again, SRC is subtracted by 1.  */
			src--;
		    }
		  else if (coding->eol_type == CODING_EOL_CR)
		    c1 = '\n';
		}
	      else if (c1 == '\n'
		       && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
		       && (coding->eol_type == CODING_EOL_CR
			   || coding->eol_type == CODING_EOL_CRLF))
		{
		  coding->result = CODING_FINISH_INCONSISTENT_EOL;
		  goto label_end_of_loop;
		}
	    }
	}
      else
        {
	  if (sjis_p)
	    {
	      if (c1 == 0x80 || c1 == 0xA0 || c1 > 0xEF)
		goto label_invalid_code;
	      if (c1 <= 0x9F || c1 >= 0xE0)
		{
		  /* SJIS -> JISX0208 */
		  ONE_MORE_BYTE (c2);
		  if (c2 < 0x40 || c2 == 0x7F || c2 > 0xFC)
		    goto label_invalid_code;
		  DECODE_SJIS (c1, c2, c1, c2);
		  charset = charset_jisx0208;
		}
	      else
		/* SJIS -> JISX0201-Kana */
		charset = charset_katakana_jisx0201;
	    }
	  else
	    {
	      /* BIG5 -> Big5 */
	      if (c1 < 0xA0 || c1 > 0xFE)
		goto label_invalid_code;
	      ONE_MORE_BYTE (c2);
	      if (c2 < 0x40 || (c2 > 0x7E && c2 < 0xA1) || c2 > 0xFE)
		goto label_invalid_code;
	      DECODE_BIG5 (c1, c2, charset, c1, c2);
	    }
	}

      c = DECODE_ISO_CHARACTER (charset, c1, c2);
      EMIT_CHAR (c);
      continue;

    label_invalid_code:
      coding->errors++;
      src = src_base;
      c = *src++;
      EMIT_CHAR (c);
    }

 label_end_of_loop:
  coding->consumed = coding->consumed_char = src_base - source;
  coding->produced = dst - destination;
  return;
}

/* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions".
   This function can encode charsets `ascii', `katakana-jisx0201',
   `japanese-jisx0208', `chinese-big5-1', and `chinese-big5-2'.  We
   are sure that all these charsets are registered as official charset
   (i.e. do not have extended leading-codes).  Characters of other
   charsets are produced without any encoding.  If SJIS_P is 1, encode
   SJIS text, else encode BIG5 text.  */

static void
encode_coding_sjis_big5 (coding, source, destination,
			 src_bytes, dst_bytes, sjis_p)
     struct coding_system *coding;
     unsigned char *source, *destination;
     int src_bytes, dst_bytes;
     int sjis_p;
{
  unsigned char *src = source;
  unsigned char *src_end = source + src_bytes;
  unsigned char *dst = destination;
  unsigned char *dst_end = destination + dst_bytes;
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source text to
     analyze multi-byte codes (within macro ONE_MORE_CHAR), or when
     there's not enough destination area to produce encoded codes
     (within macro EMIT_BYTES).  */
  unsigned char *src_base;
  Lisp_Object translation_table;

  if (NILP (Venable_character_translation))
    translation_table = Qnil;
  else
    {
      translation_table = coding->translation_table_for_encode;
      if (NILP (translation_table))
	translation_table = Vstandard_translation_table_for_encode;
    }

  while (1)
    {
      int c, charset, c1, c2;

      src_base = src;
      ONE_MORE_CHAR (c);

      /* Now encode the character C.  */
      if (SINGLE_BYTE_CHAR_P (c))
	{
	  switch (c)
	    {
	    case '\r':
	      if (!(coding->mode & CODING_MODE_SELECTIVE_DISPLAY))
		{
		  EMIT_ONE_BYTE (c);
		  break;
		}
	      c = '\n';
	    case '\n':
	      if (coding->eol_type == CODING_EOL_CRLF)
		{
		  EMIT_TWO_BYTES ('\r', c);
		  break;
		}
	      else if (coding->eol_type == CODING_EOL_CR)
		c = '\r';
	    default:
	      EMIT_ONE_BYTE (c);
	    }
	}
      else
	{
	  SPLIT_CHAR (c, charset, c1, c2);
	  if (sjis_p)
	    {
	      if (charset == charset_jisx0208
		  || charset == charset_jisx0208_1978)
		{
		  ENCODE_SJIS (c1, c2, c1, c2);
		  EMIT_TWO_BYTES (c1, c2);
		}
	      else if (charset == charset_katakana_jisx0201)
		EMIT_ONE_BYTE (c1 | 0x80);
	      else if (charset == charset_latin_jisx0201)
		EMIT_ONE_BYTE (c1);
	      else if (coding->mode & CODING_MODE_INHIBIT_UNENCODABLE_CHAR)
		{
		  EMIT_ONE_BYTE (CODING_REPLACEMENT_CHARACTER);
		  if (CHARSET_WIDTH (charset) > 1)
		    EMIT_ONE_BYTE (CODING_REPLACEMENT_CHARACTER);
		}
	      else
		/* There's no way other than producing the internal
		   codes as is.  */
		EMIT_BYTES (src_base, src);
	    }
	  else
	    {
	      if (charset == charset_big5_1 || charset == charset_big5_2)
		{
		  ENCODE_BIG5 (charset, c1, c2, c1, c2);
		  EMIT_TWO_BYTES (c1, c2);
		}
	      else if (coding->mode & CODING_MODE_INHIBIT_UNENCODABLE_CHAR)
		{
		  EMIT_ONE_BYTE (CODING_REPLACEMENT_CHARACTER);
		  if (CHARSET_WIDTH (charset) > 1)
		    EMIT_ONE_BYTE (CODING_REPLACEMENT_CHARACTER);
		}
	      else
		/* There's no way other than producing the internal
		   codes as is.  */
		EMIT_BYTES (src_base, src);
	    }
	}
      coding->consumed_char++;
    }

 label_end_of_loop:
  coding->consumed = src_base - source;
  coding->produced = coding->produced_char = dst - destination;
}


/*** 5. CCL handlers ***/

/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
   Check if a text is encoded in a coding system of which
   encoder/decoder are written in CCL program.  If it is, return
   CODING_CATEGORY_MASK_CCL, else return 0.  */

static int
detect_coding_ccl (src, src_end, multibytep)
     unsigned char *src, *src_end;
     int multibytep;
{
  unsigned char *valid;
  int c;
  /* Dummy for ONE_MORE_BYTE.  */
  struct coding_system dummy_coding;
  struct coding_system *coding = &dummy_coding;

  /* No coding system is assigned to coding-category-ccl.  */
  if (!coding_system_table[CODING_CATEGORY_IDX_CCL])
    return 0;

  valid = coding_system_table[CODING_CATEGORY_IDX_CCL]->spec.ccl.valid_codes;
  while (1)
    {
      ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep, CODING_CATEGORY_MASK_CCL);
      if (! valid[c])
	return 0;
    }
}


/*** 6. End-of-line handlers ***/

/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".  */

static void
decode_eol (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  const unsigned char *src = source;
  unsigned char *dst = destination;
  const unsigned char *src_end = src + src_bytes;
  unsigned char *dst_end = dst + dst_bytes;
  Lisp_Object translation_table;
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source code
     (within macro ONE_MORE_BYTE), or when there's not enough
     destination area to produce a character (within macro
     EMIT_CHAR).  */
  const unsigned char *src_base;
  int c;

  translation_table = Qnil;
  switch (coding->eol_type)
    {
    case CODING_EOL_CRLF:
      while (1)
	{
	  src_base = src;
	  ONE_MORE_BYTE (c);
	  if (c == '\r')
	    {
	      ONE_MORE_BYTE (c);
	      if (c != '\n')
		{
		  src--;
		  c = '\r';
		}
	    }
	  else if (c == '\n'
		   && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL))
	    {
	      coding->result = CODING_FINISH_INCONSISTENT_EOL;
	      goto label_end_of_loop;
	    }
	  EMIT_CHAR (c);
	}
      break;

    case CODING_EOL_CR:
      while (1)
	{
	  src_base = src;
	  ONE_MORE_BYTE (c);
	  if (c == '\n')
	    {
	      if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
		{
		  coding->result = CODING_FINISH_INCONSISTENT_EOL;
		  goto label_end_of_loop;
		}
	    }
	  else if (c == '\r')
	    c = '\n';
	  EMIT_CHAR (c);
	}
      break;

    default:			/* no need for EOL handling */
      while (1)
	{
	  src_base = src;
	  ONE_MORE_BYTE (c);
	  EMIT_CHAR (c);
	}
    }

 label_end_of_loop:
  coding->consumed = coding->consumed_char = src_base - source;
  coding->produced = dst - destination;
  return;
}

/* See "GENERAL NOTES about `encode_coding_XXX ()' functions".  Encode
   format of end-of-line according to `coding->eol_type'.  It also
   convert multibyte form 8-bit characters to unibyte if
   CODING->src_multibyte is nonzero.  If `coding->mode &
   CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code '\r' in source text
   also means end-of-line.  */

static void
encode_eol (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  const unsigned char *src = source;
  unsigned char *dst = destination;
  const unsigned char *src_end = src + src_bytes;
  unsigned char *dst_end = dst + dst_bytes;
  Lisp_Object translation_table;
  /* SRC_BASE remembers the start position in source in each loop.
     The loop will be exited when there's not enough source text to
     analyze multi-byte codes (within macro ONE_MORE_CHAR), or when
     there's not enough destination area to produce encoded codes
     (within macro EMIT_BYTES).  */
  const unsigned char *src_base;
  unsigned char *tmp;
  int c;
  int selective_display = coding->mode & CODING_MODE_SELECTIVE_DISPLAY;

  translation_table = Qnil;
  if (coding->src_multibyte
      && *(src_end - 1) == LEADING_CODE_8_BIT_CONTROL)
    {
      src_end--;
      src_bytes--;
      coding->result = CODING_FINISH_INSUFFICIENT_SRC;
    }

  if (coding->eol_type == CODING_EOL_CRLF)
    {
      while (src < src_end)
	{
	  src_base = src;
	  c = *src++;
	  if (c >= 0x20)
	    EMIT_ONE_BYTE (c);
	  else if (c == '\n' || (c == '\r' && selective_display))
	    EMIT_TWO_BYTES ('\r', '\n');
	  else
	    EMIT_ONE_BYTE (c);
	}
      src_base = src;
    label_end_of_loop:
      ;
    }
  else
    {
      if (!dst_bytes || src_bytes <= dst_bytes)
	{
	  safe_bcopy (src, dst, src_bytes);
	  src_base = src_end;
	  dst += src_bytes;
	}
      else
	{
	  if (coding->src_multibyte
	      && *(src + dst_bytes - 1) == LEADING_CODE_8_BIT_CONTROL)
	    dst_bytes--;
	  safe_bcopy (src, dst, dst_bytes);
	  src_base = src + dst_bytes;
	  dst = destination + dst_bytes;
	  coding->result = CODING_FINISH_INSUFFICIENT_DST;
	}
      if (coding->eol_type == CODING_EOL_CR)
	{
	  for (tmp = destination; tmp < dst; tmp++)
	    if (*tmp == '\n') *tmp = '\r';
	}
      else if (selective_display)
	{
	  for (tmp = destination; tmp < dst; tmp++)
	    if (*tmp == '\r') *tmp = '\n';
	}
    }
  if (coding->src_multibyte)
    dst = destination + str_as_unibyte (destination, dst - destination);

  coding->consumed = src_base - source;
  coding->produced = dst - destination;
  coding->produced_char = coding->produced;
}


/*** 7. C library functions ***/

/* In Emacs Lisp, a coding system is represented by a Lisp symbol which
   has a property `coding-system'.  The value of this property is a
   vector of length 5 (called the coding-vector).  Among elements of
   this vector, the first (element[0]) and the fifth (element[4])
   carry important information for decoding/encoding.  Before
   decoding/encoding, this information should be set in fields of a
   structure of type `coding_system'.

   The value of the property `coding-system' can be a symbol of another
   subsidiary coding-system.  In that case, Emacs gets coding-vector
   from that symbol.

   `element[0]' contains information to be set in `coding->type'.  The
   value and its meaning is as follows:

   0 -- coding_type_emacs_mule
   1 -- coding_type_sjis
   2 -- coding_type_iso2022
   3 -- coding_type_big5
   4 -- coding_type_ccl encoder/decoder written in CCL
   nil -- coding_type_no_conversion
   t -- coding_type_undecided (automatic conversion on decoding,
   			       no-conversion on encoding)

   `element[4]' contains information to be set in `coding->flags' and
   `coding->spec'.  The meaning varies by `coding->type'.

   If `coding->type' is `coding_type_iso2022', element[4] is a vector
   of length 32 (of which the first 13 sub-elements are used now).
   Meanings of these sub-elements are:

   sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022'
   	If the value is an integer of valid charset, the charset is
	assumed to be designated to graphic register N initially.

	If the value is minus, it is a minus value of charset which
	reserves graphic register N, which means that the charset is
	not designated initially but should be designated to graphic
	register N just before encoding a character in that charset.

	If the value is nil, graphic register N is never used on
	encoding.

   sub-element[N] where N is 4 through 11: to be set in `coding->flags'
   	Each value takes t or nil.  See the section ISO2022 of
	`coding.h' for more information.

   If `coding->type' is `coding_type_big5', element[4] is t to denote
   BIG5-ETen or nil to denote BIG5-HKU.

   If `coding->type' takes the other value, element[4] is ignored.

   Emacs Lisp's coding systems also carry information about format of
   end-of-line in a value of property `eol-type'.  If the value is
   integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2
   means CODING_EOL_CR.  If it is not integer, it should be a vector
   of subsidiary coding systems of which property `eol-type' has one
   of the above values.

*/

/* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL
   and set it in CODING.  If CODING_SYSTEM_SYMBOL is invalid, CODING
   is setup so that no conversion is necessary and return -1, else
   return 0.  */

int
setup_coding_system (coding_system, coding)
     Lisp_Object coding_system;
     struct coding_system *coding;
{
  Lisp_Object coding_spec, coding_type, eol_type, plist;
  Lisp_Object val;

  /* At first, zero clear all members.  */
  bzero (coding, sizeof (struct coding_system));

  /* Initialize some fields required for all kinds of coding systems.  */
  coding->symbol = coding_system;
  coding->heading_ascii = -1;
  coding->post_read_conversion = coding->pre_write_conversion = Qnil;
  coding->composing = COMPOSITION_DISABLED;
  coding->cmp_data = NULL;

  if (NILP (coding_system))
    goto label_invalid_coding_system;

  coding_spec = Fget (coding_system, Qcoding_system);

  if (!VECTORP (coding_spec)
      || XVECTOR (coding_spec)->size != 5
      || !CONSP (XVECTOR (coding_spec)->contents[3]))
    goto label_invalid_coding_system;

  eol_type = inhibit_eol_conversion ? Qnil : Fget (coding_system, Qeol_type);
  if (VECTORP (eol_type))
    {
      coding->eol_type = CODING_EOL_UNDECIDED;
      coding->common_flags = CODING_REQUIRE_DETECTION_MASK;
      if (system_eol_type != CODING_EOL_LF)
	coding->common_flags |= CODING_REQUIRE_ENCODING_MASK;
    }
  else if (XFASTINT (eol_type) == 1)
    {
      coding->eol_type = CODING_EOL_CRLF;
      coding->common_flags
	= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
    }
  else if (XFASTINT (eol_type) == 2)
    {
      coding->eol_type = CODING_EOL_CR;
      coding->common_flags
	= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
    }
  else
    {
      coding->common_flags = 0;
      coding->eol_type = CODING_EOL_LF;
    }

  coding_type = XVECTOR (coding_spec)->contents[0];
  /* Try short cut.  */
  if (SYMBOLP (coding_type))
    {
      if (EQ (coding_type, Qt))
	{
	  coding->type = coding_type_undecided;
	  coding->common_flags |= CODING_REQUIRE_DETECTION_MASK;
	}
      else
	coding->type = coding_type_no_conversion;
      /* Initialize this member.  Any thing other than
	 CODING_CATEGORY_IDX_UTF_16_BE and
	 CODING_CATEGORY_IDX_UTF_16_LE are ok because they have
	 special treatment in detect_eol.  */
      coding->category_idx = CODING_CATEGORY_IDX_EMACS_MULE;

      return 0;
    }

  /* Get values of coding system properties:
     `post-read-conversion', `pre-write-conversion',
     `translation-table-for-decode', `translation-table-for-encode'.  */
  plist = XVECTOR (coding_spec)->contents[3];
  /* Pre & post conversion functions should be disabled if
     inhibit_eol_conversion is nonzero.  This is the case that a code
     conversion function is called while those functions are running.  */
  if (! inhibit_pre_post_conversion)
    {
      coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion);
      coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion);
    }
  val = Fplist_get (plist, Qtranslation_table_for_decode);
  if (SYMBOLP (val))
    val = Fget (val, Qtranslation_table_for_decode);
  coding->translation_table_for_decode = CHAR_TABLE_P (val) ? val : Qnil;
  val = Fplist_get (plist, Qtranslation_table_for_encode);
  if (SYMBOLP (val))
    val = Fget (val, Qtranslation_table_for_encode);
  coding->translation_table_for_encode = CHAR_TABLE_P (val) ? val : Qnil;
  val = Fplist_get (plist, Qcoding_category);
  if (!NILP (val))
    {
      val = Fget (val, Qcoding_category_index);
      if (INTEGERP (val))
	coding->category_idx = XINT (val);
      else
	goto label_invalid_coding_system;
    }
  else
    goto label_invalid_coding_system;

  /* If the coding system has non-nil `composition' property, enable
     composition handling.  */
  val = Fplist_get (plist, Qcomposition);
  if (!NILP (val))
    coding->composing = COMPOSITION_NO;

  /* If the coding system is ascii-incompatible, record it in
     common_flags.   */
  val = Fplist_get (plist, Qascii_incompatible);
  if (! NILP (val))
    coding->common_flags |= CODING_ASCII_INCOMPATIBLE_MASK;

  switch (XFASTINT (coding_type))
    {
    case 0:
      coding->type = coding_type_emacs_mule;
      coding->common_flags
	|= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
      if (!NILP (coding->post_read_conversion))
	coding->common_flags |= CODING_REQUIRE_DECODING_MASK;
      if (!NILP (coding->pre_write_conversion))
	coding->common_flags |= CODING_REQUIRE_ENCODING_MASK;
      break;

    case 1:
      coding->type = coding_type_sjis;
      coding->common_flags
	|= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
      break;

    case 2:
      coding->type = coding_type_iso2022;
      coding->common_flags
	|= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
      {
	Lisp_Object val, temp;
	Lisp_Object *flags;
	int i, charset, reg_bits = 0;

	val = XVECTOR (coding_spec)->contents[4];

	if (!VECTORP (val) || XVECTOR (val)->size != 32)
	  goto label_invalid_coding_system;

	flags = XVECTOR (val)->contents;
	coding->flags
	  = ((NILP (flags[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM)
	     | (NILP (flags[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL)
	     | (NILP (flags[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL)
	     | (NILP (flags[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS)
	     | (NILP (flags[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT)
	     | (NILP (flags[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT)
	     | (NILP (flags[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN)
	     | (NILP (flags[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS)
	     | (NILP (flags[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION)
	     | (NILP (flags[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL)
	     | (NILP (flags[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL)
	     | (NILP (flags[15]) ? 0 : CODING_FLAG_ISO_SAFE)
	     | (NILP (flags[16]) ? 0 : CODING_FLAG_ISO_LATIN_EXTRA)
	     );

	/* Invoke graphic register 0 to plane 0.  */
	CODING_SPEC_ISO_INVOCATION (coding, 0) = 0;
	/* Invoke graphic register 1 to plane 1 if we can use full 8-bit.  */
	CODING_SPEC_ISO_INVOCATION (coding, 1)
	  = (coding->flags & CODING_FLAG_ISO_SEVEN_BITS ? -1 : 1);
	/* Not single shifting at first.  */
	CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0;
	/* Beginning of buffer should also be regarded as bol. */
	CODING_SPEC_ISO_BOL (coding) = 1;

	for (charset = 0; charset <= MAX_CHARSET; charset++)
	  CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = 255;
	val = Vcharset_revision_alist;
	while (CONSP (val))
	  {
	    charset = get_charset_id (Fcar_safe (XCAR (val)));
	    if (charset >= 0
		&& (temp = Fcdr_safe (XCAR (val)), INTEGERP (temp))
		&& (i = XINT (temp), (i >= 0 && (i + '@') < 128)))
	      CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = i;
	    val = XCDR (val);
	  }

	/* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations.
	   FLAGS[REG] can be one of below:
		integer CHARSET: CHARSET occupies register I,
		t: designate nothing to REG initially, but can be used
		  by any charsets,
		list of integer, nil, or t: designate the first
		  element (if integer) to REG initially, the remaining
		  elements (if integer) is designated to REG on request,
		  if an element is t, REG can be used by any charsets,
		nil: REG is never used.  */
	for (charset = 0; charset <= MAX_CHARSET; charset++)
	  CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
	    = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION;
	for (i = 0; i < 4; i++)
	  {
	    if ((INTEGERP (flags[i])
		 && (charset = XINT (flags[i]), CHARSET_VALID_P (charset)))
		|| (charset = get_charset_id (flags[i])) >= 0)
	      {
		CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset;
		CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) = i;
	      }
	    else if (EQ (flags[i], Qt))
	      {
		CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1;
		reg_bits |= 1 << i;
		coding->flags |= CODING_FLAG_ISO_DESIGNATION;
	      }
	    else if (CONSP (flags[i]))
	      {
		Lisp_Object tail;
		tail = flags[i];

		coding->flags |= CODING_FLAG_ISO_DESIGNATION;
		if ((INTEGERP (XCAR (tail))
		     && (charset = XINT (XCAR (tail)),
			 CHARSET_VALID_P (charset)))
		    || (charset = get_charset_id (XCAR (tail))) >= 0)
		  {
		    CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset;
		    CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i;
		  }
		else
		  CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1;
		tail = XCDR (tail);
		while (CONSP (tail))
		  {
		    if ((INTEGERP (XCAR (tail))
			 && (charset = XINT (XCAR (tail)),
			     CHARSET_VALID_P (charset)))
			|| (charset = get_charset_id (XCAR (tail))) >= 0)
		      CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
			= i;
		    else if (EQ (XCAR (tail), Qt))
		      reg_bits |= 1 << i;
		    tail = XCDR (tail);
		  }
	      }
	    else
	      CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1;

	    CODING_SPEC_ISO_DESIGNATION (coding, i)
	      = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i);
	  }

	if (reg_bits && ! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT))
	  {
	    /* REG 1 can be used only by locking shift in 7-bit env.  */
	    if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS)
	      reg_bits &= ~2;
	    if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT))
	      /* Without any shifting, only REG 0 and 1 can be used.  */
	      reg_bits &= 3;
	  }

	if (reg_bits)
	  for (charset = 0; charset <= MAX_CHARSET; charset++)
	    {
	      if (CHARSET_DEFINED_P (charset)
		  && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
		      == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))
		{
		  /* There exist some default graphic registers to be
		     used by CHARSET.  */

		  /* We had better avoid designating a charset of
		     CHARS96 to REG 0 as far as possible.  */
		  if (CHARSET_CHARS (charset) == 96)
		    CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
		      = (reg_bits & 2
			 ? 1 : (reg_bits & 4 ? 2 : (reg_bits & 8 ? 3 : 0)));
		  else
		    CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
		      = (reg_bits & 1
			 ? 0 : (reg_bits & 2 ? 1 : (reg_bits & 4 ? 2 : 3)));
		}
	    }
      }
      coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK;
      coding->spec.iso2022.last_invalid_designation_register = -1;
      break;

    case 3:
      coding->type = coding_type_big5;
      coding->common_flags
	|= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
      coding->flags
	= (NILP (XVECTOR (coding_spec)->contents[4])
	   ? CODING_FLAG_BIG5_HKU
	   : CODING_FLAG_BIG5_ETEN);
      break;

    case 4:
      coding->type = coding_type_ccl;
      coding->common_flags
	|= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
      {
	val = XVECTOR (coding_spec)->contents[4];
	if (! CONSP (val)
	    || setup_ccl_program (&(coding->spec.ccl.decoder),
				  XCAR (val)) < 0
	    || setup_ccl_program (&(coding->spec.ccl.encoder),
				  XCDR (val)) < 0)
	  goto label_invalid_coding_system;

	bzero (coding->spec.ccl.valid_codes, 256);
	val = Fplist_get (plist, Qvalid_codes);
	if (CONSP (val))
	  {
	    Lisp_Object this;

	    for (; CONSP (val); val = XCDR (val))
	      {
		this = XCAR (val);
		if (INTEGERP (this)
		    && XINT (this) >= 0 && XINT (this) < 256)
		  coding->spec.ccl.valid_codes[XINT (this)] = 1;
		else if (CONSP (this)
			 && INTEGERP (XCAR (this))
			 && INTEGERP (XCDR (this)))
		  {
		    int start = XINT (XCAR (this));
		    int end = XINT (XCDR (this));

		    if (start >= 0 && start <= end && end < 256)
		      while (start <= end)
			coding->spec.ccl.valid_codes[start++] = 1;
		  }
	      }
	  }
      }
      coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK;
      coding->spec.ccl.cr_carryover = 0;
      coding->spec.ccl.eight_bit_carryover[0] = 0;
      break;

    case 5:
      coding->type = coding_type_raw_text;
      break;

    default:
      goto label_invalid_coding_system;
    }
  return 0;

 label_invalid_coding_system:
  coding->type = coding_type_no_conversion;
  coding->category_idx = CODING_CATEGORY_IDX_BINARY;
  coding->common_flags = 0;
  coding->eol_type = CODING_EOL_UNDECIDED;
  coding->pre_write_conversion = coding->post_read_conversion = Qnil;
  return NILP (coding_system) ? 0 : -1;
}

/* Free memory blocks allocated for storing composition information.  */

void
coding_free_composition_data (coding)
     struct coding_system *coding;
{
  struct composition_data *cmp_data = coding->cmp_data, *next;

  if (!cmp_data)
    return;
  /* Memory blocks are chained.  At first, rewind to the first, then,
     free blocks one by one.  */
  while (cmp_data->prev)
    cmp_data = cmp_data->prev;
  while (cmp_data)
    {
      next = cmp_data->next;
      xfree (cmp_data);
      cmp_data = next;
    }
  coding->cmp_data = NULL;
}

/* Set `char_offset' member of all memory blocks pointed by
   coding->cmp_data to POS.  */

void
coding_adjust_composition_offset (coding, pos)
     struct coding_system *coding;
     int pos;
{
  struct composition_data *cmp_data;

  for (cmp_data = coding->cmp_data; cmp_data; cmp_data = cmp_data->next)
    cmp_data->char_offset = pos;
}

/* Setup raw-text or one of its subsidiaries in the structure
   coding_system CODING according to the already setup value eol_type
   in CODING.  CODING should be setup for some coding system in
   advance.  */

void
setup_raw_text_coding_system (coding)
     struct coding_system *coding;
{
  if (coding->type != coding_type_raw_text)
    {
      coding->symbol = Qraw_text;
      coding->type = coding_type_raw_text;
      if (coding->eol_type != CODING_EOL_UNDECIDED)
	{
	  Lisp_Object subsidiaries;
	  subsidiaries = Fget (Qraw_text, Qeol_type);

	  if (VECTORP (subsidiaries)
	      && XVECTOR (subsidiaries)->size == 3)
	    coding->symbol
	      = XVECTOR (subsidiaries)->contents[coding->eol_type];
	}
      setup_coding_system (coding->symbol, coding);
    }
  return;
}

/* Emacs has a mechanism to automatically detect a coding system if it
   is one of Emacs' internal format, ISO2022, SJIS, and BIG5.  But,
   it's impossible to distinguish some coding systems accurately
   because they use the same range of codes.  So, at first, coding
   systems are categorized into 7, those are:

   o coding-category-emacs-mule

   	The category for a coding system which has the same code range
	as Emacs' internal format.  Assigned the coding-system (Lisp
	symbol) `emacs-mule' by default.

   o coding-category-sjis

	The category for a coding system which has the same code range
	as SJIS.  Assigned the coding-system (Lisp
	symbol) `japanese-shift-jis' by default.

   o coding-category-iso-7

   	The category for a coding system which has the same code range
	as ISO2022 of 7-bit environment.  This doesn't use any locking
	shift and single shift functions.  This can encode/decode all
	charsets.  Assigned the coding-system (Lisp symbol)
	`iso-2022-7bit' by default.

   o coding-category-iso-7-tight

	Same as coding-category-iso-7 except that this can
	encode/decode only the specified charsets.

   o coding-category-iso-8-1

   	The category for a coding system which has the same code range
	as ISO2022 of 8-bit environment and graphic plane 1 used only
	for DIMENSION1 charset.  This doesn't use any locking shift
	and single shift functions.  Assigned the coding-system (Lisp
	symbol) `iso-latin-1' by default.

   o coding-category-iso-8-2

   	The category for a coding system which has the same code range
	as ISO2022 of 8-bit environment and graphic plane 1 used only
	for DIMENSION2 charset.  This doesn't use any locking shift
	and single shift functions.  Assigned the coding-system (Lisp
	symbol) `japanese-iso-8bit' by default.

   o coding-category-iso-7-else

   	The category for a coding system which has the same code range
	as ISO2022 of 7-bit environment but uses locking shift or
	single shift functions.  Assigned the coding-system (Lisp
	symbol) `iso-2022-7bit-lock' by default.

   o coding-category-iso-8-else

   	The category for a coding system which has the same code range
	as ISO2022 of 8-bit environment but uses locking shift or
	single shift functions.  Assigned the coding-system (Lisp
	symbol) `iso-2022-8bit-ss2' by default.

   o coding-category-big5

   	The category for a coding system which has the same code range
	as BIG5.  Assigned the coding-system (Lisp symbol)
	`cn-big5' by default.

   o coding-category-utf-8

	The category for a coding system which has the same code range
	as UTF-8 (cf. RFC3629).  Assigned the coding-system (Lisp
	symbol) `utf-8' by default.

   o coding-category-utf-16-be

	The category for a coding system in which a text has an
	Unicode signature (cf. Unicode Standard) in the order of BIG
	endian at the head.  Assigned the coding-system (Lisp symbol)
	`utf-16-be' by default.

   o coding-category-utf-16-le

	The category for a coding system in which a text has an
	Unicode signature (cf. Unicode Standard) in the order of
	LITTLE endian at the head.  Assigned the coding-system (Lisp
	symbol) `utf-16-le' by default.

   o coding-category-ccl

	The category for a coding system of which encoder/decoder is
	written in CCL programs.  The default value is nil, i.e., no
	coding system is assigned.

   o coding-category-binary

   	The category for a coding system not categorized in any of the
	above.  Assigned the coding-system (Lisp symbol)
	`no-conversion' by default.

   Each of them is a Lisp symbol and the value is an actual
   `coding-system' (this is also a Lisp symbol) assigned by a user.
   What Emacs does actually is to detect a category of coding system.
   Then, it uses a `coding-system' assigned to it.  If Emacs can't
   decide a single possible category, it selects a category of the
   highest priority.  Priorities of categories are also specified by a
   user in a Lisp variable `coding-category-list'.

*/

static
int ascii_skip_code[256];

/* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded.
   If it detects possible coding systems, return an integer in which
   appropriate flag bits are set.  Flag bits are defined by macros
   CODING_CATEGORY_MASK_XXX in `coding.h'.  If PRIORITIES is non-NULL,
   it should point the table `coding_priorities'.  In that case, only
   the flag bit for a coding system of the highest priority is set in
   the returned value.  If MULTIBYTEP is nonzero, 8-bit codes of the
   range 0x80..0x9F are in multibyte form.

   How many ASCII characters are at the head is returned as *SKIP.  */

static int
detect_coding_mask (source, src_bytes, priorities, skip, multibytep)
     unsigned char *source;
     int src_bytes, *priorities, *skip;
     int multibytep;
{
  register unsigned char c;
  unsigned char *src = source, *src_end = source + src_bytes;
  unsigned int mask, utf16_examined_p, iso2022_examined_p;
  int i;

  /* At first, skip all ASCII characters and control characters except
     for three ISO2022 specific control characters.  */
  ascii_skip_code[ISO_CODE_SO] = 0;
  ascii_skip_code[ISO_CODE_SI] = 0;
  ascii_skip_code[ISO_CODE_ESC] = 0;

 label_loop_detect_coding:
  while (src < src_end && ascii_skip_code[*src]) src++;
  *skip = src - source;

  if (src >= src_end)
    /* We found nothing other than ASCII.  There's nothing to do.  */
    return 0;

  c = *src;
  /* The text seems to be encoded in some multilingual coding system.
     Now, try to find in which coding system the text is encoded.  */
  if (c < 0x80)
    {
      /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */
      /* C is an ISO2022 specific control code of C0.  */
      mask = detect_coding_iso2022 (src, src_end, multibytep);
      if (mask == 0)
	{
	  /* No valid ISO2022 code follows C.  Try again.  */
	  src++;
	  if (c == ISO_CODE_ESC)
	    ascii_skip_code[ISO_CODE_ESC] = 1;
	  else
	    ascii_skip_code[ISO_CODE_SO] = ascii_skip_code[ISO_CODE_SI] = 1;
	  goto label_loop_detect_coding;
	}
      if (priorities)
	{
	  for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++)
	    {
	      if (mask & priorities[i])
		return priorities[i];
	    }
	  return CODING_CATEGORY_MASK_RAW_TEXT;
	}
    }
  else
    {
      int try;

      if (multibytep && c == LEADING_CODE_8_BIT_CONTROL)
	c = src[1] - 0x20;

      if (c < 0xA0)
	{
	  /* C is the first byte of SJIS character code,
	     or a leading-code of Emacs' internal format (emacs-mule),
	     or the first byte of UTF-16.  */
	  try = (CODING_CATEGORY_MASK_SJIS
		  | CODING_CATEGORY_MASK_EMACS_MULE
		  | CODING_CATEGORY_MASK_UTF_16_BE
		  | CODING_CATEGORY_MASK_UTF_16_LE);

	  /* Or, if C is a special latin extra code,
	     or is an ISO2022 specific control code of C1 (SS2 or SS3),
	     or is an ISO2022 control-sequence-introducer (CSI),
	     we should also consider the possibility of ISO2022 codings.  */
	  if ((VECTORP (Vlatin_extra_code_table)
	       && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c]))
	      || (c == ISO_CODE_SS2 || c == ISO_CODE_SS3)
	      || (c == ISO_CODE_CSI
		  && (src < src_end
		      && (*src == ']'
			  || ((*src == '0' || *src == '1' || *src == '2')
			      && src + 1 < src_end
			      && src[1] == ']')))))
	    try |= (CODING_CATEGORY_MASK_ISO_8_ELSE
		     | CODING_CATEGORY_MASK_ISO_8BIT);
	}
      else
	/* C is a character of ISO2022 in graphic plane right,
	   or a SJIS's 1-byte character code (i.e. JISX0201),
	   or the first byte of BIG5's 2-byte code,
	   or the first byte of UTF-8/16.  */
	try = (CODING_CATEGORY_MASK_ISO_8_ELSE
		| CODING_CATEGORY_MASK_ISO_8BIT
		| CODING_CATEGORY_MASK_SJIS
		| CODING_CATEGORY_MASK_BIG5
	        | CODING_CATEGORY_MASK_UTF_8
	        | CODING_CATEGORY_MASK_UTF_16_BE
	        | CODING_CATEGORY_MASK_UTF_16_LE);

      /* Or, we may have to consider the possibility of CCL.  */
      if (coding_system_table[CODING_CATEGORY_IDX_CCL]
	  && (coding_system_table[CODING_CATEGORY_IDX_CCL]
	      ->spec.ccl.valid_codes)[c])
	try |= CODING_CATEGORY_MASK_CCL;

      mask = 0;
      utf16_examined_p = iso2022_examined_p = 0;
      if (priorities)
	{
	  for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++)
	    {
	      if (!iso2022_examined_p
		  && (priorities[i] & try & CODING_CATEGORY_MASK_ISO))
		{
		  mask |= detect_coding_iso2022 (src, src_end, multibytep);
		  iso2022_examined_p = 1;
		}
	      else if (priorities[i] & try & CODING_CATEGORY_MASK_SJIS)
		mask |= detect_coding_sjis (src, src_end, multibytep);
	      else if (priorities[i] & try & CODING_CATEGORY_MASK_UTF_8)
		mask |= detect_coding_utf_8 (src, src_end, multibytep);
	      else if (!utf16_examined_p
		       && (priorities[i] & try &
			   CODING_CATEGORY_MASK_UTF_16_BE_LE))
		{
		  mask |= detect_coding_utf_16 (src, src_end, multibytep);
		  utf16_examined_p = 1;
		}
	      else if (priorities[i] & try & CODING_CATEGORY_MASK_BIG5)
		mask |= detect_coding_big5 (src, src_end, multibytep);
	      else if (priorities[i] & try & CODING_CATEGORY_MASK_EMACS_MULE)
		mask |= detect_coding_emacs_mule (src, src_end, multibytep);
	      else if (priorities[i] & try & CODING_CATEGORY_MASK_CCL)
		mask |= detect_coding_ccl (src, src_end, multibytep);
	      else if (priorities[i] & CODING_CATEGORY_MASK_RAW_TEXT)
		mask |= CODING_CATEGORY_MASK_RAW_TEXT;
	      else if (priorities[i] & CODING_CATEGORY_MASK_BINARY)
		mask |= CODING_CATEGORY_MASK_BINARY;
	      if (mask & priorities[i])
		return priorities[i];
	    }
	  return CODING_CATEGORY_MASK_RAW_TEXT;
	}
      if (try & CODING_CATEGORY_MASK_ISO)
	mask |= detect_coding_iso2022 (src, src_end, multibytep);
      if (try & CODING_CATEGORY_MASK_SJIS)
	mask |= detect_coding_sjis (src, src_end, multibytep);
      if (try & CODING_CATEGORY_MASK_BIG5)
	mask |= detect_coding_big5 (src, src_end, multibytep);
      if (try & CODING_CATEGORY_MASK_UTF_8)
	mask |= detect_coding_utf_8 (src, src_end, multibytep);
      if (try & CODING_CATEGORY_MASK_UTF_16_BE_LE)
	mask |= detect_coding_utf_16 (src, src_end, multibytep);
      if (try & CODING_CATEGORY_MASK_EMACS_MULE)
	mask |= detect_coding_emacs_mule (src, src_end, multibytep);
      if (try & CODING_CATEGORY_MASK_CCL)
	mask |= detect_coding_ccl (src, src_end, multibytep);
    }
  return (mask | CODING_CATEGORY_MASK_RAW_TEXT | CODING_CATEGORY_MASK_BINARY);
}

/* Detect how a text of length SRC_BYTES pointed by SRC is encoded.
   The information of the detected coding system is set in CODING.  */

void
detect_coding (coding, src, src_bytes)
     struct coding_system *coding;
     const unsigned char *src;
     int src_bytes;
{
  unsigned int idx;
  int skip, mask;
  Lisp_Object val;

  val = Vcoding_category_list;
  mask = detect_coding_mask (src, src_bytes, coding_priorities, &skip,
			     coding->src_multibyte);
  coding->heading_ascii = skip;

  if (!mask) return;

  /* We found a single coding system of the highest priority in MASK.  */
  idx = 0;
  while (mask && ! (mask & 1)) mask >>= 1, idx++;
  if (! mask)
    idx = CODING_CATEGORY_IDX_RAW_TEXT;

  val = SYMBOL_VALUE (XVECTOR (Vcoding_category_table)->contents[idx]);

  if (coding->eol_type != CODING_EOL_UNDECIDED)
    {
      Lisp_Object tmp;

      tmp = Fget (val, Qeol_type);
      if (VECTORP (tmp))
	val = XVECTOR (tmp)->contents[coding->eol_type];
    }

  /* Setup this new coding system while preserving some slots.  */
  {
    int src_multibyte = coding->src_multibyte;
    int dst_multibyte = coding->dst_multibyte;

    setup_coding_system (val, coding);
    coding->src_multibyte = src_multibyte;
    coding->dst_multibyte = dst_multibyte;
    coding->heading_ascii = skip;
  }
}

/* Detect how end-of-line of a text of length SRC_BYTES pointed by
   SOURCE is encoded.  Return one of CODING_EOL_LF, CODING_EOL_CRLF,
   CODING_EOL_CR, and CODING_EOL_UNDECIDED.

   How many non-eol characters are at the head is returned as *SKIP.  */

#define MAX_EOL_CHECK_COUNT 3

static int
detect_eol_type (source, src_bytes, skip)
     unsigned char *source;
     int src_bytes, *skip;
{
  unsigned char *src = source, *src_end = src + src_bytes;
  unsigned char c;
  int total = 0;		/* How many end-of-lines are found so far.  */
  int eol_type = CODING_EOL_UNDECIDED;
  int this_eol_type;

  *skip = 0;

  while (src < src_end && total < MAX_EOL_CHECK_COUNT)
    {
      c = *src++;
      if (c == '\n' || c == '\r')
	{
	  if (*skip == 0)
	    *skip = src - 1 - source;
	  total++;
	  if (c == '\n')
	    this_eol_type = CODING_EOL_LF;
	  else if (src >= src_end || *src != '\n')
	    this_eol_type = CODING_EOL_CR;
	  else
	    this_eol_type = CODING_EOL_CRLF, src++;

	  if (eol_type == CODING_EOL_UNDECIDED)
	    /* This is the first end-of-line.  */
	    eol_type = this_eol_type;
	  else if (eol_type != this_eol_type)
	    {
	      /* The found type is different from what found before.  */
	      eol_type = CODING_EOL_INCONSISTENT;
	      break;
	    }
	}
    }

  if (*skip == 0)
    *skip = src_end - source;
  return eol_type;
}

/* Like detect_eol_type, but detect EOL type in 2-octet
   big-endian/little-endian format for coding systems utf-16-be and
   utf-16-le.  */

static int
detect_eol_type_in_2_octet_form (source, src_bytes, skip, big_endian_p)
     unsigned char *source;
     int src_bytes, *skip, big_endian_p;
{
  unsigned char *src = source, *src_end = src + src_bytes;
  unsigned int c1, c2;
  int total = 0;		/* How many end-of-lines are found so far.  */
  int eol_type = CODING_EOL_UNDECIDED;
  int this_eol_type;
  int msb, lsb;

  if (big_endian_p)
    msb = 0, lsb = 1;
  else
    msb = 1, lsb = 0;

  *skip = 0;

  while ((src + 1) < src_end && total < MAX_EOL_CHECK_COUNT)
    {
      c1 = (src[msb] << 8) | (src[lsb]);
      src += 2;

      if (c1 == '\n' || c1 == '\r')
	{
	  if (*skip == 0)
	    *skip = src - 2 - source;
	  total++;
	  if (c1 == '\n')
	    {
	      this_eol_type = CODING_EOL_LF;
	    }
	  else
	    {
	      if ((src + 1) >= src_end)
		{
		  this_eol_type = CODING_EOL_CR;
		}
	      else
		{
		  c2 = (src[msb] << 8) | (src[lsb]);
		  if (c2 == '\n')
		    this_eol_type = CODING_EOL_CRLF, src += 2;
		  else
		    this_eol_type = CODING_EOL_CR;
		}
	    }

	  if (eol_type == CODING_EOL_UNDECIDED)
	    /* This is the first end-of-line.  */
	    eol_type = this_eol_type;
	  else if (eol_type != this_eol_type)
	    {
	      /* The found type is different from what found before.  */
	      eol_type = CODING_EOL_INCONSISTENT;
	      break;
	    }
	}
    }

  if (*skip == 0)
    *skip = src_end - source;
  return eol_type;
}

/* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC
   is encoded.  If it detects an appropriate format of end-of-line, it
   sets the information in *CODING.  */

void
detect_eol (coding, src, src_bytes)
     struct coding_system *coding;
     const unsigned char *src;
     int src_bytes;
{
  Lisp_Object val;
  int skip;
  int eol_type;

  switch (coding->category_idx)
    {
    case CODING_CATEGORY_IDX_UTF_16_BE:
      eol_type = detect_eol_type_in_2_octet_form (src, src_bytes, &skip, 1);
      break;
    case CODING_CATEGORY_IDX_UTF_16_LE:
      eol_type = detect_eol_type_in_2_octet_form (src, src_bytes, &skip, 0);
      break;
    default:
      eol_type = detect_eol_type (src, src_bytes, &skip);
      break;
    }

  if (coding->heading_ascii > skip)
    coding->heading_ascii = skip;
  else
    skip = coding->heading_ascii;

  if (eol_type == CODING_EOL_UNDECIDED)
    return;
  if (eol_type == CODING_EOL_INCONSISTENT)
    {
#if 0
      /* This code is suppressed until we find a better way to
	 distinguish raw text file and binary file.  */

      /* If we have already detected that the coding is raw-text, the
	 coding should actually be no-conversion.  */
      if (coding->type == coding_type_raw_text)
	{
	  setup_coding_system (Qno_conversion, coding);
	  return;
	}
      /* Else, let's decode only text code anyway.  */
#endif /* 0 */
      eol_type = CODING_EOL_LF;
    }

  val = Fget (coding->symbol, Qeol_type);
  if (VECTORP (val) && XVECTOR (val)->size == 3)
    {
      int src_multibyte = coding->src_multibyte;
      int dst_multibyte = coding->dst_multibyte;
      struct composition_data *cmp_data = coding->cmp_data;

      setup_coding_system (XVECTOR (val)->contents[eol_type], coding);
      coding->src_multibyte = src_multibyte;
      coding->dst_multibyte = dst_multibyte;
      coding->heading_ascii = skip;
      coding->cmp_data = cmp_data;
    }
}

#define CONVERSION_BUFFER_EXTRA_ROOM 256

#define DECODING_BUFFER_MAG(coding)			\
  (coding->type == coding_type_iso2022			\
   ? 3							\
   : (coding->type == coding_type_ccl			\
      ? coding->spec.ccl.decoder.buf_magnification	\
      : 2))

/* Return maximum size (bytes) of a buffer enough for decoding
   SRC_BYTES of text encoded in CODING.  */

int
decoding_buffer_size (coding, src_bytes)
     struct coding_system *coding;
     int src_bytes;
{
  return (src_bytes * DECODING_BUFFER_MAG (coding)
	  + CONVERSION_BUFFER_EXTRA_ROOM);
}

/* Return maximum size (bytes) of a buffer enough for encoding
   SRC_BYTES of text to CODING.  */

int
encoding_buffer_size (coding, src_bytes)
     struct coding_system *coding;
     int src_bytes;
{
  int magnification;

  if (coding->type == coding_type_ccl)
    {
      magnification = coding->spec.ccl.encoder.buf_magnification;
      if (coding->eol_type == CODING_EOL_CRLF)
	magnification *= 2;
    }
  else if (CODING_REQUIRE_ENCODING (coding))
    magnification = 3;
  else
    magnification = 1;

  return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM);
}

/* Working buffer for code conversion.  */
struct conversion_buffer
{
  int size;			/* size of data.  */
  int on_stack;			/* 1 if allocated by alloca.  */
  unsigned char *data;
};

/* Allocate LEN bytes of memory for BUF (struct conversion_buffer).  */
#define allocate_conversion_buffer(buf, len)		\
  do {							\
    if (len < MAX_ALLOCA)				\
      {							\
	buf.data = (unsigned char *) alloca (len);	\
	buf.on_stack = 1;				\
      }							\
    else						\
      {							\
	buf.data = (unsigned char *) xmalloc (len);	\
	buf.on_stack = 0;				\
      }							\
    buf.size = len;					\
  } while (0)

/* Double the allocated memory for *BUF.  */
static void
extend_conversion_buffer (buf)
     struct conversion_buffer *buf;
{
  if (buf->on_stack)
    {
      unsigned char *save = buf->data;
      buf->data = (unsigned char *) xmalloc (buf->size * 2);
      bcopy (save, buf->data, buf->size);
      buf->on_stack = 0;
    }
  else
    {
      buf->data = (unsigned char *) xrealloc (buf->data, buf->size * 2);
    }
  buf->size *= 2;
}

/* Free the allocated memory for BUF if it is not on stack.  */
static void
free_conversion_buffer (buf)
     struct conversion_buffer *buf;
{
  if (!buf->on_stack)
    xfree (buf->data);
}

int
ccl_coding_driver (coding, source, destination, src_bytes, dst_bytes, encodep)
     struct coding_system *coding;
     unsigned char *source, *destination;
     int src_bytes, dst_bytes, encodep;
{
  struct ccl_program *ccl
    = encodep ? &coding->spec.ccl.encoder : &coding->spec.ccl.decoder;
  unsigned char *dst = destination;

  ccl->suppress_error = coding->suppress_error;
  ccl->last_block = coding->mode & CODING_MODE_LAST_BLOCK;
  if (encodep)
    {
      /* On encoding, EOL format is converted within ccl_driver.  For
	 that, setup proper information in the structure CCL.  */
      ccl->eol_type = coding->eol_type;
      if (ccl->eol_type ==CODING_EOL_UNDECIDED)
	ccl->eol_type = CODING_EOL_LF;
      ccl->cr_consumed = coding->spec.ccl.cr_carryover;
      ccl->eight_bit_control = coding->dst_multibyte;
    }
  else
    ccl->eight_bit_control = 1;
  ccl->multibyte = coding->src_multibyte;
  if (coding->spec.ccl.eight_bit_carryover[0] != 0)
    {
      /* Move carryover bytes to DESTINATION.  */
      unsigned char *p = coding->spec.ccl.eight_bit_carryover;
      while (*p)
	*dst++ = *p++;
      coding->spec.ccl.eight_bit_carryover[0] = 0;
      if (dst_bytes)
	dst_bytes -= dst - destination;
    }

  coding->produced = (ccl_driver (ccl, source, dst, src_bytes, dst_bytes,
				  &(coding->consumed))
		      + dst - destination);

  if (encodep)
    {
      coding->produced_char = coding->produced;
      coding->spec.ccl.cr_carryover = ccl->cr_consumed;
    }
  else if (!ccl->eight_bit_control)
    {
      /* The produced bytes forms a valid multibyte sequence. */
      coding->produced_char
	= multibyte_chars_in_text (destination, coding->produced);
      coding->spec.ccl.eight_bit_carryover[0] = 0;
    }
  else
    {
      /* On decoding, the destination should always multibyte.  But,
	 CCL program might have been generated an invalid multibyte
	 sequence.  Here we make such a sequence valid as
	 multibyte.  */
      int bytes
	= dst_bytes ? dst_bytes : source + coding->consumed - destination;

      if ((coding->consumed < src_bytes
	   || !ccl->last_block)
	  && coding->produced >= 1
	  && destination[coding->produced - 1] >= 0x80)
	{
	  /* We should not convert the tailing 8-bit codes to
	     multibyte form even if they doesn't form a valid
	     multibyte sequence.  They may form a valid sequence in
	     the next call.  */
	  int carryover = 0;

	  if (destination[coding->produced - 1] < 0xA0)
	    carryover = 1;
	  else if (coding->produced >= 2)
	    {
	      if (destination[coding->produced - 2] >= 0x80)
		{
		  if (destination[coding->produced - 2] < 0xA0)
		    carryover = 2;
		  else if (coding->produced >= 3
			   && destination[coding->produced - 3] >= 0x80
			   && destination[coding->produced - 3] < 0xA0)
		    carryover = 3;
		}
	    }
	  if (carryover > 0)
	    {
	      BCOPY_SHORT (destination + coding->produced - carryover,
			   coding->spec.ccl.eight_bit_carryover,
			   carryover);
	      coding->spec.ccl.eight_bit_carryover[carryover] = 0;
	      coding->produced -= carryover;
	    }
	}
      coding->produced = str_as_multibyte (destination, bytes,
					   coding->produced,
					   &(coding->produced_char));
    }

  switch (ccl->status)
    {
    case CCL_STAT_SUSPEND_BY_SRC:
      coding->result = CODING_FINISH_INSUFFICIENT_SRC;
      break;
    case CCL_STAT_SUSPEND_BY_DST:
      coding->result = CODING_FINISH_INSUFFICIENT_DST;
      break;
    case CCL_STAT_QUIT:
    case CCL_STAT_INVALID_CMD:
      coding->result = CODING_FINISH_INTERRUPT;
      break;
    default:
      coding->result = CODING_FINISH_NORMAL;
      break;
    }
  return coding->result;
}

/* Decode EOL format of the text at PTR of BYTES length destructively
   according to CODING->eol_type.  This is called after the CCL
   program produced a decoded text at PTR.  If we do CRLF->LF
   conversion, update CODING->produced and CODING->produced_char.  */

static void
decode_eol_post_ccl (coding, ptr, bytes)
     struct coding_system *coding;
     unsigned char *ptr;
     int bytes;
{
  Lisp_Object val, saved_coding_symbol;
  unsigned char *pend = ptr + bytes;
  int dummy;

  /* Remember the current coding system symbol.  We set it back when
     an inconsistent EOL is found so that `last-coding-system-used' is
     set to the coding system that doesn't specify EOL conversion.  */
  saved_coding_symbol = coding->symbol;

  coding->spec.ccl.cr_carryover = 0;
  if (coding->eol_type == CODING_EOL_UNDECIDED)
    {
      /* Here, to avoid the call of setup_coding_system, we directly
	 call detect_eol_type.  */
      coding->eol_type = detect_eol_type (ptr, bytes, &dummy);
      if (coding->eol_type == CODING_EOL_INCONSISTENT)
	coding->eol_type = CODING_EOL_LF;
      if (coding->eol_type != CODING_EOL_UNDECIDED)
	{
	  val = Fget (coding->symbol, Qeol_type);
	  if (VECTORP (val) && XVECTOR (val)->size == 3)
	    coding->symbol = XVECTOR (val)->contents[coding->eol_type];
	}
      coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL;
    }

  if (coding->eol_type == CODING_EOL_LF
      || coding->eol_type == CODING_EOL_UNDECIDED)
    {
      /* We have nothing to do.  */
      ptr = pend;
    }
  else if (coding->eol_type == CODING_EOL_CRLF)
    {
      unsigned char *pstart = ptr, *p = ptr;

      if (! (coding->mode & CODING_MODE_LAST_BLOCK)
	  && *(pend - 1) == '\r')
	{
	  /* If the last character is CR, we can't handle it here
	     because LF will be in the not-yet-decoded source text.
	     Record that the CR is not yet processed.  */
	  coding->spec.ccl.cr_carryover = 1;
	  coding->produced--;
	  coding->produced_char--;
	  pend--;
	}
      while (ptr < pend)
	{
	  if (*ptr == '\r')
	    {
	      if (ptr + 1 < pend && *(ptr + 1) == '\n')
		{
		  *p++ = '\n';
		  ptr += 2;
		}
	      else
		{
		  if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
		    goto undo_eol_conversion;
		  *p++ = *ptr++;
		}
	    }
	  else if (*ptr == '\n'
		   && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
	    goto undo_eol_conversion;
	  else
	    *p++ = *ptr++;
	  continue;

	undo_eol_conversion:
	  /* We have faced with inconsistent EOL format at PTR.
	     Convert all LFs before PTR back to CRLFs.  */
	  for (p--, ptr--; p >= pstart; p--)
	    {
	      if (*p == '\n')
		*ptr-- = '\n', *ptr-- = '\r';
	      else
		*ptr-- = *p;
	    }
	  /*  If carryover is recorded, cancel it because we don't
	      convert CRLF anymore.  */
	  if (coding->spec.ccl.cr_carryover)
	    {
	      coding->spec.ccl.cr_carryover = 0;
	      coding->produced++;
	      coding->produced_char++;
	      pend++;
	    }
	  p = ptr = pend;
	  coding->eol_type = CODING_EOL_LF;
	  coding->symbol = saved_coding_symbol;
	}
      if (p < pend)
	{
	  /* As each two-byte sequence CRLF was converted to LF, (PEND
	     - P) is the number of deleted characters.  */
	  coding->produced -= pend - p;
	  coding->produced_char -= pend - p;
	}
    }
  else			/* i.e. coding->eol_type == CODING_EOL_CR */
    {
      unsigned char *p = ptr;

      for (; ptr < pend; ptr++)
	{
	  if (*ptr == '\r')
	    *ptr = '\n';
	  else if (*ptr == '\n'
		   && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
	    {
	      for (; p < ptr; p++)
		{
		  if (*p == '\n')
		    *p = '\r';
		}
	      ptr = pend;
	      coding->eol_type = CODING_EOL_LF;
	      coding->symbol = saved_coding_symbol;
	    }
	}
    }
}

/* See "GENERAL NOTES about `decode_coding_XXX ()' functions".  Before
   decoding, it may detect coding system and format of end-of-line if
   those are not yet decided.  The source should be unibyte, the
   result is multibyte if CODING->dst_multibyte is nonzero, else
   unibyte.  */

int
decode_coding (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  int extra = 0;

  if (coding->type == coding_type_undecided)
    detect_coding (coding, source, src_bytes);

  if (coding->eol_type == CODING_EOL_UNDECIDED
      && coding->type != coding_type_ccl)
    {
      detect_eol (coding, source, src_bytes);
      /* We had better recover the original eol format if we
	 encounter an inconsistent eol format while decoding.  */
      coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL;
    }

  coding->produced = coding->produced_char = 0;
  coding->consumed = coding->consumed_char = 0;
  coding->errors = 0;
  coding->result = CODING_FINISH_NORMAL;

  switch (coding->type)
    {
    case coding_type_sjis:
      decode_coding_sjis_big5 (coding, source, destination,
			       src_bytes, dst_bytes, 1);
      break;

    case coding_type_iso2022:
      decode_coding_iso2022 (coding, source, destination,
			     src_bytes, dst_bytes);
      break;

    case coding_type_big5:
      decode_coding_sjis_big5 (coding, source, destination,
			       src_bytes, dst_bytes, 0);
      break;

    case coding_type_emacs_mule:
      decode_coding_emacs_mule (coding, source, destination,
				src_bytes, dst_bytes);
      break;

    case coding_type_ccl:
      if (coding->spec.ccl.cr_carryover)
	{
	  /* Put the CR which was not processed by the previous call
	     of decode_eol_post_ccl in DESTINATION.  It will be
	     decoded together with the following LF by the call to
	     decode_eol_post_ccl below.  */
	  *destination = '\r';
	  coding->produced++;
	  coding->produced_char++;
	  dst_bytes--;
	  extra = coding->spec.ccl.cr_carryover;
	}
      ccl_coding_driver (coding, source, destination + extra,
			 src_bytes, dst_bytes, 0);
      if (coding->eol_type != CODING_EOL_LF)
	{
	  coding->produced += extra;
	  coding->produced_char += extra;
	  decode_eol_post_ccl (coding, destination, coding->produced);
	}
      break;

    default:
      decode_eol (coding, source, destination, src_bytes, dst_bytes);
    }

  if (coding->result == CODING_FINISH_INSUFFICIENT_SRC
      && coding->mode & CODING_MODE_LAST_BLOCK
      && coding->consumed == src_bytes)
    coding->result = CODING_FINISH_NORMAL;

  if (coding->mode & CODING_MODE_LAST_BLOCK
      && coding->result == CODING_FINISH_INSUFFICIENT_SRC)
    {
      const unsigned char *src = source + coding->consumed;
      unsigned char *dst = destination + coding->produced;

      src_bytes -= coding->consumed;
      coding->errors++;
      if (COMPOSING_P (coding))
	DECODE_COMPOSITION_END ('1');
      while (src_bytes--)
	{
	  int c = *src++;
	  dst += CHAR_STRING (c, dst);
	  coding->produced_char++;
	}
      coding->consumed = coding->consumed_char = src - source;
      coding->produced = dst - destination;
      coding->result = CODING_FINISH_NORMAL;
    }

  if (!coding->dst_multibyte)
    {
      coding->produced = str_as_unibyte (destination, coding->produced);
      coding->produced_char = coding->produced;
    }

  return coding->result;
}

/* See "GENERAL NOTES about `encode_coding_XXX ()' functions".  The
   multibyteness of the source is CODING->src_multibyte, the
   multibyteness of the result is always unibyte.  */

int
encode_coding (coding, source, destination, src_bytes, dst_bytes)
     struct coding_system *coding;
     const unsigned char *source;
     unsigned char *destination;
     int src_bytes, dst_bytes;
{
  coding->produced = coding->produced_char = 0;
  coding->consumed = coding->consumed_char = 0;
  coding->errors = 0;
  coding->result = CODING_FINISH_NORMAL;
  if (coding->eol_type == CODING_EOL_UNDECIDED)
    coding->eol_type = CODING_EOL_LF;

  switch (coding->type)
    {
    case coding_type_sjis:
      encode_coding_sjis_big5 (coding, source, destination,
			       src_bytes, dst_bytes, 1);
      break;

    case coding_type_iso2022:
      encode_coding_iso2022 (coding, source, destination,
			     src_bytes, dst_bytes);
      break;

    case coding_type_big5:
      encode_coding_sjis_big5 (coding, source, destination,
			       src_bytes, dst_bytes, 0);
      break;

    case coding_type_emacs_mule:
      encode_coding_emacs_mule (coding, source, destination,
				src_bytes, dst_bytes);
      break;

    case coding_type_ccl:
      ccl_coding_driver (coding, source, destination,
			 src_bytes, dst_bytes, 1);
      break;

    default:
      encode_eol (coding, source, destination, src_bytes, dst_bytes);
    }

  if (coding->mode & CODING_MODE_LAST_BLOCK
      && coding->result == CODING_FINISH_INSUFFICIENT_SRC)
    {
      const unsigned char *src = source + coding->consumed;
      unsigned char *dst = destination + coding->produced;

      if (coding->type == coding_type_iso2022)
	ENCODE_RESET_PLANE_AND_REGISTER;
      if (COMPOSING_P (coding))
	*dst++ = ISO_CODE_ESC, *dst++ = '1';
      if (coding->consumed < src_bytes)
	{
	  int len = src_bytes - coding->consumed;

	  BCOPY_SHORT (src, dst, len);
	  if (coding->src_multibyte)
	    len = str_as_unibyte (dst, len);
	  dst += len;
	  coding->consumed = src_bytes;
	}
      coding->produced = coding->produced_char = dst - destination;
      coding->result = CODING_FINISH_NORMAL;
    }

  if (coding->result == CODING_FINISH_INSUFFICIENT_SRC
      && coding->consumed == src_bytes)
    coding->result = CODING_FINISH_NORMAL;

  return coding->result;
}

/* Scan text in the region between *BEG and *END (byte positions),
   skip characters which we don't have to decode by coding system
   CODING at the head and tail, then set *BEG and *END to the region
   of the text we actually have to convert.  The caller should move
   the gap out of the region in advance if the region is from a
   buffer.

   If STR is not NULL, *BEG and *END are indices into STR.  */

static void
shrink_decoding_region (beg, end, coding, str)
     int *beg, *end;
     struct coding_system *coding;
     unsigned char *str;
{
  unsigned char *begp_orig, *begp, *endp_orig, *endp, c;
  int eol_conversion;
  Lisp_Object translation_table;

  if (coding->type == coding_type_ccl
      || coding->type == coding_type_undecided
      || coding->eol_type != CODING_EOL_LF
      || !NILP (coding->post_read_conversion)
      || coding->composing != COMPOSITION_DISABLED)
    {
      /* We can't skip any data.  */
      return;
    }
  if (coding->type == coding_type_no_conversion
      || coding->type == coding_type_raw_text
      || coding->type == coding_type_emacs_mule)
    {
      /* We need no conversion, but don't have to skip any data here.
         Decoding routine handles them effectively anyway.  */
      return;
    }

  translation_table = coding->translation_table_for_decode;
  if (NILP (translation_table) && !NILP (Venable_character_translation))
    translation_table = Vstandard_translation_table_for_decode;
  if (CHAR_TABLE_P (translation_table))
    {
      int i;
      for (i = 0; i < 128; i++)
	if (!NILP (CHAR_TABLE_REF (translation_table, i)))
	  break;
      if (i < 128)
	/* Some ASCII character should be translated.  We give up
	   shrinking.  */
	return;
    }

  if (coding->heading_ascii >= 0)
    /* Detection routine has already found how much we can skip at the
       head.  */
    *beg += coding->heading_ascii;

  if (str)
    {
      begp_orig = begp = str + *beg;
      endp_orig = endp = str + *end;
    }
  else
    {
      begp_orig = begp = BYTE_POS_ADDR (*beg);
      endp_orig = endp = begp + *end - *beg;
    }

  eol_conversion = (coding->eol_type == CODING_EOL_CR
		    || coding->eol_type == CODING_EOL_CRLF);

  switch (coding->type)
    {
    case coding_type_sjis:
    case coding_type_big5:
      /* We can skip all ASCII characters at the head.  */
      if (coding->heading_ascii < 0)
	{
	  if (eol_conversion)
	    while (begp < endp && *begp < 0x80 && *begp != '\r') begp++;
	  else
	    while (begp < endp && *begp < 0x80) begp++;
	}
      /* We can skip all ASCII characters at the tail except for the
	 second byte of SJIS or BIG5 code.  */
      if (eol_conversion)
	while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\r') endp--;
      else
	while (begp < endp && endp[-1] < 0x80) endp--;
      /* Do not consider LF as ascii if preceded by CR, since that
	 confuses eol decoding. */
      if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n')
	endp++;
      if (begp < endp && endp < endp_orig && endp[-1] >= 0x80)
	endp++;
      break;

    case coding_type_iso2022:
      if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, 0) != CHARSET_ASCII)
	/* We can't skip any data.  */
	break;
      if (coding->heading_ascii < 0)
	{
	  /* We can skip all ASCII characters at the head except for a
	     few control codes.  */
	  while (begp < endp && (c = *begp) < 0x80
		 && c != ISO_CODE_CR && c != ISO_CODE_SO
		 && c != ISO_CODE_SI && c != ISO_CODE_ESC
		 && (!eol_conversion || c != ISO_CODE_LF))
	    begp++;
	}
      switch (coding->category_idx)
	{
	case CODING_CATEGORY_IDX_ISO_8_1:
	case CODING_CATEGORY_IDX_ISO_8_2:
	  /* We can skip all ASCII characters at the tail.  */
	  if (eol_conversion)
	    while (begp < endp && (c = endp[-1]) < 0x80 && c != '\r') endp--;
	  else
	    while (begp < endp && endp[-1] < 0x80) endp--;
	  /* Do not consider LF as ascii if preceded by CR, since that
             confuses eol decoding. */
	  if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n')
	    endp++;
	  break;

	case CODING_CATEGORY_IDX_ISO_7:
	case CODING_CATEGORY_IDX_ISO_7_TIGHT:
	  {
	    /* We can skip all characters at the tail except for 8-bit
	       codes and ESC and the following 2-byte at the tail.  */
	    unsigned char *eight_bit = NULL;

	    if (eol_conversion)
	      while (begp < endp
		     && (c = endp[-1]) != ISO_CODE_ESC && c != '\r')
		{
		  if (!eight_bit && c & 0x80) eight_bit = endp;
		  endp--;
		}
	    else
	      while (begp < endp
		     && (c = endp[-1]) != ISO_CODE_ESC)
		{
		  if (!eight_bit && c & 0x80) eight_bit = endp;
		  endp--;
		}
	    /* Do not consider LF as ascii if preceded by CR, since that
	       confuses eol decoding. */
	    if (begp < endp && endp < endp_orig
		&& endp[-1] == '\r' && endp[0] == '\n')
	      endp++;
	    if (begp < endp && endp[-1] == ISO_CODE_ESC)
	      {
		if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B')
		  /* This is an ASCII designation sequence.  We can
		     surely skip the tail.  But, if we have
		     encountered an 8-bit code, skip only the codes
		     after that.  */
		  endp = eight_bit ? eight_bit : endp + 2;
		else
		  /* Hmmm, we can't skip the tail.  */
		  endp = endp_orig;
	      }
	    else if (eight_bit)
	      endp = eight_bit;
	  }
	}
      break;

    default:
      abort ();
    }
  *beg += begp - begp_orig;
  *end += endp - endp_orig;
  return;
}

/* Like shrink_decoding_region but for encoding.  */

static void
shrink_encoding_region (beg, end, coding, str)
     int *beg, *end;
     struct coding_system *coding;
     unsigned char *str;
{
  unsigned char *begp_orig, *begp, *endp_orig, *endp;
  int eol_conversion;
  Lisp_Object translation_table;

  if (coding->type == coding_type_ccl
      || coding->eol_type == CODING_EOL_CRLF
      || coding->eol_type == CODING_EOL_CR
      || (coding->cmp_data && coding->cmp_data->used > 0))
    {
      /* We can't skip any data.  */
      return;
    }
  if (coding->type == coding_type_no_conversion
      || coding->type == coding_type_raw_text
      || coding->type == coding_type_emacs_mule
      || coding->type == coding_type_undecided)
    {
      /* We need no conversion, but don't have to skip any data here.
         Encoding routine handles them effectively anyway.  */
      return;
    }

  translation_table = coding->translation_table_for_encode;
  if (NILP (translation_table) && !NILP (Venable_character_translation))
    translation_table = Vstandard_translation_table_for_encode;
  if (CHAR_TABLE_P (translation_table))
    {
      int i;
      for (i = 0; i < 128; i++)
	if (!NILP (CHAR_TABLE_REF (translation_table, i)))
	  break;
      if (i < 128)
	/* Some ASCII character should be translated.  We give up
	   shrinking.  */
	return;
    }

  if (str)
    {
      begp_orig = begp = str + *beg;
      endp_orig = endp = str + *end;
    }
  else
    {
      begp_orig = begp = BYTE_POS_ADDR (*beg);
      endp_orig = endp = begp + *end - *beg;
    }

  eol_conversion = (coding->eol_type == CODING_EOL_CR
		    || coding->eol_type == CODING_EOL_CRLF);

  /* Here, we don't have to check coding->pre_write_conversion because
     the caller is expected to have handled it already.  */
  switch (coding->type)
    {
    case coding_type_iso2022:
      if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, 0) != CHARSET_ASCII)
	/* We can't skip any data.  */
	break;
      if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL)
	{
	  unsigned char *bol = begp;
	  while (begp < endp && *begp < 0x80)
	    {
	      begp++;
	      if (begp[-1] == '\n')
		bol = begp;
	    }
	  begp = bol;
	  goto label_skip_tail;
	}
      /* fall down ... */

    case coding_type_sjis:
    case coding_type_big5:
      /* We can skip all ASCII characters at the head and tail.  */
      if (eol_conversion)
	while (begp < endp && *begp < 0x80 && *begp != '\n') begp++;
      else
	while (begp < endp && *begp < 0x80) begp++;
    label_skip_tail:
      if (eol_conversion)
	while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\n') endp--;
      else
	while (begp < endp && *(endp - 1) < 0x80) endp--;
      break;

    default:
      abort ();
    }

  *beg += begp - begp_orig;
  *end += endp - endp_orig;
  return;
}

/* As shrinking conversion region requires some overhead, we don't try
   shrinking if the length of conversion region is less than this
   value.  */
static int shrink_conversion_region_threshhold = 1024;

#define SHRINK_CONVERSION_REGION(beg, end, coding, str, encodep)	\
  do {									\
    if (*(end) - *(beg) > shrink_conversion_region_threshhold)		\
      {									\
        if (encodep) shrink_encoding_region (beg, end, coding, str);	\
        else shrink_decoding_region (beg, end, coding, str);		\
      }									\
  } while (0)

/* ARG is (CODING BUFFER ...) where CODING is what to be set in
   Vlast_coding_system_used and the remaining elements are buffers to
   kill.  */
static Lisp_Object
code_convert_region_unwind (arg)
     Lisp_Object arg;
{
  struct gcpro gcpro1;
  GCPRO1 (arg);

  inhibit_pre_post_conversion = 0;
  Vlast_coding_system_used = XCAR (arg);
  for (arg = XCDR (arg); ! NILP (arg); arg = XCDR (arg))
    Fkill_buffer (XCAR (arg));

  UNGCPRO;
  return Qnil;
}

/* Store information about all compositions in the range FROM and TO
   of OBJ in memory blocks pointed by CODING->cmp_data.  OBJ is a
   buffer or a string, defaults to the current buffer.  */

void
coding_save_composition (coding, from, to, obj)
     struct coding_system *coding;
     int from, to;
     Lisp_Object obj;
{
  Lisp_Object prop;
  int start, end;

  if (coding->composing == COMPOSITION_DISABLED)
    return;
  if (!coding->cmp_data)
    coding_allocate_composition_data (coding, from);
  if (!find_composition (from, to, &start, &end, &prop, obj)
      || end > to)
    return;
  if (start < from
      && (!find_composition (end, to, &start, &end, &prop, obj)
	  || end > to))
    return;
  coding->composing = COMPOSITION_NO;
  do
    {
      if (COMPOSITION_VALID_P (start, end, prop))
	{
	  enum composition_method method = COMPOSITION_METHOD (prop);
	  if (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH
	      >= COMPOSITION_DATA_SIZE)
	    coding_allocate_composition_data (coding, from);
	  /* For relative composition, we remember start and end
             positions, for the other compositions, we also remember
             components.  */
	  CODING_ADD_COMPOSITION_START (coding, start - from, method);
	  if (method != COMPOSITION_RELATIVE)
	    {
	      /* We must store a*/
	      Lisp_Object val, ch;

	      val = COMPOSITION_COMPONENTS (prop);
	      if (CONSP (val))
		while (CONSP (val))
		  {
		    ch = XCAR (val), val = XCDR (val);
		    CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (ch));
		  }
	      else if (VECTORP (val) || STRINGP (val))
		{
		  int len = (VECTORP (val)
			     ? XVECTOR (val)->size : SCHARS (val));
		  int i;
		  for (i = 0; i < len; i++)
		    {
		      ch = (STRINGP (val)
			    ? Faref (val, make_number (i))
			    : XVECTOR (val)->contents[i]);
		      CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (ch));
		    }
		}
	      else		/* INTEGERP (val) */
		CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (val));
	    }
	  CODING_ADD_COMPOSITION_END (coding, end - from);
	}
      start = end;
    }
  while (start < to
	 && find_composition (start, to, &start, &end, &prop, obj)
	 && end <= to);

  /* Make coding->cmp_data point to the first memory block.  */
  while (coding->cmp_data->prev)
    coding->cmp_data = coding->cmp_data->prev;
  coding->cmp_data_start = 0;
}

/* Reflect the saved information about compositions to OBJ.
   CODING->cmp_data points to a memory block for the information.  OBJ
   is a buffer or a string, defaults to the current buffer.  */

void
coding_restore_composition (coding, obj)
     struct coding_system *coding;
     Lisp_Object obj;
{
  struct composition_data *cmp_data = coding->cmp_data;

  if (!cmp_data)
    return;

  while (cmp_data->prev)
    cmp_data = cmp_data->prev;

  while (cmp_data)
    {
      int i;

      for (i = 0; i < cmp_data->used && cmp_data->data[i] > 0;
	   i += cmp_data->data[i])
	{
	  int *data = cmp_data->data + i;
	  enum composition_method method = (enum composition_method) data[3];
	  Lisp_Object components;

	  if (data[0] < 0 || i + data[0] > cmp_data->used)
	    /* Invalid composition data.  */
	    break;

	  if (method == COMPOSITION_RELATIVE)
	    components = Qnil;
	  else
	    {
	      int len = data[0] - 4, j;
	      Lisp_Object args[MAX_COMPOSITION_COMPONENTS * 2 - 1];

	      if (method == COMPOSITION_WITH_RULE_ALTCHARS
		  && len % 2 == 0)
		len --;
	      if (len < 1)
		/* Invalid composition data.  */
		break;
	      for (j = 0; j < len; j++)
		args[j] = make_number (data[4 + j]);
	      components = (method == COMPOSITION_WITH_ALTCHARS
			    ? Fstring (len, args)
			    : Fvector (len, args));
	    }
	  compose_text (data[1], data[2], components, Qnil, obj);
	}
      cmp_data = cmp_data->next;
    }
}

/* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the
   text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by
   coding system CODING, and return the status code of code conversion
   (currently, this value has no meaning).

   How many characters (and bytes) are converted to how many
   characters (and bytes) are recorded in members of the structure
   CODING.

   If REPLACE is nonzero, we do various things as if the original text
   is deleted and a new text is inserted.  See the comments in
   replace_range (insdel.c) to know what we are doing.

   If REPLACE is zero, it is assumed that the source text is unibyte.
   Otherwise, it is assumed that the source text is multibyte.  */

int
code_convert_region (from, from_byte, to, to_byte, coding, encodep, replace)
     int from, from_byte, to, to_byte, encodep, replace;
     struct coding_system *coding;
{
  int len = to - from, len_byte = to_byte - from_byte;
  int nchars_del = 0, nbytes_del = 0;
  int require, inserted, inserted_byte;
  int head_skip, tail_skip, total_skip = 0;
  Lisp_Object saved_coding_symbol;
  int first = 1;
  unsigned char *src, *dst;
  Lisp_Object deletion;
  int orig_point = PT, orig_len = len;
  int prev_Z;
  int multibyte_p = !NILP (current_buffer->enable_multibyte_characters);

  deletion = Qnil;
  saved_coding_symbol = coding->symbol;

  if (from < PT && PT < to)
    {
      TEMP_SET_PT_BOTH (from, from_byte);
      orig_point = from;
    }

  if (replace)
    {
      int saved_from = from;
      int saved_inhibit_modification_hooks;

      prepare_to_modify_buffer (from, to, &from);
      if (saved_from != from)
	{
	  to = from + len;
	  from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to);
	  len_byte = to_byte - from_byte;
	}

      /* The code conversion routine can not preserve text properties
	 for now.  So, we must remove all text properties in the
	 region.  Here, we must suppress all modification hooks.  */
      saved_inhibit_modification_hooks = inhibit_modification_hooks;
      inhibit_modification_hooks = 1;
      Fset_text_properties (make_number (from), make_number (to), Qnil, Qnil);
      inhibit_modification_hooks = saved_inhibit_modification_hooks;
    }

  coding->heading_ascii = 0;

  if (! encodep && CODING_REQUIRE_DETECTION (coding))
    {
      /* We must detect encoding of text and eol format.  */

      if (from < GPT && to > GPT)
	move_gap_both (from, from_byte);
      if (coding->type == coding_type_undecided)
	{
	  detect_coding (coding, BYTE_POS_ADDR (from_byte), len_byte);
	  if (coding->type == coding_type_undecided)
	    {
	      /* It seems that the text contains only ASCII, but we
		 should not leave it undecided because the deeper
		 decoding routine (decode_coding) tries to detect the
		 encodings again in vain.  */
	      coding->type = coding_type_emacs_mule;
	      coding->category_idx = CODING_CATEGORY_IDX_EMACS_MULE;
	      /* As emacs-mule decoder will handle composition, we
		 need this setting to allocate coding->cmp_data
		 later.  */
	      coding->composing = COMPOSITION_NO;
	    }
	}
      if (coding->eol_type == CODING_EOL_UNDECIDED
	  && coding->type != coding_type_ccl)
	{
	  detect_eol (coding, BYTE_POS_ADDR (from_byte), len_byte);
	  if (coding->eol_type == CODING_EOL_UNDECIDED)
	    coding->eol_type = CODING_EOL_LF;
	  /* We had better recover the original eol format if we
	     encounter an inconsistent eol format while decoding.  */
	  coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL;
	}
    }

  /* Now we convert the text.  */

  /* For encoding, we must process pre-write-conversion in advance.  */
  if (! inhibit_pre_post_conversion
      && encodep
      && SYMBOLP (coding->pre_write_conversion)
      && ! NILP (Ffboundp (coding->pre_write_conversion)))
    {
      /* The function in pre-write-conversion may put a new text in a
         new buffer.  */
      struct buffer *prev = current_buffer;
      Lisp_Object new;

      record_unwind_protect (code_convert_region_unwind,
			     Fcons (Vlast_coding_system_used, Qnil));
      /* We should not call any more pre-write/post-read-conversion
         functions while this pre-write-conversion is running.  */
      inhibit_pre_post_conversion = 1;
      call2 (coding->pre_write_conversion,
	     make_number (from), make_number (to));
      inhibit_pre_post_conversion = 0;
      /* Discard the unwind protect.  */
      specpdl_ptr--;

      if (current_buffer != prev)
	{
	  len = ZV - BEGV;
	  new = Fcurrent_buffer ();
	  set_buffer_internal_1 (prev);
	  del_range_2 (from, from_byte, to, to_byte, 0);
	  TEMP_SET_PT_BOTH (from, from_byte);
	  insert_from_buffer (XBUFFER (new), 1, len, 0);
	  Fkill_buffer (new);
	  if (orig_point >= to)
	    orig_point += len - orig_len;
	  else if (orig_point > from)
	    orig_point = from;
	  orig_len = len;
	  to = from + len;
	  from_byte = CHAR_TO_BYTE (from);
	  to_byte = CHAR_TO_BYTE (to);
	  len_byte = to_byte - from_byte;
	  TEMP_SET_PT_BOTH (from, from_byte);
	}
    }

  if (replace)
    {
      if (! EQ (current_buffer->undo_list, Qt))
	deletion = make_buffer_string_both (from, from_byte, to, to_byte, 1);
      else
	{
	  nchars_del = to - from;
	  nbytes_del = to_byte - from_byte;
	}
    }

  if (coding->composing != COMPOSITION_DISABLED)
    {
      if (encodep)
	coding_save_composition (coding, from, to, Fcurrent_buffer ());
      else
	coding_allocate_composition_data (coding, from);
    }

  /* Try to skip the heading and tailing ASCIIs.  We can't skip them
     if we must run CCL program or there are compositions to
     encode.  */
  if (coding->type != coding_type_ccl
      && (! coding->cmp_data || coding->cmp_data->used == 0))
    {
      int from_byte_orig = from_byte, to_byte_orig = to_byte;

      if (from < GPT && GPT < to)
	move_gap_both (from, from_byte);
      SHRINK_CONVERSION_REGION (&from_byte, &to_byte, coding, NULL, encodep);
      if (from_byte == to_byte
	  && (encodep || NILP (coding->post_read_conversion))
	  && ! CODING_REQUIRE_FLUSHING (coding))
	{
	  coding->produced = len_byte;
	  coding->produced_char = len;
	  if (!replace)
	    /* We must record and adjust for this new text now.  */
	    adjust_after_insert (from, from_byte_orig, to, to_byte_orig, len);
	  coding_free_composition_data (coding);
	  return 0;
	}

      head_skip = from_byte - from_byte_orig;
      tail_skip = to_byte_orig - to_byte;
      total_skip = head_skip + tail_skip;
      from += head_skip;
      to -= tail_skip;
      len -= total_skip; len_byte -= total_skip;
    }

  /* For conversion, we must put the gap before the text in addition to
     making the gap larger for efficient decoding.  The required gap
     size starts from 2000 which is the magic number used in make_gap.
     But, after one batch of conversion, it will be incremented if we
     find that it is not enough .  */
  require = 2000;

  if (GAP_SIZE  < require)
    make_gap (require - GAP_SIZE);
  move_gap_both (from, from_byte);

  inserted = inserted_byte = 0;

  GAP_SIZE += len_byte;
  ZV -= len;
  Z -= len;
  ZV_BYTE -= len_byte;
  Z_BYTE -= len_byte;

  if (GPT - BEG < BEG_UNCHANGED)
    BEG_UNCHANGED = GPT - BEG;
  if (Z - GPT < END_UNCHANGED)
    END_UNCHANGED = Z - GPT;

  if (!encodep && coding->src_multibyte)
    {
      /* Decoding routines expects that the source text is unibyte.
	 We must convert 8-bit characters of multibyte form to
	 unibyte.  */
      int len_byte_orig = len_byte;
      len_byte = str_as_unibyte (GAP_END_ADDR - len_byte, len_byte);
      if (len_byte < len_byte_orig)
	safe_bcopy (GAP_END_ADDR - len_byte_orig, GAP_END_ADDR - len_byte,
		    len_byte);
      coding->src_multibyte = 0;
    }

  for (;;)
    {
      int result;

      /* The buffer memory is now:
	 +--------+converted-text+---------+-------original-text-------+---+
	 |<-from->|<--inserted-->|---------|<--------len_byte--------->|---|
		  |<---------------------- GAP ----------------------->|  */
      src = GAP_END_ADDR - len_byte;
      dst = GPT_ADDR + inserted_byte;

      if (encodep)
	result = encode_coding (coding, src, dst, len_byte, 0);
      else
	{
	  if (coding->composing != COMPOSITION_DISABLED)
	    coding->cmp_data->char_offset = from + inserted;
	  result = decode_coding (coding, src, dst, len_byte, 0);
	}

      /* The buffer memory is now:
	 +--------+-------converted-text----+--+------original-text----+---+
	 |<-from->|<-inserted->|<-produced->|--|<-(len_byte-consumed)->|---|
		  |<---------------------- GAP ----------------------->|  */

      inserted += coding->produced_char;
      inserted_byte += coding->produced;
      len_byte -= coding->consumed;

      if (result == CODING_FINISH_INSUFFICIENT_CMP)
	{
	  coding_allocate_composition_data (coding, from + inserted);
	  continue;
	}

      src += coding->consumed;
      dst += coding->produced;

      if (result == CODING_FINISH_NORMAL)
	{
	  src += len_byte;
	  break;
	}
      if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL)
	{
	  unsigned char *pend = dst, *p = pend - inserted_byte;
	  Lisp_Object eol_type;

	  /* Encode LFs back to the original eol format (CR or CRLF).  */
	  if (coding->eol_type == CODING_EOL_CR)
	    {
	      while (p < pend) if (*p++ == '\n') p[-1] = '\r';
	    }
	  else
	    {
	      int count = 0;

	      while (p < pend) if (*p++ == '\n') count++;
	      if (src - dst < count)
		{
		  /* We don't have sufficient room for encoding LFs
		     back to CRLF.  We must record converted and
		     not-yet-converted text back to the buffer
		     content, enlarge the gap, then record them out of
		     the buffer contents again.  */
		  int add = len_byte + inserted_byte;

		  GAP_SIZE -= add;
		  ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add;
		  GPT += inserted_byte; GPT_BYTE += inserted_byte;
		  make_gap (count - GAP_SIZE);
		  GAP_SIZE += add;
		  ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add;
		  GPT -= inserted_byte; GPT_BYTE -= inserted_byte;
		  /* Don't forget to update SRC, DST, and PEND.  */
		  src = GAP_END_ADDR - len_byte;
		  dst = GPT_ADDR + inserted_byte;
		  pend = dst;
		}
	      inserted += count;
	      inserted_byte += count;
	      coding->produced += count;
	      p = dst = pend + count;
	      while (count)
		{
		  *--p = *--pend;
		  if (*p == '\n') count--, *--p = '\r';
		}
	    }

	  /* Suppress eol-format conversion in the further conversion.  */
	  coding->eol_type = CODING_EOL_LF;

	  /* Set the coding system symbol to that for Unix-like EOL.  */
	  eol_type = Fget (saved_coding_symbol, Qeol_type);
	  if (VECTORP (eol_type)
	      && XVECTOR (eol_type)->size == 3
	      && SYMBOLP (XVECTOR (eol_type)->contents[CODING_EOL_LF]))
	    coding->symbol = XVECTOR (eol_type)->contents[CODING_EOL_LF];
	  else
	    coding->symbol = saved_coding_symbol;

	  continue;
	}
      if (len_byte <= 0)
	{
	  if (coding->type != coding_type_ccl
	      || coding->mode & CODING_MODE_LAST_BLOCK)
	    break;
	  coding->mode |= CODING_MODE_LAST_BLOCK;
	  continue;
	}
      if (result == CODING_FINISH_INSUFFICIENT_SRC)
	{
	  /* The source text ends in invalid codes.  Let's just
	     make them valid buffer contents, and finish conversion.  */
	  if (multibyte_p)
	    {
	      unsigned char *start = dst;

	      inserted += len_byte;
	      while (len_byte--)
		{
		  int c = *src++;
		  dst += CHAR_STRING (c, dst);
		}

	      inserted_byte += dst - start;
	    }
	  else
	    {
	      inserted += len_byte;
	      inserted_byte += len_byte;
	      while (len_byte--)
		*dst++ = *src++;
	    }
	  break;
	}
      if (result == CODING_FINISH_INTERRUPT)
	{
	  /* The conversion procedure was interrupted by a user.  */
	  break;
	}
      /* Now RESULT == CODING_FINISH_INSUFFICIENT_DST  */
      if (coding->consumed < 1)
	{
	  /* It's quite strange to require more memory without
	     consuming any bytes.  Perhaps CCL program bug.  */
	  break;
	}
      if (first)
	{
	  /* We have just done the first batch of conversion which was
	     stopped because of insufficient gap.  Let's reconsider the
	     required gap size (i.e. SRT - DST) now.

	     We have converted ORIG bytes (== coding->consumed) into
	     NEW bytes (coding->produced).  To convert the remaining
	     LEN bytes, we may need REQUIRE bytes of gap, where:
		REQUIRE + LEN_BYTE = LEN_BYTE * (NEW / ORIG)
		REQUIRE = LEN_BYTE * (NEW - ORIG) / ORIG
	     Here, we are sure that NEW >= ORIG.  */

	  if (coding->produced <= coding->consumed)
	    {
	      /* This happens because of CCL-based coding system with
		 eol-type CRLF.  */
	      require = 0;
	    }
	  else
	    {
	      float ratio = coding->produced - coding->consumed;
	      ratio /= coding->consumed;
	      require = len_byte * ratio;
	    }
	  first = 0;
	}
      if ((src - dst) < (require + 2000))
	{
	  /* See the comment above the previous call of make_gap.  */
	  int add = len_byte + inserted_byte;

	  GAP_SIZE -= add;
	  ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add;
	  GPT += inserted_byte; GPT_BYTE += inserted_byte;
	  make_gap (require + 2000);
	  GAP_SIZE += add;
	  ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add;
	  GPT -= inserted_byte; GPT_BYTE -= inserted_byte;
	}
    }
  if (src - dst > 0) *dst = 0; /* Put an anchor.  */

  if (encodep && coding->dst_multibyte)
    {
      /* The output is unibyte.  We must convert 8-bit characters to
	 multibyte form.  */
      if (inserted_byte * 2 > GAP_SIZE)
	{
	  GAP_SIZE -= inserted_byte;
	  ZV += inserted_byte; Z += inserted_byte;
	  ZV_BYTE += inserted_byte; Z_BYTE += inserted_byte;
	  GPT += inserted_byte; GPT_BYTE += inserted_byte;
	  make_gap (inserted_byte - GAP_SIZE);
	  GAP_SIZE += inserted_byte;
	  ZV -= inserted_byte; Z -= inserted_byte;
	  ZV_BYTE -= inserted_byte; Z_BYTE -= inserted_byte;
	  GPT -= inserted_byte; GPT_BYTE -= inserted_byte;
	}
      inserted_byte = str_to_multibyte (GPT_ADDR, GAP_SIZE, inserted_byte);
    }

  /* If we shrank the conversion area, adjust it now.  */
  if (total_skip > 0)
    {
      if (tail_skip > 0)
	safe_bcopy (GAP_END_ADDR, GPT_ADDR + inserted_byte, tail_skip);
      inserted += total_skip; inserted_byte += total_skip;
      GAP_SIZE += total_skip;
      GPT -= head_skip; GPT_BYTE -= head_skip;
      ZV -= total_skip; ZV_BYTE -= total_skip;
      Z -= total_skip; Z_BYTE -= total_skip;
      from -= head_skip; from_byte -= head_skip;
      to += tail_skip; to_byte += tail_skip;
    }

  prev_Z = Z;
  if (! EQ (current_buffer->undo_list, Qt))
    adjust_after_replace (from, from_byte, deletion, inserted, inserted_byte);
  else
    adjust_after_replace_noundo (from, from_byte, nchars_del, nbytes_del,
				 inserted, inserted_byte);
  inserted = Z - prev_Z;

  if (!encodep && coding->cmp_data && coding->cmp_data->used)
    coding_restore_composition (coding, Fcurrent_buffer ());
  coding_free_composition_data (coding);

  if (! inhibit_pre_post_conversion
      && ! encodep && ! NILP (coding->post_read_conversion))
    {
      Lisp_Object val;
      Lisp_Object saved_coding_system;

      if (from != PT)
	TEMP_SET_PT_BOTH (from, from_byte);
      prev_Z = Z;
      record_unwind_protect (code_convert_region_unwind,
			     Fcons (Vlast_coding_system_used, Qnil));
      saved_coding_system = Vlast_coding_system_used;
      Vlast_coding_system_used = coding->symbol;
      /* We should not call any more pre-write/post-read-conversion
         functions while this post-read-conversion is running.  */
      inhibit_pre_post_conversion = 1;
      val = call1 (coding->post_read_conversion, make_number (inserted));
      inhibit_pre_post_conversion = 0;
      coding->symbol = Vlast_coding_system_used;
      Vlast_coding_system_used = saved_coding_system;
      /* Discard the unwind protect.  */
      specpdl_ptr--;
      CHECK_NUMBER (val);
      inserted += Z - prev_Z;
    }

  if (orig_point >= from)
    {
      if (orig_point >= from + orig_len)
	orig_point += inserted - orig_len;
      else
	orig_point = from;
      TEMP_SET_PT (orig_point);
    }

  if (replace)
    {
      signal_after_change (from, to - from, inserted);
      update_compositions (from, from + inserted, CHECK_BORDER);
    }

  {
    coding->consumed = to_byte - from_byte;
    coding->consumed_char = to - from;
    coding->produced = inserted_byte;
    coding->produced_char = inserted;
  }

  return 0;
}

/* Name (or base name) of work buffer for code conversion.  */
static Lisp_Object Vcode_conversion_workbuf_name;

/* Set the current buffer to the working buffer prepared for
   code-conversion.  MULTIBYTE specifies the multibyteness of the
   buffer.  Return the buffer we set if it must be killed after use.
   Otherwise return Qnil.  */

static Lisp_Object
set_conversion_work_buffer (multibyte)
     int multibyte;
{
  Lisp_Object buffer, buffer_to_kill;
  struct buffer *buf;

  buffer = Fget_buffer_create (Vcode_conversion_workbuf_name);
  buf = XBUFFER (buffer);
  if (buf == current_buffer)
    {
      /* As we are already in the work buffer, we must generate a new
	 buffer for the work.  */
      Lisp_Object name;

      name = Fgenerate_new_buffer_name (Vcode_conversion_workbuf_name, Qnil);
      buffer = buffer_to_kill = Fget_buffer_create (name);
      buf = XBUFFER (buffer);
    }
  else
    buffer_to_kill = Qnil;

  delete_all_overlays (buf);
  buf->directory = current_buffer->directory;
  buf->read_only = Qnil;
  buf->filename = Qnil;
  buf->undo_list = Qt;
  eassert (buf->overlays_before == NULL);
  eassert (buf->overlays_after == NULL);
  set_buffer_internal (buf);
  if (BEG != BEGV || Z != ZV)
    Fwiden ();
  del_range_2 (BEG, BEG_BYTE, Z, Z_BYTE, 0);
  buf->enable_multibyte_characters = multibyte ? Qt : Qnil;
  return buffer_to_kill;
}

Lisp_Object
run_pre_post_conversion_on_str (str, coding, encodep)
     Lisp_Object str;
     struct coding_system *coding;
     int encodep;
{
  int count = SPECPDL_INDEX ();
  struct gcpro gcpro1, gcpro2;
  int multibyte = STRING_MULTIBYTE (str);
  Lisp_Object old_deactivate_mark;
  Lisp_Object buffer_to_kill;
  Lisp_Object unwind_arg;

  record_unwind_protect (Fset_buffer, Fcurrent_buffer ());
  /* It is not crucial to specbind this.  */
  old_deactivate_mark = Vdeactivate_mark;
  GCPRO2 (str, old_deactivate_mark);

  /* We must insert the contents of STR as is without
     unibyte<->multibyte conversion.  For that, we adjust the
     multibyteness of the working buffer to that of STR.  */
  buffer_to_kill = set_conversion_work_buffer (multibyte);
  if (NILP (buffer_to_kill))
    unwind_arg = Fcons (Vlast_coding_system_used, Qnil);
  else
    unwind_arg = list2 (Vlast_coding_system_used, buffer_to_kill);
  record_unwind_protect (code_convert_region_unwind, unwind_arg);

  insert_from_string (str, 0, 0,
		      SCHARS (str), SBYTES (str), 0);
  UNGCPRO;
  inhibit_pre_post_conversion = 1;
  if (encodep)
    {
      struct buffer *prev = current_buffer;

      call2 (coding->pre_write_conversion, make_number (BEG), make_number (Z));
      if (prev != current_buffer)
	/* We must kill the current buffer too.  */
	Fsetcdr (unwind_arg, Fcons (Fcurrent_buffer (), XCDR (unwind_arg)));
    }
  else
    {
      Vlast_coding_system_used = coding->symbol;
      TEMP_SET_PT_BOTH (BEG, BEG_BYTE);
      call1 (coding->post_read_conversion, make_number (Z - BEG));
      coding->symbol = Vlast_coding_system_used;
    }
  inhibit_pre_post_conversion = 0;
  Vdeactivate_mark = old_deactivate_mark;
  str = make_buffer_string (BEG, Z, 1);
  return unbind_to (count, str);
}


/* Run pre-write-conversion function of CODING on NCHARS/NBYTES
   text in *STR.  *SIZE is the allocated bytes for STR.  As it
   is intended that this function is called from encode_terminal_code,
   the pre-write-conversion function is run by safe_call and thus
   "Error during redisplay: ..." is logged when an error occurs.

   Store the resulting text in *STR and set CODING->produced_char and
   CODING->produced to the number of characters and bytes
   respectively.  If the size of *STR is too small, enlarge it by
   xrealloc and update *STR and *SIZE.  */

void
run_pre_write_conversin_on_c_str (str, size, nchars, nbytes, coding)
     unsigned char **str;
     int *size, nchars, nbytes;
     struct coding_system *coding;
{
  struct gcpro gcpro1, gcpro2;
  struct buffer *cur = current_buffer;
  struct buffer *prev;
  Lisp_Object old_deactivate_mark, old_last_coding_system_used;
  Lisp_Object args[3];
  Lisp_Object buffer_to_kill;

  /* It is not crucial to specbind this.  */
  old_deactivate_mark = Vdeactivate_mark;
  old_last_coding_system_used = Vlast_coding_system_used;
  GCPRO2 (old_deactivate_mark, old_last_coding_system_used);

  /* We must insert the contents of STR as is without
     unibyte<->multibyte conversion.  For that, we adjust the
     multibyteness of the working buffer to that of STR.  */
  buffer_to_kill = set_conversion_work_buffer (coding->src_multibyte);
  insert_1_both (*str, nchars, nbytes, 0, 0, 0);
  UNGCPRO;
  inhibit_pre_post_conversion = 1;
  prev = current_buffer;
  args[0] = coding->pre_write_conversion;
  args[1] = make_number (BEG);
  args[2] = make_number (Z);
  safe_call (3, args);
  inhibit_pre_post_conversion = 0;
  Vdeactivate_mark = old_deactivate_mark;
  Vlast_coding_system_used = old_last_coding_system_used;
  coding->produced_char = Z - BEG;
  coding->produced = Z_BYTE - BEG_BYTE;
  if (coding->produced > *size)
    {
      *size = coding->produced;
      *str = xrealloc (*str, *size);
    }
  if (BEG < GPT && GPT < Z)
    move_gap (BEG);
  bcopy (BEG_ADDR, *str, coding->produced);
  coding->src_multibyte
    = ! NILP (current_buffer->enable_multibyte_characters);
  if (prev != current_buffer)
    Fkill_buffer (Fcurrent_buffer ());
  set_buffer_internal (cur);
  if (! NILP (buffer_to_kill))
    Fkill_buffer (buffer_to_kill);
}


Lisp_Object
decode_coding_string (str, coding, nocopy)
     Lisp_Object str;
     struct coding_system *coding;
     int nocopy;
{
  int len;
  struct conversion_buffer buf;
  int from, to_byte;
  Lisp_Object saved_coding_symbol;
  int result;
  int require_decoding;
  int shrinked_bytes = 0;
  Lisp_Object newstr;
  int consumed, consumed_char, produced, produced_char;

  from = 0;
  to_byte = SBYTES (str);

  saved_coding_symbol = coding->symbol;
  coding->src_multibyte = STRING_MULTIBYTE (str);
  coding->dst_multibyte = 1;
  coding->heading_ascii = 0;

  if (CODING_REQUIRE_DETECTION (coding))
    {
      /* See the comments in code_convert_region.  */
      if (coding->type == coding_type_undecided)
	{
	  detect_coding (coding, SDATA (str), to_byte);
	  if (coding->type == coding_type_undecided)
	    {
	      coding->type = coding_type_emacs_mule;
	      coding->category_idx = CODING_CATEGORY_IDX_EMACS_MULE;
	      /* As emacs-mule decoder will handle composition, we
		 need this setting to allocate coding->cmp_data
		 later.  */
	      coding->composing = COMPOSITION_NO;
	    }
	}
      if (coding->eol_type == CODING_EOL_UNDECIDED
	  && coding->type != coding_type_ccl)
	{
	  saved_coding_symbol = coding->symbol;
	  detect_eol (coding, SDATA (str), to_byte);
	  if (coding->eol_type == CODING_EOL_UNDECIDED)
	    coding->eol_type = CODING_EOL_LF;
	  /* We had better recover the original eol format if we
	     encounter an inconsistent eol format while decoding.  */
	  coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL;
	}
    }

  if (coding->type == coding_type_no_conversion
      || coding->type == coding_type_raw_text)
    coding->dst_multibyte = 0;

  require_decoding = CODING_REQUIRE_DECODING (coding);

  if (STRING_MULTIBYTE (str))
    {
      /* Decoding routines expect the source text to be unibyte.  */
      str = Fstring_as_unibyte (str);
      to_byte = SBYTES (str);
      nocopy = 1;
      coding->src_multibyte = 0;
    }

  /* Try to skip the heading and tailing ASCIIs.  */
  if (require_decoding && coding->type != coding_type_ccl)
    {
      SHRINK_CONVERSION_REGION (&from, &to_byte, coding, SDATA (str),
				0);
      if (from == to_byte)
	require_decoding = 0;
      shrinked_bytes = from + (SBYTES (str) - to_byte);
    }

  if (!require_decoding
      && !(SYMBOLP (coding->post_read_conversion)
	   && !NILP (Ffboundp (coding->post_read_conversion))))
    {
      coding->consumed = SBYTES (str);
      coding->consumed_char = SCHARS (str);
      if (coding->dst_multibyte)
	{
	  str = Fstring_as_multibyte (str);
	  nocopy = 1;
	}
      coding->produced = SBYTES (str);
      coding->produced_char = SCHARS (str);
      return (nocopy ? str : Fcopy_sequence (str));
    }

  if (coding->composing != COMPOSITION_DISABLED)
    coding_allocate_composition_data (coding, from);
  len = decoding_buffer_size (coding, to_byte - from);
  allocate_conversion_buffer (buf, len);

  consumed = consumed_char = produced = produced_char = 0;
  while (1)
    {
      result = decode_coding (coding, SDATA (str) + from + consumed,
			      buf.data + produced, to_byte - from - consumed,
			      buf.size - produced);
      consumed += coding->consumed;
      consumed_char += coding->consumed_char;
      produced += coding->produced;
      produced_char += coding->produced_char;
      if (result == CODING_FINISH_NORMAL
	  || result == CODING_FINISH_INTERRUPT
	  || (result == CODING_FINISH_INSUFFICIENT_SRC
	      && coding->consumed == 0))
	break;
      if (result == CODING_FINISH_INSUFFICIENT_CMP)
	coding_allocate_composition_data (coding, from + produced_char);
      else if (result == CODING_FINISH_INSUFFICIENT_DST)
	extend_conversion_buffer (&buf);
      else if (result == CODING_FINISH_INCONSISTENT_EOL)
	{
	  Lisp_Object eol_type;

	  /* Recover the original EOL format.  */
	  if (coding->eol_type == CODING_EOL_CR)
	    {
	      unsigned char *p;
	      for (p = buf.data; p < buf.data + produced; p++)
		if (*p == '\n') *p = '\r';
	    }
	  else if (coding->eol_type == CODING_EOL_CRLF)
	    {
	      int num_eol = 0;
	      unsigned char *p0, *p1;
	      for (p0 = buf.data, p1 = p0 + produced; p0 < p1; p0++)
		if (*p0 == '\n') num_eol++;
	      if (produced + num_eol >= buf.size)
		extend_conversion_buffer (&buf);
	      for (p0 = buf.data + produced, p1 = p0 + num_eol; p0 > buf.data;)
		{
		  *--p1 = *--p0;
		  if (*p0 == '\n') *--p1 = '\r';
		}
	      produced += num_eol;
	      produced_char += num_eol;
	    }
	  /* Suppress eol-format conversion in the further conversion.  */
	  coding->eol_type = CODING_EOL_LF;

	  /* Set the coding system symbol to that for Unix-like EOL.  */
	  eol_type = Fget (saved_coding_symbol, Qeol_type);
	  if (VECTORP (eol_type)
	      && XVECTOR (eol_type)->size == 3
	      && SYMBOLP (XVECTOR (eol_type)->contents[CODING_EOL_LF]))
	    coding->symbol = XVECTOR (eol_type)->contents[CODING_EOL_LF];
	  else
	    coding->symbol = saved_coding_symbol;


	}
    }

  coding->consumed = consumed;
  coding->consumed_char = consumed_char;
  coding->produced = produced;
  coding->produced_char = produced_char;

  if (coding->dst_multibyte)
    newstr = make_uninit_multibyte_string (produced_char + shrinked_bytes,
					   produced + shrinked_bytes);
  else
    newstr = make_uninit_string (produced + shrinked_bytes);
  if (from > 0)
    STRING_COPYIN (newstr, 0, SDATA (str), from);
  STRING_COPYIN (newstr, from, buf.data, produced);
  if (shrinked_bytes > from)
    STRING_COPYIN (newstr, from + produced,
		   SDATA (str) + to_byte,
		   shrinked_bytes - from);
  free_conversion_buffer (&buf);

  coding->consumed += shrinked_bytes;
  coding->consumed_char += shrinked_bytes;
  coding->produced += shrinked_bytes;
  coding->produced_char += shrinked_bytes;

  if (coding->cmp_data && coding->cmp_data->used)
    coding_restore_composition (coding, newstr);
  coding_free_composition_data (coding);

  if (SYMBOLP (coding->post_read_conversion)
      && !NILP (Ffboundp (coding->post_read_conversion)))
    newstr = run_pre_post_conversion_on_str (newstr, coding, 0);

  return newstr;
}

Lisp_Object
encode_coding_string (str, coding, nocopy)
     Lisp_Object str;
     struct coding_system *coding;
     int nocopy;
{
  int len;
  struct conversion_buffer buf;
  int from, to, to_byte;
  int result;
  int shrinked_bytes = 0;
  Lisp_Object newstr;
  int consumed, consumed_char, produced, produced_char;

  if (SYMBOLP (coding->pre_write_conversion)
      && !NILP (Ffboundp (coding->pre_write_conversion)))
    {
      str = run_pre_post_conversion_on_str (str, coding, 1);
      /* As STR is just newly generated, we don't have to copy it
	 anymore.  */
      nocopy = 1;
    }

  from = 0;
  to = SCHARS (str);
  to_byte = SBYTES (str);

  /* Encoding routines determine the multibyteness of the source text
     by coding->src_multibyte.  */
  coding->src_multibyte = SCHARS (str) < SBYTES (str);
  coding->dst_multibyte = 0;
  if (! CODING_REQUIRE_ENCODING (coding))
    goto no_need_of_encoding;

  if (coding->composing != COMPOSITION_DISABLED)
    coding_save_composition (coding, from, to, str);

  /* Try to skip the heading and tailing ASCIIs.  We can't skip them
     if we must run CCL program or there are compositions to
     encode.  */
  coding->heading_ascii = 0;
  if (coding->type != coding_type_ccl
      && (! coding->cmp_data || coding->cmp_data->used == 0))
    {
      SHRINK_CONVERSION_REGION (&from, &to_byte, coding, SDATA (str),
				1);
      if (from == to_byte)
	{
	  coding_free_composition_data (coding);
	  goto no_need_of_encoding;
	}
      shrinked_bytes = from + (SBYTES (str) - to_byte);
    }

  len = encoding_buffer_size (coding, to_byte - from);
  allocate_conversion_buffer (buf, len);

  consumed = consumed_char = produced = produced_char = 0;
  while (1)
    {
      result = encode_coding (coding, SDATA (str) + from + consumed,
			      buf.data + produced, to_byte - from - consumed,
			      buf.size - produced);
      consumed += coding->consumed;
      consumed_char += coding->consumed_char;
      produced += coding->produced;
      produced_char += coding->produced_char;
      if (result == CODING_FINISH_NORMAL
	  || result == CODING_FINISH_INTERRUPT
	  || (result == CODING_FINISH_INSUFFICIENT_SRC
	      && coding->consumed == 0))
	break;
      /* Now result should be CODING_FINISH_INSUFFICIENT_DST.  */
      extend_conversion_buffer (&buf);
    }

  coding->consumed = consumed;
  coding->consumed_char = consumed_char;
  coding->produced = produced;
  coding->produced_char = produced_char;

  newstr = make_uninit_string (produced + shrinked_bytes);
  if (from > 0)
    STRING_COPYIN (newstr, 0, SDATA (str), from);
  STRING_COPYIN (newstr, from, buf.data, produced);
  if (shrinked_bytes > from)
    STRING_COPYIN (newstr, from + produced,
		   SDATA (str) + to_byte,
		   shrinked_bytes - from);

  free_conversion_buffer (&buf);
  coding_free_composition_data (coding);

  return newstr;

 no_need_of_encoding:
  coding->consumed = SBYTES (str);
  coding->consumed_char = SCHARS (str);
  if (STRING_MULTIBYTE (str))
    {
      if (nocopy)
	/* We are sure that STR doesn't contain a multibyte
	   character.  */
	STRING_SET_UNIBYTE (str);
      else
	{
	  str = Fstring_as_unibyte (str);
	  nocopy = 1;
	}
    }
  coding->produced = SBYTES (str);
  coding->produced_char = SCHARS (str);
  return (nocopy ? str : Fcopy_sequence (str));
}


#ifdef emacs
/*** 8. Emacs Lisp library functions ***/

DEFUN ("coding-system-p", Fcoding_system_p, Scoding_system_p, 1, 1, 0,
       doc: /* Return t if OBJECT is nil or a coding-system.
See the documentation of `make-coding-system' for information
about coding-system objects.  */)
     (obj)
     Lisp_Object obj;
{
  if (NILP (obj))
    return Qt;
  if (!SYMBOLP (obj))
    return Qnil;
  if (! NILP (Fget (obj, Qcoding_system_define_form)))
    return Qt;
  /* Get coding-spec vector for OBJ.  */
  obj = Fget (obj, Qcoding_system);
  return ((VECTORP (obj) && XVECTOR (obj)->size == 5)
	  ? Qt : Qnil);
}

DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system,
       Sread_non_nil_coding_system, 1, 1, 0,
       doc: /* Read a coding system from the minibuffer, prompting with string PROMPT.  */)
     (prompt)
     Lisp_Object prompt;
{
  Lisp_Object val;
  do
    {
      val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil,
			      Qt, Qnil, Qcoding_system_history, Qnil, Qnil);
    }
  while (SCHARS (val) == 0);
  return (Fintern (val, Qnil));
}

DEFUN ("read-coding-system", Fread_coding_system, Sread_coding_system, 1, 2, 0,
       doc: /* Read a coding system from the minibuffer, prompting with string PROMPT.
If the user enters null input, return second argument DEFAULT-CODING-SYSTEM.  */)
     (prompt, default_coding_system)
     Lisp_Object prompt, default_coding_system;
{
  Lisp_Object val;
  if (SYMBOLP (default_coding_system))
    default_coding_system = SYMBOL_NAME (default_coding_system);
  val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil,
			  Qt, Qnil, Qcoding_system_history,
			  default_coding_system, Qnil);
  return (SCHARS (val) == 0 ? Qnil : Fintern (val, Qnil));
}

DEFUN ("check-coding-system", Fcheck_coding_system, Scheck_coding_system,
       1, 1, 0,
       doc: /* Check validity of CODING-SYSTEM.
If valid, return CODING-SYSTEM, else signal a `coding-system-error' error.
It is valid if it is nil or a symbol with a non-nil `coding-system' property.
The value of this property should be a vector of length 5.  */)
     (coding_system)
     Lisp_Object coding_system;
{
  Lisp_Object define_form;

  define_form = Fget (coding_system, Qcoding_system_define_form);
  if (! NILP (define_form))
    {
      Fput (coding_system, Qcoding_system_define_form, Qnil);
      safe_eval (define_form);
    }
  if (!NILP (Fcoding_system_p (coding_system)))
    return coding_system;
  xsignal1 (Qcoding_system_error, coding_system);
}

Lisp_Object
detect_coding_system (src, src_bytes, highest, multibytep)
     const unsigned char *src;
     int src_bytes, highest;
     int multibytep;
{
  int coding_mask, eol_type;
  Lisp_Object val, tmp;
  int dummy;

  coding_mask = detect_coding_mask (src, src_bytes, NULL, &dummy, multibytep);
  eol_type  = detect_eol_type (src, src_bytes, &dummy);
  if (eol_type == CODING_EOL_INCONSISTENT)
    eol_type = CODING_EOL_UNDECIDED;

  if (!coding_mask)
    {
      val = Qundecided;
      if (eol_type != CODING_EOL_UNDECIDED)
	{
	  Lisp_Object val2;
	  val2 = Fget (Qundecided, Qeol_type);
	  if (VECTORP (val2))
	    val = XVECTOR (val2)->contents[eol_type];
	}
      return (highest ? val : Fcons (val, Qnil));
    }

  /* At first, gather possible coding systems in VAL.  */
  val = Qnil;
  for (tmp = Vcoding_category_list; CONSP (tmp); tmp = XCDR (tmp))
    {
      Lisp_Object category_val, category_index;

      category_index = Fget (XCAR (tmp), Qcoding_category_index);
      category_val = Fsymbol_value (XCAR (tmp));
      if (!NILP (category_val)
	  && NATNUMP (category_index)
	  && (coding_mask & (1 << XFASTINT (category_index))))
	{
	  val = Fcons (category_val, val);
	  if (highest)
	    break;
	}
    }
  if (!highest)
    val = Fnreverse (val);

  /* Then, replace the elements with subsidiary coding systems.  */
  for (tmp = val; CONSP (tmp); tmp = XCDR (tmp))
    {
      if (eol_type != CODING_EOL_UNDECIDED
	  && eol_type != CODING_EOL_INCONSISTENT)
	{
	  Lisp_Object eol;
	  eol = Fget (XCAR (tmp), Qeol_type);
	  if (VECTORP (eol))
	    XSETCAR (tmp, XVECTOR (eol)->contents[eol_type]);
	}
    }
  return (highest ? XCAR (val) : val);
}

DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region,
       2, 3, 0,
       doc: /* Detect how the byte sequence in the region is encoded.
Return a list of possible coding systems used on decoding a byte
sequence containing the bytes in the region between START and END when
the coding system `undecided' is specified.  The list is ordered by
priority decided in the current language environment.

If only ASCII characters are found (except for such ISO-2022 control
characters ISO-2022 as ESC), it returns a list of single element
`undecided' or its subsidiary coding system according to a detected
end-of-line format.

If optional argument HIGHEST is non-nil, return the coding system of
highest priority.  */)
     (start, end, highest)
     Lisp_Object start, end, highest;
{
  int from, to;
  int from_byte, to_byte;
  int include_anchor_byte = 0;

  CHECK_NUMBER_COERCE_MARKER (start);
  CHECK_NUMBER_COERCE_MARKER (end);

  validate_region (&start, &end);
  from = XINT (start), to = XINT (end);
  from_byte = CHAR_TO_BYTE (from);
  to_byte = CHAR_TO_BYTE (to);

  if (from < GPT && to >= GPT)
    move_gap_both (to, to_byte);
  /* If we an anchor byte `\0' follows the region, we include it in
     the detecting source.  Then code detectors can handle the tailing
     byte sequence more accurately.

     Fix me: This is not a perfect solution.  It is better that we
     add one more argument, say LAST_BLOCK, to all detect_coding_XXX.
  */
  if (to == Z || (to == GPT && GAP_SIZE > 0))
    include_anchor_byte = 1;
  return detect_coding_system (BYTE_POS_ADDR (from_byte),
			       to_byte - from_byte + include_anchor_byte,
			       !NILP (highest),
			       !NILP (current_buffer
				      ->enable_multibyte_characters));
}

DEFUN ("detect-coding-string", Fdetect_coding_string, Sdetect_coding_string,
       1, 2, 0,
       doc: /* Detect how the byte sequence in STRING is encoded.
Return a list of possible coding systems used on decoding a byte
sequence containing the bytes in STRING when the coding system
`undecided' is specified.  The list is ordered by priority decided in
the current language environment.

If only ASCII characters are found (except for such ISO-2022 control
characters ISO-2022 as ESC), it returns a list of single element
`undecided' or its subsidiary coding system according to a detected
end-of-line format.

If optional argument HIGHEST is non-nil, return the coding system of
highest priority.  */)
     (string, highest)
     Lisp_Object string, highest;
{
  CHECK_STRING (string);

  return detect_coding_system (SDATA (string),
			       /* "+ 1" is to include the anchor byte
				  `\0'.  With this, code detectors can
				  handle the tailing bytes more
				  accurately.  */
			       SBYTES (string) + 1,
			       !NILP (highest),
			       STRING_MULTIBYTE (string));
}

/*  Subroutine for Ffind_coding_systems_region_internal.

    Return a list of coding systems that safely encode the multibyte
    text between P and PEND.  SAFE_CODINGS, if non-nil, is an alist of
    possible coding systems.  If it is nil, it means that we have not
    yet found any coding systems.

    WORK_TABLE a char-table of which element is set to t once the
    element is looked up.

    If a non-ASCII single byte char is found, set
    *single_byte_char_found to 1.  */

static Lisp_Object
find_safe_codings (p, pend, safe_codings, work_table, single_byte_char_found)
     unsigned char *p, *pend;
     Lisp_Object safe_codings, work_table;
     int *single_byte_char_found;
{
  int c, len;
  Lisp_Object val, ch;
  Lisp_Object prev, tail;

  if (NILP (safe_codings))
    goto done_safe_codings;
  while (p < pend)
    {
      c = STRING_CHAR_AND_LENGTH (p, pend - p, len);
      p += len;
      if (ASCII_BYTE_P (c))
	/* We can ignore ASCII characters here.  */
	continue;
      if (SINGLE_BYTE_CHAR_P (c))
	*single_byte_char_found = 1;
      /* Check the safe coding systems for C.  */
      ch = make_number (c);
      val = Faref (work_table, ch);
      if (EQ (val, Qt))
	/* This element was already checked.  Ignore it.  */
	continue;
      /* Remember that we checked this element.  */
      Faset (work_table, ch, Qt);

      for (prev = tail = safe_codings; CONSP (tail); tail = XCDR (tail))
	{
	  Lisp_Object elt, translation_table, hash_table, accept_latin_extra;
	  int encodable;

	  elt = XCAR (tail);
	  if (CONSP (XCDR (elt)))
	    {
	      /* This entry has this format now:
		 ( CODING SAFE-CHARS TRANSLATION-TABLE HASH-TABLE
		          ACCEPT-LATIN-EXTRA ) */
	      val = XCDR (elt);
	      encodable = ! NILP (Faref (XCAR (val), ch));
	      if (! encodable)
		{
		  val = XCDR (val);
		  translation_table = XCAR (val);
		  hash_table = XCAR (XCDR (val));
		  accept_latin_extra = XCAR (XCDR (XCDR (val)));
		}
	    }
	  else
	    {
	      /* This entry has this format now: ( CODING . SAFE-CHARS) */
	      encodable = ! NILP (Faref (XCDR (elt), ch));
	      if (! encodable)
		{
		  /* Transform the format to:
		     ( CODING SAFE-CHARS TRANSLATION-TABLE HASH-TABLE
		       ACCEPT-LATIN-EXTRA )  */
		  val = Fget (XCAR (elt), Qcoding_system);
		  translation_table
		    = Fplist_get (AREF (val, 3),
				  Qtranslation_table_for_encode);
		  if (SYMBOLP (translation_table))
		    translation_table = Fget (translation_table,
					      Qtranslation_table);
		  hash_table
		    = (CHAR_TABLE_P (translation_table)
		       ? XCHAR_TABLE (translation_table)->extras[1]
		       : Qnil);
		  accept_latin_extra
		    = ((EQ (AREF (val, 0), make_number (2))
			&& VECTORP (AREF (val, 4)))
		       ? AREF (AREF (val, 4), 16)
		       : Qnil);
		  XSETCAR (tail, list5 (XCAR (elt), XCDR (elt),
					translation_table, hash_table,
					accept_latin_extra));
		}
	    }

	  if (! encodable
	      && ((CHAR_TABLE_P (translation_table)
		   && ! NILP (Faref (translation_table, ch)))
		  || (HASH_TABLE_P (hash_table)
		      && ! NILP (Fgethash (ch, hash_table, Qnil)))
		  || (SINGLE_BYTE_CHAR_P (c)
		      && ! NILP (accept_latin_extra)
		      && VECTORP (Vlatin_extra_code_table)
		      && ! NILP (AREF (Vlatin_extra_code_table, c)))))
	    encodable = 1;
	  if (encodable)
	    prev = tail;
	  else
	    {
	      /* Exclude this coding system from SAFE_CODINGS.  */
	      if (EQ (tail, safe_codings))
		{
		  safe_codings = XCDR (safe_codings);
		  if (NILP (safe_codings))
		    goto done_safe_codings;
		}
	      else
		XSETCDR (prev, XCDR (tail));
	    }
	}
    }

 done_safe_codings:
  /* If the above loop was terminated before P reaches PEND, it means
     SAFE_CODINGS was set to nil.  If we have not yet found an
     non-ASCII single-byte char, check it now.  */
  if (! *single_byte_char_found)
    while (p < pend)
      {
	c = STRING_CHAR_AND_LENGTH (p, pend - p, len);
	p += len;
	if (! ASCII_BYTE_P (c)
	    && SINGLE_BYTE_CHAR_P (c))
	  {
	    *single_byte_char_found = 1;
	    break;
	  }
      }
  return safe_codings;
}

DEFUN ("find-coding-systems-region-internal",
       Ffind_coding_systems_region_internal,
       Sfind_coding_systems_region_internal, 2, 2, 0,
       doc: /* Internal use only.  */)
     (start, end)
     Lisp_Object start, end;
{
  Lisp_Object work_table, safe_codings;
  int non_ascii_p = 0;
  int single_byte_char_found = 0;
  const unsigned char *p1, *p1end, *p2, *p2end, *p;

  if (STRINGP (start))
    {
      if (!STRING_MULTIBYTE (start))
	return Qt;
      p1 = SDATA (start), p1end = p1 + SBYTES (start);
      p2 = p2end = p1end;
      if (SCHARS (start) != SBYTES (start))
	non_ascii_p = 1;
    }
  else
    {
      int from, to, stop;

      CHECK_NUMBER_COERCE_MARKER (start);
      CHECK_NUMBER_COERCE_MARKER (end);
      if (XINT (start) < BEG || XINT (end) > Z || XINT (start) > XINT (end))
	args_out_of_range (start, end);
      if (NILP (current_buffer->enable_multibyte_characters))
	return Qt;
      from = CHAR_TO_BYTE (XINT (start));
      to = CHAR_TO_BYTE (XINT (end));
      stop = from < GPT_BYTE && GPT_BYTE < to ? GPT_BYTE : to;
      p1 = BYTE_POS_ADDR (from), p1end = p1 + (stop - from);
      if (stop == to)
	p2 = p2end = p1end;
      else
	p2 = BYTE_POS_ADDR (stop), p2end = p2 + (to - stop);
      if (XINT (end) - XINT (start) != to - from)
	non_ascii_p = 1;
    }

  if (!non_ascii_p)
    {
      /* We are sure that the text contains no multibyte character.
	 Check if it contains eight-bit-graphic.  */
      p = p1;
      for (p = p1; p < p1end && ASCII_BYTE_P (*p); p++);
      if (p == p1end)
	{
	  for (p = p2; p < p2end && ASCII_BYTE_P (*p); p++);
	  if (p == p2end)
	    return Qt;
	}
    }

  /* The text contains non-ASCII characters.  */

  work_table = Fmake_char_table (Qchar_coding_system, Qnil);
  safe_codings = Fcopy_sequence (XCDR (Vcoding_system_safe_chars));

  safe_codings = find_safe_codings (p1, p1end, safe_codings, work_table,
				    &single_byte_char_found);
  if (p2 < p2end)
    safe_codings = find_safe_codings (p2, p2end, safe_codings, work_table,
				      &single_byte_char_found);
  if (EQ (safe_codings, XCDR (Vcoding_system_safe_chars)))
    safe_codings = Qt;
  else
    {
      /* Turn safe_codings to a list of coding systems... */
      Lisp_Object val;

      if (single_byte_char_found)
	/* ... and append these for eight-bit chars.  */
	val = Fcons (Qraw_text,
		     Fcons (Qemacs_mule, Fcons (Qno_conversion, Qnil)));
      else
	/* ... and append generic coding systems.  */
	val = Fcopy_sequence (XCAR (Vcoding_system_safe_chars));

      for (; CONSP (safe_codings); safe_codings = XCDR (safe_codings))
	val = Fcons (XCAR (XCAR (safe_codings)), val);
      safe_codings = val;
    }

  return safe_codings;
}


/* Search from position POS for such characters that are unencodable
   accoding to SAFE_CHARS, and return a list of their positions.  P
   points where in the memory the character at POS exists.  Limit the
   search at PEND or when Nth unencodable characters are found.

   If SAFE_CHARS is a char table, an element for an unencodable
   character is nil.

   If SAFE_CHARS is nil, all non-ASCII characters are unencodable.

   Otherwise, SAFE_CHARS is t, and only eight-bit-contrl and
   eight-bit-graphic characters are unencodable.  */

static Lisp_Object
unencodable_char_position (safe_chars, pos, p, pend, n)
     Lisp_Object safe_chars;
     int pos;
     unsigned char *p, *pend;
     int n;
{
  Lisp_Object pos_list;

  pos_list = Qnil;
  while (p < pend)
    {
      int len;
      int c = STRING_CHAR_AND_LENGTH (p, MAX_MULTIBYTE_LENGTH, len);

      if (c >= 128
	  && (CHAR_TABLE_P (safe_chars)
	      ? NILP (CHAR_TABLE_REF (safe_chars, c))
	      : (NILP (safe_chars) || c < 256)))
	{
	  pos_list = Fcons (make_number (pos), pos_list);
	  if (--n <= 0)
	    break;
	}
      pos++;
      p += len;
    }
  return Fnreverse (pos_list);
}


DEFUN ("unencodable-char-position", Funencodable_char_position,
       Sunencodable_char_position, 3, 5, 0,
       doc: /*
Return position of first un-encodable character in a region.
START and END specfiy the region and CODING-SYSTEM specifies the
encoding to check.  Return nil if CODING-SYSTEM does encode the region.

If optional 4th argument COUNT is non-nil, it specifies at most how
many un-encodable characters to search.  In this case, the value is a
list of positions.

If optional 5th argument STRING is non-nil, it is a string to search
for un-encodable characters.  In that case, START and END are indexes
to the string.  */)
     (start, end, coding_system, count, string)
     Lisp_Object start, end, coding_system, count, string;
{
  int n;
  Lisp_Object safe_chars;
  struct coding_system coding;
  Lisp_Object positions;
  int from, to;
  unsigned char *p, *pend;

  if (NILP (string))
    {
      validate_region (&start, &end);
      from = XINT (start);
      to = XINT (end);
      if (NILP (current_buffer->enable_multibyte_characters))
	return Qnil;
      p = CHAR_POS_ADDR (from);
      if (to == GPT)
	pend = GPT_ADDR;
      else
	pend = CHAR_POS_ADDR (to);
    }
  else
    {
      CHECK_STRING (string);
      CHECK_NATNUM (start);
      CHECK_NATNUM (end);
      from = XINT (start);
      to = XINT (end);
      if (from > to
	  || to > SCHARS (string))
	args_out_of_range_3 (string, start, end);
      if (! STRING_MULTIBYTE (string))
	return Qnil;
      p = SDATA (string) + string_char_to_byte (string, from);
      pend = SDATA (string) + string_char_to_byte (string, to);
    }

  setup_coding_system (Fcheck_coding_system (coding_system), &coding);

  if (NILP (count))
    n = 1;
  else
    {
      CHECK_NATNUM (count);
      n = XINT (count);
    }

  if (coding.type == coding_type_no_conversion
      || coding.type == coding_type_raw_text)
    return Qnil;

  if (coding.type == coding_type_undecided)
    safe_chars = Qnil;
  else
    safe_chars = coding_safe_chars (coding_system);

  if (STRINGP (string)
      || from >= GPT || to <= GPT)
    positions = unencodable_char_position (safe_chars, from, p, pend, n);
  else
    {
      Lisp_Object args[2];

      args[0] = unencodable_char_position (safe_chars, from, p, GPT_ADDR, n);
      n -= XINT (Flength (args[0]));
      if (n <= 0)
	positions = args[0];
      else
	{
	  args[1] = unencodable_char_position (safe_chars, GPT, GAP_END_ADDR,
					       pend, n);
	  positions = Fappend (2, args);
	}
    }

  return  (NILP (count) ? Fcar (positions) : positions);
}


Lisp_Object
code_convert_region1 (start, end, coding_system, encodep)
     Lisp_Object start, end, coding_system;
     int encodep;
{
  struct coding_system coding;
  int from, to;

  CHECK_NUMBER_COERCE_MARKER (start);
  CHECK_NUMBER_COERCE_MARKER (end);
  CHECK_SYMBOL (coding_system);

  validate_region (&start, &end);
  from = XFASTINT (start);
  to = XFASTINT (end);

  if (NILP (coding_system))
    return make_number (to - from);

  if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0)
    error ("Invalid coding system: %s", SDATA (SYMBOL_NAME (coding_system)));

  coding.mode |= CODING_MODE_LAST_BLOCK;
  coding.src_multibyte = coding.dst_multibyte
    = !NILP (current_buffer->enable_multibyte_characters);
  code_convert_region (from, CHAR_TO_BYTE (from), to, CHAR_TO_BYTE (to),
		       &coding, encodep, 1);
  Vlast_coding_system_used = coding.symbol;
  return make_number (coding.produced_char);
}

DEFUN ("decode-coding-region", Fdecode_coding_region, Sdecode_coding_region,
       3, 3, "r\nzCoding system: ",
       doc: /* Decode the current region from the specified coding system.
When called from a program, takes three arguments:
START, END, and CODING-SYSTEM.  START and END are buffer positions.
This function sets `last-coding-system-used' to the precise coding system
used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
not fully specified.)
It returns the length of the decoded text.  */)
     (start, end, coding_system)
     Lisp_Object start, end, coding_system;
{
  return code_convert_region1 (start, end, coding_system, 0);
}

DEFUN ("encode-coding-region", Fencode_coding_region, Sencode_coding_region,
       3, 3, "r\nzCoding system: ",
       doc: /* Encode the current region into the specified coding system.
When called from a program, takes three arguments:
START, END, and CODING-SYSTEM.  START and END are buffer positions.
This function sets `last-coding-system-used' to the precise coding system
used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
not fully specified.)
It returns the length of the encoded text.  */)
     (start, end, coding_system)
     Lisp_Object start, end, coding_system;
{
  return code_convert_region1 (start, end, coding_system, 1);
}

Lisp_Object
code_convert_string1 (string, coding_system, nocopy, encodep)
     Lisp_Object string, coding_system, nocopy;
     int encodep;
{
  struct coding_system coding;

  CHECK_STRING (string);
  CHECK_SYMBOL (coding_system);

  if (NILP (coding_system))
    return (NILP (nocopy) ? Fcopy_sequence (string) : string);

  if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0)
    error ("Invalid coding system: %s", SDATA (SYMBOL_NAME (coding_system)));

  coding.mode |= CODING_MODE_LAST_BLOCK;
  string = (encodep
	    ? encode_coding_string (string, &coding, !NILP (nocopy))
	    : decode_coding_string (string, &coding, !NILP (nocopy)));
  Vlast_coding_system_used = coding.symbol;

  return string;
}

DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string,
       2, 3, 0,
       doc: /* Decode STRING which is encoded in CODING-SYSTEM, and return the result.
Optional arg NOCOPY non-nil means it is OK to return STRING itself
if the decoding operation is trivial.
This function sets `last-coding-system-used' to the precise coding system
used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
not fully specified.)  */)
     (string, coding_system, nocopy)
     Lisp_Object string, coding_system, nocopy;
{
  return code_convert_string1 (string, coding_system, nocopy, 0);
}

DEFUN ("encode-coding-string", Fencode_coding_string, Sencode_coding_string,
       2, 3, 0,
       doc: /* Encode STRING to CODING-SYSTEM, and return the result.
Optional arg NOCOPY non-nil means it is OK to return STRING itself
if the encoding operation is trivial.
This function sets `last-coding-system-used' to the precise coding system
used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
not fully specified.)  */)
     (string, coding_system, nocopy)
     Lisp_Object string, coding_system, nocopy;
{
  return code_convert_string1 (string, coding_system, nocopy, 1);
}

/* Encode or decode STRING according to CODING_SYSTEM.
   Do not set Vlast_coding_system_used.

   This function is called only from macros DECODE_FILE and
   ENCODE_FILE, thus we ignore character composition.  */

Lisp_Object
code_convert_string_norecord (string, coding_system, encodep)
     Lisp_Object string, coding_system;
     int encodep;
{
  struct coding_system coding;

  CHECK_STRING (string);
  CHECK_SYMBOL (coding_system);

  if (NILP (coding_system))
    return string;

  if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0)
    error ("Invalid coding system: %s", SDATA (SYMBOL_NAME (coding_system)));

  coding.composing = COMPOSITION_DISABLED;
  coding.mode |= CODING_MODE_LAST_BLOCK;
  return (encodep
	  ? encode_coding_string (string, &coding, 1)
	  : decode_coding_string (string, &coding, 1));
}

DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0,
       doc: /* Decode a Japanese character which has CODE in shift_jis encoding.
Return the corresponding character.  */)
     (code)
     Lisp_Object code;
{
  unsigned char c1, c2, s1, s2;
  Lisp_Object val;

  CHECK_NUMBER (code);
  s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF;
  if (s1 == 0)
    {
      if (s2 < 0x80)
	XSETFASTINT (val, s2);
      else if (s2 >= 0xA0 || s2 <= 0xDF)
	XSETFASTINT (val, MAKE_CHAR (charset_katakana_jisx0201, s2, 0));
      else
	error ("Invalid Shift JIS code: %x", XFASTINT (code));
    }
  else
    {
      if ((s1 < 0x80 || (s1 > 0x9F && s1 < 0xE0) || s1 > 0xEF)
	  || (s2 < 0x40 || s2 == 0x7F || s2 > 0xFC))
	error ("Invalid Shift JIS code: %x", XFASTINT (code));
      DECODE_SJIS (s1, s2, c1, c2);
      XSETFASTINT (val, MAKE_CHAR (charset_jisx0208, c1, c2));
    }
  return val;
}

DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0,
       doc: /* Encode a Japanese character CH to shift_jis encoding.
Return the corresponding code in SJIS.  */)
     (ch)
     Lisp_Object ch;
{
  int charset, c1, c2, s1, s2;
  Lisp_Object val;

  CHECK_NUMBER (ch);
  SPLIT_CHAR (XFASTINT (ch), charset, c1, c2);
  if (charset == CHARSET_ASCII)
    {
      val = ch;
    }
  else if (charset == charset_jisx0208
	   && c1 > 0x20 && c1 < 0x7F && c2 > 0x20 && c2 < 0x7F)
    {
      ENCODE_SJIS (c1, c2, s1, s2);
      XSETFASTINT (val, (s1 << 8) | s2);
    }
  else if (charset == charset_katakana_jisx0201
	   && c1 > 0x20 && c2 < 0xE0)
    {
      XSETFASTINT (val, c1 | 0x80);
    }
  else
    error ("Can't encode to shift_jis: %d", XFASTINT (ch));
  return val;
}

DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0,
       doc: /* Decode a Big5 character which has CODE in BIG5 coding system.
Return the corresponding character.  */)
     (code)
     Lisp_Object code;
{
  int charset;
  unsigned char b1, b2, c1, c2;
  Lisp_Object val;

  CHECK_NUMBER (code);
  b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF;
  if (b1 == 0)
    {
      if (b2 >= 0x80)
	error ("Invalid BIG5 code: %x", XFASTINT (code));
      val = code;
    }
  else
    {
      if ((b1 < 0xA1 || b1 > 0xFE)
	  || (b2 < 0x40 || (b2 > 0x7E && b2 < 0xA1) || b2 > 0xFE))
	error ("Invalid BIG5 code: %x", XFASTINT (code));
      DECODE_BIG5 (b1, b2, charset, c1, c2);
      XSETFASTINT (val, MAKE_CHAR (charset, c1, c2));
    }
  return val;
}

DEFUN ("encode-big5-char", Fencode_big5_char, Sencode_big5_char, 1, 1, 0,
       doc: /* Encode the Big5 character CH to BIG5 coding system.
Return the corresponding character code in Big5.  */)
     (ch)
     Lisp_Object ch;
{
  int charset, c1, c2, b1, b2;
  Lisp_Object val;

  CHECK_NUMBER (ch);
  SPLIT_CHAR (XFASTINT (ch), charset, c1, c2);
  if (charset == CHARSET_ASCII)
    {
      val = ch;
    }
  else if ((charset == charset_big5_1
	    && (XFASTINT (ch) >= 0x250a1 && XFASTINT (ch) <= 0x271ec))
	   || (charset == charset_big5_2
	       && XFASTINT (ch) >= 0x290a1 && XFASTINT (ch) <= 0x2bdb2))
    {
      ENCODE_BIG5 (charset, c1, c2, b1, b2);
      XSETFASTINT (val, (b1 << 8) | b2);
    }
  else
    error ("Can't encode to Big5: %d", XFASTINT (ch));
  return val;
}

DEFUN ("set-terminal-coding-system-internal", Fset_terminal_coding_system_internal,
       Sset_terminal_coding_system_internal, 1, 1, 0,
       doc: /* Internal use only.  */)
     (coding_system)
     Lisp_Object coding_system;
{
  CHECK_SYMBOL (coding_system);
  setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding);
  /* We had better not send unsafe characters to terminal.  */
  terminal_coding.mode |= CODING_MODE_INHIBIT_UNENCODABLE_CHAR;
  /* Character composition should be disabled.  */
  terminal_coding.composing = COMPOSITION_DISABLED;
  /* Error notification should be suppressed.  */
  terminal_coding.suppress_error = 1;
  terminal_coding.src_multibyte = 1;
  terminal_coding.dst_multibyte = 0;
  return Qnil;
}

DEFUN ("set-safe-terminal-coding-system-internal", Fset_safe_terminal_coding_system_internal,
       Sset_safe_terminal_coding_system_internal, 1, 1, 0,
       doc: /* Internal use only.  */)
     (coding_system)
     Lisp_Object coding_system;
{
  CHECK_SYMBOL (coding_system);
  setup_coding_system (Fcheck_coding_system (coding_system),
		       &safe_terminal_coding);
  /* Character composition should be disabled.  */
  safe_terminal_coding.composing = COMPOSITION_DISABLED;
  /* Error notification should be suppressed.  */
  safe_terminal_coding.suppress_error = 1;
  safe_terminal_coding.src_multibyte = 1;
  safe_terminal_coding.dst_multibyte = 0;
  return Qnil;
}

DEFUN ("terminal-coding-system", Fterminal_coding_system,
       Sterminal_coding_system, 0, 0, 0,
       doc: /* Return coding system specified for terminal output.  */)
     ()
{
  return terminal_coding.symbol;
}

DEFUN ("set-keyboard-coding-system-internal", Fset_keyboard_coding_system_internal,
       Sset_keyboard_coding_system_internal, 1, 1, 0,
       doc: /* Internal use only.  */)
     (coding_system)
     Lisp_Object coding_system;
{
  CHECK_SYMBOL (coding_system);
  setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding);
  /* Character composition should be disabled.  */
  keyboard_coding.composing = COMPOSITION_DISABLED;
  return Qnil;
}

DEFUN ("keyboard-coding-system", Fkeyboard_coding_system,
       Skeyboard_coding_system, 0, 0, 0,
       doc: /* Return coding system specified for decoding keyboard input.  */)
     ()
{
  return keyboard_coding.symbol;
}


DEFUN ("find-operation-coding-system", Ffind_operation_coding_system,
       Sfind_operation_coding_system,  1, MANY, 0,
       doc: /* Choose a coding system for an operation based on the target name.
The value names a pair of coding systems: (DECODING-SYSTEM . ENCODING-SYSTEM).
DECODING-SYSTEM is the coding system to use for decoding
\(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system
for encoding (in case OPERATION does encoding).

The first argument OPERATION specifies an I/O primitive:
  For file I/O, `insert-file-contents' or `write-region'.
  For process I/O, `call-process', `call-process-region', or `start-process'.
  For network I/O, `open-network-stream'.

The remaining arguments should be the same arguments that were passed
to the primitive.  Depending on which primitive, one of those arguments
is selected as the TARGET.  For example, if OPERATION does file I/O,
whichever argument specifies the file name is TARGET.

TARGET has a meaning which depends on OPERATION:
  For file I/O, TARGET is a file name (except for the special case below).
  For process I/O, TARGET is a process name.
  For network I/O, TARGET is a service name or a port number

This function looks up what specified for TARGET in,
`file-coding-system-alist', `process-coding-system-alist',
or `network-coding-system-alist' depending on OPERATION.
They may specify a coding system, a cons of coding systems,
or a function symbol to call.
In the last case, we call the function with one argument,
which is a list of all the arguments given to this function.
If the function can't decide a coding system, it can return
`undecided' so that the normal code-detection is performed.

If OPERATION is `insert-file-contents', the argument corresponding to
TARGET may be a cons (FILENAME . BUFFER).  In that case, FILENAME is a
file name to look up, and BUFFER is a buffer that contains the file's
contents (not yet decoded).  If `file-coding-system-alist' specifies a
function to call for FILENAME, that function should examine the
contents of BUFFER instead of reading the file.

usage: (find-operation-coding-system OPERATION ARGUMENTS...)  */)
     (nargs, args)
     int nargs;
     Lisp_Object *args;
{
  Lisp_Object operation, target_idx, target, val;
  register Lisp_Object chain;

  if (nargs < 2)
    error ("Too few arguments");
  operation = args[0];
  if (!SYMBOLP (operation)
      || !INTEGERP (target_idx = Fget (operation, Qtarget_idx)))
    error ("Invalid first argument");
  if (nargs < 1 + XINT (target_idx))
    error ("Too few arguments for operation: %s",
	   SDATA (SYMBOL_NAME (operation)));
  /* For write-region, if the 6th argument (i.e. VISIT, the 5th
     argument to write-region) is string, it must be treated as a
     target file name.  */
  if (EQ (operation, Qwrite_region)
      && nargs > 5
      && STRINGP (args[5]))
    target_idx = make_number (4);
  target = args[XINT (target_idx) + 1];
  if (!(STRINGP (target)
	|| (EQ (operation, Qinsert_file_contents) && CONSP (target)
	    && STRINGP (XCAR (target)) && BUFFERP (XCDR (target)))
	|| (EQ (operation, Qopen_network_stream) && INTEGERP (target))))
    error ("Invalid argument %d", XINT (target_idx) + 1);
  if (CONSP (target))
    target = XCAR (target);

  chain = ((EQ (operation, Qinsert_file_contents)
	    || EQ (operation, Qwrite_region))
	   ? Vfile_coding_system_alist
	   : (EQ (operation, Qopen_network_stream)
	      ? Vnetwork_coding_system_alist
	      : Vprocess_coding_system_alist));
  if (NILP (chain))
    return Qnil;

  for (; CONSP (chain); chain = XCDR (chain))
    {
      Lisp_Object elt;
      elt = XCAR (chain);

      if (CONSP (elt)
	  && ((STRINGP (target)
	       && STRINGP (XCAR (elt))
	       && fast_string_match (XCAR (elt), target) >= 0)
	      || (INTEGERP (target) && EQ (target, XCAR (elt)))))
	{
	  val = XCDR (elt);
	  /* Here, if VAL is both a valid coding system and a valid
             function symbol, we return VAL as a coding system.  */
	  if (CONSP (val))
	    return val;
	  if (! SYMBOLP (val))
	    return Qnil;
	  if (! NILP (Fcoding_system_p (val)))
	    return Fcons (val, val);
	  if (! NILP (Ffboundp (val)))
	    {
	      /* We use call1 rather than safe_call1
		 so as to get bug reports about functions called here
		 which don't handle the current interface.  */
	      val = call1 (val, Flist (nargs, args));
	      if (CONSP (val))
		return val;
	      if (SYMBOLP (val) && ! NILP (Fcoding_system_p (val)))
		return Fcons (val, val);
	    }
	  return Qnil;
	}
    }
  return Qnil;
}

DEFUN ("update-coding-systems-internal",  Fupdate_coding_systems_internal,
       Supdate_coding_systems_internal, 0, 0, 0,
       doc: /* Update internal database for ISO2022 and CCL based coding systems.
When values of any coding categories are changed, you must
call this function.  */)
     ()
{
  int i;

  for (i = CODING_CATEGORY_IDX_EMACS_MULE; i < CODING_CATEGORY_IDX_MAX; i++)
    {
      Lisp_Object val;

      val = SYMBOL_VALUE (XVECTOR (Vcoding_category_table)->contents[i]);
      if (!NILP (val))
	{
	  if (! coding_system_table[i])
	    coding_system_table[i] = ((struct coding_system *)
				      xmalloc (sizeof (struct coding_system)));
	  setup_coding_system (val, coding_system_table[i]);
	}
      else if (coding_system_table[i])
	{
	  xfree (coding_system_table[i]);
	  coding_system_table[i] = NULL;
	}
    }

  return Qnil;
}

DEFUN ("set-coding-priority-internal", Fset_coding_priority_internal,
       Sset_coding_priority_internal, 0, 0, 0,
       doc: /* Update internal database for the current value of `coding-category-list'.
This function is internal use only.  */)
     ()
{
  int i = 0, idx;
  Lisp_Object val;

  val = Vcoding_category_list;

  while (CONSP (val) && i < CODING_CATEGORY_IDX_MAX)
    {
      if (! SYMBOLP (XCAR (val)))
	break;
      idx = XFASTINT (Fget (XCAR (val), Qcoding_category_index));
      if (idx >= CODING_CATEGORY_IDX_MAX)
	break;
      coding_priorities[i++] = (1 << idx);
      val = XCDR (val);
    }
  /* If coding-category-list is valid and contains all coding
     categories, `i' should be CODING_CATEGORY_IDX_MAX now.  If not,
     the following code saves Emacs from crashing.  */
  while (i < CODING_CATEGORY_IDX_MAX)
    coding_priorities[i++] = CODING_CATEGORY_MASK_RAW_TEXT;

  return Qnil;
}

DEFUN ("define-coding-system-internal", Fdefine_coding_system_internal,
       Sdefine_coding_system_internal, 1, 1, 0,
       doc: /* Register CODING-SYSTEM as a base coding system.
This function is internal use only.  */)
     (coding_system)
     Lisp_Object coding_system;
{
  Lisp_Object safe_chars, slot;

  if (NILP (Fcheck_coding_system (coding_system)))
    xsignal1 (Qcoding_system_error, coding_system);

  safe_chars = coding_safe_chars (coding_system);
  if (! EQ (safe_chars, Qt) && ! CHAR_TABLE_P (safe_chars))
    error ("No valid safe-chars property for %s",
	   SDATA (SYMBOL_NAME (coding_system)));

  if (EQ (safe_chars, Qt))
    {
      if (NILP (Fmemq (coding_system, XCAR (Vcoding_system_safe_chars))))
	XSETCAR (Vcoding_system_safe_chars,
		 Fcons (coding_system, XCAR (Vcoding_system_safe_chars)));
    }
  else
    {
      slot = Fassq (coding_system, XCDR (Vcoding_system_safe_chars));
      if (NILP (slot))
	XSETCDR (Vcoding_system_safe_chars,
		 nconc2 (XCDR (Vcoding_system_safe_chars),
			 Fcons (Fcons (coding_system, safe_chars), Qnil)));
      else
	XSETCDR (slot, safe_chars);
    }
  return Qnil;
}

#endif /* emacs */


/*** 9. Post-amble ***/

void
init_coding_once ()
{
  int i;

  /* Emacs' internal format specific initialize routine.  */
  for (i = 0; i <= 0x20; i++)
    emacs_code_class[i] = EMACS_control_code;
  emacs_code_class[0x0A] = EMACS_linefeed_code;
  emacs_code_class[0x0D] = EMACS_carriage_return_code;
  for (i = 0x21 ; i < 0x7F; i++)
    emacs_code_class[i] = EMACS_ascii_code;
  emacs_code_class[0x7F] = EMACS_control_code;
  for (i = 0x80; i < 0xFF; i++)
    emacs_code_class[i] = EMACS_invalid_code;
  emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3;
  emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3;
  emacs_code_class[LEADING_CODE_PRIVATE_21] = EMACS_leading_code_4;
  emacs_code_class[LEADING_CODE_PRIVATE_22] = EMACS_leading_code_4;

  /* ISO2022 specific initialize routine.  */
  for (i = 0; i < 0x20; i++)
    iso_code_class[i] = ISO_control_0;
  for (i = 0x21; i < 0x7F; i++)
    iso_code_class[i] = ISO_graphic_plane_0;
  for (i = 0x80; i < 0xA0; i++)
    iso_code_class[i] = ISO_control_1;
  for (i = 0xA1; i < 0xFF; i++)
    iso_code_class[i] = ISO_graphic_plane_1;
  iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F;
  iso_code_class[0xA0] = iso_code_class[0xFF] = ISO_0xA0_or_0xFF;
  iso_code_class[ISO_CODE_CR] = ISO_carriage_return;
  iso_code_class[ISO_CODE_SO] = ISO_shift_out;
  iso_code_class[ISO_CODE_SI] = ISO_shift_in;
  iso_code_class[ISO_CODE_SS2_7] = ISO_single_shift_2_7;
  iso_code_class[ISO_CODE_ESC] = ISO_escape;
  iso_code_class[ISO_CODE_SS2] = ISO_single_shift_2;
  iso_code_class[ISO_CODE_SS3] = ISO_single_shift_3;
  iso_code_class[ISO_CODE_CSI] = ISO_control_sequence_introducer;

  setup_coding_system (Qnil, &keyboard_coding);
  setup_coding_system (Qnil, &terminal_coding);
  setup_coding_system (Qnil, &safe_terminal_coding);
  setup_coding_system (Qnil, &default_buffer_file_coding);

  bzero (coding_system_table, sizeof coding_system_table);

  bzero (ascii_skip_code, sizeof ascii_skip_code);
  for (i = 0; i < 128; i++)
    ascii_skip_code[i] = 1;

#if defined (MSDOS) || defined (WINDOWSNT)
  system_eol_type = CODING_EOL_CRLF;
#else
  system_eol_type = CODING_EOL_LF;
#endif

  inhibit_pre_post_conversion = 0;
}

#ifdef emacs

void
syms_of_coding ()
{
  staticpro (&Vcode_conversion_workbuf_name);
  Vcode_conversion_workbuf_name = build_string (" *code-conversion-work*");

  Qtarget_idx = intern ("target-idx");
  staticpro (&Qtarget_idx);

  Qcoding_system_history = intern ("coding-system-history");
  staticpro (&Qcoding_system_history);
  Fset (Qcoding_system_history, Qnil);

  /* Target FILENAME is the first argument.  */
  Fput (Qinsert_file_contents, Qtarget_idx, make_number (0));
  /* Target FILENAME is the third argument.  */
  Fput (Qwrite_region, Qtarget_idx, make_number (2));

  Qcall_process = intern ("call-process");
  staticpro (&Qcall_process);
  /* Target PROGRAM is the first argument.  */
  Fput (Qcall_process, Qtarget_idx, make_number (0));

  Qcall_process_region = intern ("call-process-region");
  staticpro (&Qcall_process_region);
  /* Target PROGRAM is the third argument.  */
  Fput (Qcall_process_region, Qtarget_idx, make_number (2));

  Qstart_process = intern ("start-process");
  staticpro (&Qstart_process);
  /* Target PROGRAM is the third argument.  */
  Fput (Qstart_process, Qtarget_idx, make_number (2));

  Qopen_network_stream = intern ("open-network-stream");
  staticpro (&Qopen_network_stream);
  /* Target SERVICE is the fourth argument.  */
  Fput (Qopen_network_stream, Qtarget_idx, make_number (3));

  Qcoding_system = intern ("coding-system");
  staticpro (&Qcoding_system);

  Qeol_type = intern ("eol-type");
  staticpro (&Qeol_type);

  Qbuffer_file_coding_system = intern ("buffer-file-coding-system");
  staticpro (&Qbuffer_file_coding_system);

  Qpost_read_conversion = intern ("post-read-conversion");
  staticpro (&Qpost_read_conversion);

  Qpre_write_conversion = intern ("pre-write-conversion");
  staticpro (&Qpre_write_conversion);

  Qno_conversion = intern ("no-conversion");
  staticpro (&Qno_conversion);

  Qundecided = intern ("undecided");
  staticpro (&Qundecided);

  Qcoding_system_p = intern ("coding-system-p");
  staticpro (&Qcoding_system_p);

  Qcoding_system_error = intern ("coding-system-error");
  staticpro (&Qcoding_system_error);

  Fput (Qcoding_system_error, Qerror_conditions,
	Fcons (Qcoding_system_error, Fcons (Qerror, Qnil)));
  Fput (Qcoding_system_error, Qerror_message,
	build_string ("Invalid coding system"));

  Qcoding_category = intern ("coding-category");
  staticpro (&Qcoding_category);
  Qcoding_category_index = intern ("coding-category-index");
  staticpro (&Qcoding_category_index);

  Vcoding_category_table
    = Fmake_vector (make_number (CODING_CATEGORY_IDX_MAX), Qnil);
  staticpro (&Vcoding_category_table);
  {
    int i;
    for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++)
      {
	XVECTOR (Vcoding_category_table)->contents[i]
	  = intern (coding_category_name[i]);
	Fput (XVECTOR (Vcoding_category_table)->contents[i],
	      Qcoding_category_index, make_number (i));
      }
  }

  Vcoding_system_safe_chars = Fcons (Qnil, Qnil);
  staticpro (&Vcoding_system_safe_chars);

  Qtranslation_table = intern ("translation-table");
  staticpro (&Qtranslation_table);
  Fput (Qtranslation_table, Qchar_table_extra_slots, make_number (2));

  Qtranslation_table_id = intern ("translation-table-id");
  staticpro (&Qtranslation_table_id);

  Qtranslation_table_for_decode = intern ("translation-table-for-decode");
  staticpro (&Qtranslation_table_for_decode);

  Qtranslation_table_for_encode = intern ("translation-table-for-encode");
  staticpro (&Qtranslation_table_for_encode);

  Qsafe_chars = intern ("safe-chars");
  staticpro (&Qsafe_chars);

  Qchar_coding_system = intern ("char-coding-system");
  staticpro (&Qchar_coding_system);

  /* Intern this now in case it isn't already done.
     Setting this variable twice is harmless.
     But don't staticpro it here--that is done in alloc.c.  */
  Qchar_table_extra_slots = intern ("char-table-extra-slots");
  Fput (Qsafe_chars, Qchar_table_extra_slots, make_number (0));
  Fput (Qchar_coding_system, Qchar_table_extra_slots, make_number (0));

  Qvalid_codes = intern ("valid-codes");
  staticpro (&Qvalid_codes);

  Qascii_incompatible = intern ("ascii-incompatible");
  staticpro (&Qascii_incompatible);

  Qemacs_mule = intern ("emacs-mule");
  staticpro (&Qemacs_mule);

  Qraw_text = intern ("raw-text");
  staticpro (&Qraw_text);

  Qutf_8 = intern ("utf-8");
  staticpro (&Qutf_8);

  Qcoding_system_define_form = intern ("coding-system-define-form");
  staticpro (&Qcoding_system_define_form);

  defsubr (&Scoding_system_p);
  defsubr (&Sread_coding_system);
  defsubr (&Sread_non_nil_coding_system);
  defsubr (&Scheck_coding_system);
  defsubr (&Sdetect_coding_region);
  defsubr (&Sdetect_coding_string);
  defsubr (&Sfind_coding_systems_region_internal);
  defsubr (&Sunencodable_char_position);
  defsubr (&Sdecode_coding_region);
  defsubr (&Sencode_coding_region);
  defsubr (&Sdecode_coding_string);
  defsubr (&Sencode_coding_string);
  defsubr (&Sdecode_sjis_char);
  defsubr (&Sencode_sjis_char);
  defsubr (&Sdecode_big5_char);
  defsubr (&Sencode_big5_char);
  defsubr (&Sset_terminal_coding_system_internal);
  defsubr (&Sset_safe_terminal_coding_system_internal);
  defsubr (&Sterminal_coding_system);
  defsubr (&Sset_keyboard_coding_system_internal);
  defsubr (&Skeyboard_coding_system);
  defsubr (&Sfind_operation_coding_system);
  defsubr (&Supdate_coding_systems_internal);
  defsubr (&Sset_coding_priority_internal);
  defsubr (&Sdefine_coding_system_internal);

  DEFVAR_LISP ("coding-system-list", &Vcoding_system_list,
	       doc: /* List of coding systems.

Do not alter the value of this variable manually.  This variable should be
updated by the functions `make-coding-system' and
`define-coding-system-alias'.  */);
  Vcoding_system_list = Qnil;

  DEFVAR_LISP ("coding-system-alist", &Vcoding_system_alist,
	       doc: /* Alist of coding system names.
Each element is one element list of coding system name.
This variable is given to `completing-read' as TABLE argument.

Do not alter the value of this variable manually.  This variable should be
updated by the functions `make-coding-system' and
`define-coding-system-alias'.  */);
  Vcoding_system_alist = Qnil;

  DEFVAR_LISP ("coding-category-list", &Vcoding_category_list,
	       doc: /* List of coding-categories (symbols) ordered by priority.

On detecting a coding system, Emacs tries code detection algorithms
associated with each coding-category one by one in this order.  When
one algorithm agrees with a byte sequence of source text, the coding
system bound to the corresponding coding-category is selected.

Don't modify this variable directly, but use `set-coding-priority'.  */);
  {
    int i;

    Vcoding_category_list = Qnil;
    for (i = CODING_CATEGORY_IDX_MAX - 1; i >= 0; i--)
      Vcoding_category_list
	= Fcons (XVECTOR (Vcoding_category_table)->contents[i],
		 Vcoding_category_list);
  }

  DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read,
	       doc: /* Specify the coding system for read operations.
It is useful to bind this variable with `let', but do not set it globally.
If the value is a coding system, it is used for decoding on read operation.
If not, an appropriate element is used from one of the coding system alists:
There are three such tables, `file-coding-system-alist',
`process-coding-system-alist', and `network-coding-system-alist'.  */);
  Vcoding_system_for_read = Qnil;

  DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write,
	       doc: /* Specify the coding system for write operations.
Programs bind this variable with `let', but you should not set it globally.
If the value is a coding system, it is used for encoding of output,
when writing it to a file and when sending it to a file or subprocess.

If this does not specify a coding system, an appropriate element
is used from one of the coding system alists:
There are three such tables, `file-coding-system-alist',
`process-coding-system-alist', and `network-coding-system-alist'.
For output to files, if the above procedure does not specify a coding system,
the value of `buffer-file-coding-system' is used.  */);
  Vcoding_system_for_write = Qnil;

  DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used,
	       doc: /* Coding system used in the latest file or process I/O.
Also set by `encode-coding-region', `decode-coding-region',
`encode-coding-string' and `decode-coding-string'.  */);
  Vlast_coding_system_used = Qnil;

  DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion,
	       doc: /* *Non-nil means always inhibit code conversion of end-of-line format.
See info node `Coding Systems' and info node `Text and Binary' concerning
such conversion.  */);
  inhibit_eol_conversion = 0;

  DEFVAR_BOOL ("inherit-process-coding-system", &inherit_process_coding_system,
	       doc: /* Non-nil means process buffer inherits coding system of process output.
Bind it to t if the process output is to be treated as if it were a file
read from some filesystem.  */);
  inherit_process_coding_system = 0;

  DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist,
	       doc: /* Alist to decide a coding system to use for a file I/O operation.
The format is ((PATTERN . VAL) ...),
where PATTERN is a regular expression matching a file name,
VAL is a coding system, a cons of coding systems, or a function symbol.
If VAL is a coding system, it is used for both decoding and encoding
the file contents.
If VAL is a cons of coding systems, the car part is used for decoding,
and the cdr part is used for encoding.
If VAL is a function symbol, the function must return a coding system
or a cons of coding systems which are used as above.  The function is
called with an argument that is a list of the arguments with which
`find-operation-coding-system' was called.  If the function can't decide
a coding system, it can return `undecided' so that the normal
code-detection is performed.

See also the function `find-operation-coding-system'
and the variable `auto-coding-alist'.  */);
  Vfile_coding_system_alist = Qnil;

  DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist,
    doc: /* Alist to decide a coding system to use for a process I/O operation.
The format is ((PATTERN . VAL) ...),
where PATTERN is a regular expression matching a program name,
VAL is a coding system, a cons of coding systems, or a function symbol.
If VAL is a coding system, it is used for both decoding what received
from the program and encoding what sent to the program.
If VAL is a cons of coding systems, the car part is used for decoding,
and the cdr part is used for encoding.
If VAL is a function symbol, the function must return a coding system
or a cons of coding systems which are used as above.

See also the function `find-operation-coding-system'.  */);
  Vprocess_coding_system_alist = Qnil;

  DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist,
    doc: /* Alist to decide a coding system to use for a network I/O operation.
The format is ((PATTERN . VAL) ...),
where PATTERN is a regular expression matching a network service name
or is a port number to connect to,
VAL is a coding system, a cons of coding systems, or a function symbol.
If VAL is a coding system, it is used for both decoding what received
from the network stream and encoding what sent to the network stream.
If VAL is a cons of coding systems, the car part is used for decoding,
and the cdr part is used for encoding.
If VAL is a function symbol, the function must return a coding system
or a cons of coding systems which are used as above.

See also the function `find-operation-coding-system'.  */);
  Vnetwork_coding_system_alist = Qnil;

  DEFVAR_LISP ("locale-coding-system", &Vlocale_coding_system,
	       doc: /* Coding system to use with system messages.
Also used for decoding keyboard input on X Window system.  */);
  Vlocale_coding_system = Qnil;

  /* The eol mnemonics are reset in startup.el system-dependently.  */
  DEFVAR_LISP ("eol-mnemonic-unix", &eol_mnemonic_unix,
	       doc: /* *String displayed in mode line for UNIX-like (LF) end-of-line format.  */);
  eol_mnemonic_unix = build_string (":");

  DEFVAR_LISP ("eol-mnemonic-dos", &eol_mnemonic_dos,
	       doc: /* *String displayed in mode line for DOS-like (CRLF) end-of-line format.  */);
  eol_mnemonic_dos = build_string ("\\");

  DEFVAR_LISP ("eol-mnemonic-mac", &eol_mnemonic_mac,
	       doc: /* *String displayed in mode line for MAC-like (CR) end-of-line format.  */);
  eol_mnemonic_mac = build_string ("/");

  DEFVAR_LISP ("eol-mnemonic-undecided", &eol_mnemonic_undecided,
	       doc: /* *String displayed in mode line when end-of-line format is not yet determined.  */);
  eol_mnemonic_undecided = build_string (":");

  DEFVAR_LISP ("enable-character-translation", &Venable_character_translation,
	       doc: /* *Non-nil enables character translation while encoding and decoding.  */);
  Venable_character_translation = Qt;

  DEFVAR_LISP ("standard-translation-table-for-decode",
	       &Vstandard_translation_table_for_decode,
	       doc: /* Table for translating characters while decoding.  */);
  Vstandard_translation_table_for_decode = Qnil;

  DEFVAR_LISP ("standard-translation-table-for-encode",
	       &Vstandard_translation_table_for_encode,
	       doc: /* Table for translating characters while encoding.  */);
  Vstandard_translation_table_for_encode = Qnil;

  DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist,
	       doc: /* Alist of charsets vs revision numbers.
While encoding, if a charset (car part of an element) is found,
designate it with the escape sequence identifying revision (cdr part of the element).  */);
  Vcharset_revision_alist = Qnil;

  DEFVAR_LISP ("default-process-coding-system",
	       &Vdefault_process_coding_system,
	       doc: /* Cons of coding systems used for process I/O by default.
The car part is used for decoding a process output,
the cdr part is used for encoding a text to be sent to a process.  */);
  Vdefault_process_coding_system = Qnil;

  DEFVAR_LISP ("latin-extra-code-table", &Vlatin_extra_code_table,
	       doc: /* Table of extra Latin codes in the range 128..159 (inclusive).
This is a vector of length 256.
If Nth element is non-nil, the existence of code N in a file
\(or output of subprocess) doesn't prevent it to be detected as
a coding system of ISO 2022 variant which has a flag
`accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file
or reading output of a subprocess.
Only 128th through 159th elements has a meaning.  */);
  Vlatin_extra_code_table = Fmake_vector (make_number (256), Qnil);

  DEFVAR_LISP ("select-safe-coding-system-function",
	       &Vselect_safe_coding_system_function,
	       doc: /* Function to call to select safe coding system for encoding a text.

If set, this function is called to force a user to select a proper
coding system which can encode the text in the case that a default
coding system used in each operation can't encode the text.

The default value is `select-safe-coding-system' (which see).  */);
  Vselect_safe_coding_system_function = Qnil;

  DEFVAR_BOOL ("coding-system-require-warning",
	       &coding_system_require_warning,
	       doc: /* Internal use only.
If non-nil, on writing a file, `select-safe-coding-system-function' is
called even if `coding-system-for-write' is non-nil.  The command
`universal-coding-system-argument' binds this variable to t temporarily.  */);
  coding_system_require_warning = 0;


  DEFVAR_BOOL ("inhibit-iso-escape-detection",
	       &inhibit_iso_escape_detection,
	       doc: /* If non-nil, Emacs ignores ISO2022's escape sequence on code detection.

By default, on reading a file, Emacs tries to detect how the text is
encoded.  This code detection is sensitive to escape sequences.  If
the sequence is valid as ISO2022, the code is determined as one of
the ISO2022 encodings, and the file is decoded by the corresponding
coding system (e.g. `iso-2022-7bit').

However, there may be a case that you want to read escape sequences in
a file as is.  In such a case, you can set this variable to non-nil.
Then, as the code detection ignores any escape sequences, no file is
detected as encoded in some ISO2022 encoding.  The result is that all
escape sequences become visible in a buffer.

The default value is nil, and it is strongly recommended not to change
it.  That is because many Emacs Lisp source files that contain
non-ASCII characters are encoded by the coding system `iso-2022-7bit'
in Emacs's distribution, and they won't be decoded correctly on
reading if you suppress escape sequence detection.

The other way to read escape sequences in a file without decoding is
to explicitly specify some coding system that doesn't use ISO2022's
escape sequence (e.g `latin-1') on reading by \\[universal-coding-system-argument].  */);
  inhibit_iso_escape_detection = 0;

  DEFVAR_LISP ("translation-table-for-input", &Vtranslation_table_for_input,
	       doc: /* Char table for translating self-inserting characters.
This is applied to the result of input methods, not their input.  See also
`keyboard-translate-table'.  */);
    Vtranslation_table_for_input = Qnil;
}

char *
emacs_strerror (error_number)
     int error_number;
{
  char *str;

  synchronize_system_messages_locale ();
  str = strerror (error_number);

  if (! NILP (Vlocale_coding_system))
    {
      Lisp_Object dec = code_convert_string_norecord (build_string (str),
						      Vlocale_coding_system,
						      0);
      str = (char *) SDATA (dec);
    }

  return str;
}

#endif /* emacs */

/* arch-tag: 3a3a2b01-5ff6-4071-9afe-f5b808d9229d
   (do not change this comment) */