cpython-withatomic / Python / formatter_unicode.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
/* implements the unicode (as opposed to string) version of the
   built-in formatters for string, int, float.  that is, the versions
   of int.__float__, etc., that take and return unicode objects */

#include "Python.h"
#include <locale.h>

/* Raises an exception about an unknown presentation type for this
 * type. */

static void
unknown_presentation_type(Py_UCS4 presentation_type,
                          const char* type_name)
{
    /* %c might be out-of-range, hence the two cases. */
    if (presentation_type > 32 && presentation_type < 128)
        PyErr_Format(PyExc_ValueError,
                     "Unknown format code '%c' "
                     "for object of type '%.200s'",
                     (char)presentation_type,
                     type_name);
    else
        PyErr_Format(PyExc_ValueError,
                     "Unknown format code '\\x%x' "
                     "for object of type '%.200s'",
                     (unsigned int)presentation_type,
                     type_name);
}

static void
invalid_comma_type(Py_UCS4 presentation_type)
{
    if (presentation_type > 32 && presentation_type < 128)
        PyErr_Format(PyExc_ValueError,
                     "Cannot specify ',' with '%c'.",
                     (char)presentation_type);
    else
        PyErr_Format(PyExc_ValueError,
                     "Cannot specify ',' with '\\x%x'.",
                     (unsigned int)presentation_type);
}

/*
    get_integer consumes 0 or more decimal digit characters from an
    input string, updates *result with the corresponding positive
    integer, and returns the number of digits consumed.

    returns -1 on error.
*/
static int
get_integer(PyObject *str, Py_ssize_t *pos, Py_ssize_t end,
                  Py_ssize_t *result)
{
    Py_ssize_t accumulator, digitval;
    int numdigits;
    accumulator = numdigits = 0;
    for (;;(*pos)++, numdigits++) {
        if (*pos >= end)
            break;
        digitval = Py_UNICODE_TODECIMAL(PyUnicode_READ_CHAR(str, *pos));
        if (digitval < 0)
            break;
        /*
           Detect possible overflow before it happens:

              accumulator * 10 + digitval > PY_SSIZE_T_MAX if and only if
              accumulator > (PY_SSIZE_T_MAX - digitval) / 10.
        */
        if (accumulator > (PY_SSIZE_T_MAX - digitval) / 10) {
            PyErr_Format(PyExc_ValueError,
                         "Too many decimal digits in format string");
            return -1;
        }
        accumulator = accumulator * 10 + digitval;
    }
    *result = accumulator;
    return numdigits;
}

/************************************************************************/
/*********** standard format specifier parsing **************************/
/************************************************************************/

/* returns true if this character is a specifier alignment token */
Py_LOCAL_INLINE(int)
is_alignment_token(Py_UCS4 c)
{
    switch (c) {
    case '<': case '>': case '=': case '^':
        return 1;
    default:
        return 0;
    }
}

/* returns true if this character is a sign element */
Py_LOCAL_INLINE(int)
is_sign_element(Py_UCS4 c)
{
    switch (c) {
    case ' ': case '+': case '-':
        return 1;
    default:
        return 0;
    }
}


typedef struct {
    Py_UCS4 fill_char;
    Py_UCS4 align;
    int alternate;
    Py_UCS4 sign;
    Py_ssize_t width;
    int thousands_separators;
    Py_ssize_t precision;
    Py_UCS4 type;
} InternalFormatSpec;

#if 0
/* Occassionally useful for debugging. Should normally be commented out. */
static void
DEBUG_PRINT_FORMAT_SPEC(InternalFormatSpec *format)
{
    printf("internal format spec: fill_char %d\n", format->fill_char);
    printf("internal format spec: align %d\n", format->align);
    printf("internal format spec: alternate %d\n", format->alternate);
    printf("internal format spec: sign %d\n", format->sign);
    printf("internal format spec: width %zd\n", format->width);
    printf("internal format spec: thousands_separators %d\n",
           format->thousands_separators);
    printf("internal format spec: precision %zd\n", format->precision);
    printf("internal format spec: type %c\n", format->type);
    printf("\n");
}
#endif


/*
  ptr points to the start of the format_spec, end points just past its end.
  fills in format with the parsed information.
  returns 1 on success, 0 on failure.
  if failure, sets the exception
*/
static int
parse_internal_render_format_spec(PyObject *format_spec,
                                  Py_ssize_t start, Py_ssize_t end,
                                  InternalFormatSpec *format,
                                  char default_type,
                                  char default_align)
{
    Py_ssize_t pos = start;
    /* end-pos is used throughout this code to specify the length of
       the input string */
#define READ_spec(index) PyUnicode_READ_CHAR(format_spec, index)

    Py_ssize_t consumed;
    int align_specified = 0;

    format->fill_char = '\0';
    format->align = default_align;
    format->alternate = 0;
    format->sign = '\0';
    format->width = -1;
    format->thousands_separators = 0;
    format->precision = -1;
    format->type = default_type;

    /* If the second char is an alignment token,
       then parse the fill char */
    if (end-pos >= 2 && is_alignment_token(READ_spec(pos+1))) {
        format->align = READ_spec(pos+1);
        format->fill_char = READ_spec(pos);
        align_specified = 1;
        pos += 2;
    }
    else if (end-pos >= 1 && is_alignment_token(READ_spec(pos))) {
        format->align = READ_spec(pos);
        align_specified = 1;
        ++pos;
    }

    /* Parse the various sign options */
    if (end-pos >= 1 && is_sign_element(READ_spec(pos))) {
        format->sign = READ_spec(pos);
        ++pos;
    }

    /* If the next character is #, we're in alternate mode.  This only
       applies to integers. */
    if (end-pos >= 1 && READ_spec(pos) == '#') {
        format->alternate = 1;
        ++pos;
    }

    /* The special case for 0-padding (backwards compat) */
    if (format->fill_char == '\0' && end-pos >= 1 && READ_spec(pos) == '0') {
        format->fill_char = '0';
        if (!align_specified) {
            format->align = '=';
        }
        ++pos;
    }

    consumed = get_integer(format_spec, &pos, end, &format->width);
    if (consumed == -1)
        /* Overflow error. Exception already set. */
        return 0;

    /* If consumed is 0, we didn't consume any characters for the
       width. In that case, reset the width to -1, because
       get_integer() will have set it to zero. -1 is how we record
       that the width wasn't specified. */
    if (consumed == 0)
        format->width = -1;

    /* Comma signifies add thousands separators */
    if (end-pos && READ_spec(pos) == ',') {
        format->thousands_separators = 1;
        ++pos;
    }

    /* Parse field precision */
    if (end-pos && READ_spec(pos) == '.') {
        ++pos;

        consumed = get_integer(format_spec, &pos, end, &format->precision);
        if (consumed == -1)
            /* Overflow error. Exception already set. */
            return 0;

        /* Not having a precision after a dot is an error. */
        if (consumed == 0) {
            PyErr_Format(PyExc_ValueError,
                         "Format specifier missing precision");
            return 0;
        }

    }

    /* Finally, parse the type field. */

    if (end-pos > 1) {
        /* More than one char remain, invalid format specifier. */
        PyErr_Format(PyExc_ValueError, "Invalid format specifier");
        return 0;
    }

    if (end-pos == 1) {
        format->type = READ_spec(pos);
        ++pos;
    }

    /* Do as much validating as we can, just by looking at the format
       specifier.  Do not take into account what type of formatting
       we're doing (int, float, string). */

    if (format->thousands_separators) {
        switch (format->type) {
        case 'd':
        case 'e':
        case 'f':
        case 'g':
        case 'E':
        case 'G':
        case '%':
        case 'F':
        case '\0':
            /* These are allowed. See PEP 378.*/
            break;
        default:
            invalid_comma_type(format->type);
            return 0;
        }
    }

    assert (format->align <= 127);
    assert (format->sign <= 127);
    return 1;
}

/* Calculate the padding needed. */
static void
calc_padding(Py_ssize_t nchars, Py_ssize_t width, Py_UCS4 align,
             Py_ssize_t *n_lpadding, Py_ssize_t *n_rpadding,
             Py_ssize_t *n_total)
{
    if (width >= 0) {
        if (nchars > width)
            *n_total = nchars;
        else
            *n_total = width;
    }
    else {
        /* not specified, use all of the chars and no more */
        *n_total = nchars;
    }

    /* Figure out how much leading space we need, based on the
       aligning */
    if (align == '>')
        *n_lpadding = *n_total - nchars;
    else if (align == '^')
        *n_lpadding = (*n_total - nchars) / 2;
    else if (align == '<' || align == '=')
        *n_lpadding = 0;
    else {
        /* We should never have an unspecified alignment. */
        *n_lpadding = 0;
        assert(0);
    }

    *n_rpadding = *n_total - nchars - *n_lpadding;
}

/* Do the padding, and return a pointer to where the caller-supplied
   content goes. */
static Py_ssize_t
fill_padding(PyObject *s, Py_ssize_t start, Py_ssize_t nchars,
             Py_UCS4 fill_char, Py_ssize_t n_lpadding,
             Py_ssize_t n_rpadding)
{
    /* Pad on left. */
    if (n_lpadding)
        PyUnicode_Fill(s, start, start + n_lpadding, fill_char);

    /* Pad on right. */
    if (n_rpadding)
        PyUnicode_Fill(s, start + nchars + n_lpadding,
                       start + nchars + n_lpadding + n_rpadding, fill_char);

    /* Pointer to the user content. */
    return start + n_lpadding;
}

/************************************************************************/
/*********** common routines for numeric formatting *********************/
/************************************************************************/

/* Locale type codes. */
#define LT_CURRENT_LOCALE 0
#define LT_DEFAULT_LOCALE 1
#define LT_NO_LOCALE 2

/* Locale info needed for formatting integers and the part of floats
   before and including the decimal. Note that locales only support
   8-bit chars, not unicode. */
typedef struct {
    PyObject *decimal_point;
    PyObject *thousands_sep;
    const char *grouping;
} LocaleInfo;

#define STATIC_LOCALE_INFO_INIT {0, 0, 0}

/* describes the layout for an integer, see the comment in
   calc_number_widths() for details */
typedef struct {
    Py_ssize_t n_lpadding;
    Py_ssize_t n_prefix;
    Py_ssize_t n_spadding;
    Py_ssize_t n_rpadding;
    char sign;
    Py_ssize_t n_sign;      /* number of digits needed for sign (0/1) */
    Py_ssize_t n_grouped_digits; /* Space taken up by the digits, including
                                    any grouping chars. */
    Py_ssize_t n_decimal;   /* 0 if only an integer */
    Py_ssize_t n_remainder; /* Digits in decimal and/or exponent part,
                               excluding the decimal itself, if
                               present. */

    /* These 2 are not the widths of fields, but are needed by
       STRINGLIB_GROUPING. */
    Py_ssize_t n_digits;    /* The number of digits before a decimal
                               or exponent. */
    Py_ssize_t n_min_width; /* The min_width we used when we computed
                               the n_grouped_digits width. */
} NumberFieldWidths;


/* Given a number of the form:
   digits[remainder]
   where ptr points to the start and end points to the end, find where
    the integer part ends. This could be a decimal, an exponent, both,
    or neither.
   If a decimal point is present, set *has_decimal and increment
    remainder beyond it.
   Results are undefined (but shouldn't crash) for improperly
    formatted strings.
*/
static void
parse_number(PyObject *s, Py_ssize_t pos, Py_ssize_t end,
             Py_ssize_t *n_remainder, int *has_decimal)
{
    Py_ssize_t remainder;

    while (pos<end && isdigit(PyUnicode_READ_CHAR(s, pos)))
        ++pos;
    remainder = pos;

    /* Does remainder start with a decimal point? */
    *has_decimal = pos<end && PyUnicode_READ_CHAR(s, remainder) == '.';

    /* Skip the decimal point. */
    if (*has_decimal)
        remainder++;

    *n_remainder = end - remainder;
}

/* not all fields of format are used.  for example, precision is
   unused.  should this take discrete params in order to be more clear
   about what it does?  or is passing a single format parameter easier
   and more efficient enough to justify a little obfuscation? */
static Py_ssize_t
calc_number_widths(NumberFieldWidths *spec, Py_ssize_t n_prefix,
                   Py_UCS4 sign_char, PyObject *number, Py_ssize_t n_start,
                   Py_ssize_t n_end, Py_ssize_t n_remainder,
                   int has_decimal, const LocaleInfo *locale,
                   const InternalFormatSpec *format, Py_UCS4 *maxchar)
{
    Py_ssize_t n_non_digit_non_padding;
    Py_ssize_t n_padding;

    spec->n_digits = n_end - n_start - n_remainder - (has_decimal?1:0);
    spec->n_lpadding = 0;
    spec->n_prefix = n_prefix;
    spec->n_decimal = has_decimal ? PyUnicode_GET_LENGTH(locale->decimal_point) : 0;
    spec->n_remainder = n_remainder;
    spec->n_spadding = 0;
    spec->n_rpadding = 0;
    spec->sign = '\0';
    spec->n_sign = 0;

    /* the output will look like:
       |                                                                                         |
       | <lpadding> <sign> <prefix> <spadding> <grouped_digits> <decimal> <remainder> <rpadding> |
       |                                                                                         |

       sign is computed from format->sign and the actual
       sign of the number

       prefix is given (it's for the '0x' prefix)

       digits is already known

       the total width is either given, or computed from the
       actual digits

       only one of lpadding, spadding, and rpadding can be non-zero,
       and it's calculated from the width and other fields
    */

    /* compute the various parts we're going to write */
    switch (format->sign) {
    case '+':
        /* always put a + or - */
        spec->n_sign = 1;
        spec->sign = (sign_char == '-' ? '-' : '+');
        break;
    case ' ':
        spec->n_sign = 1;
        spec->sign = (sign_char == '-' ? '-' : ' ');
        break;
    default:
        /* Not specified, or the default (-) */
        if (sign_char == '-') {
            spec->n_sign = 1;
            spec->sign = '-';
        }
    }

    /* The number of chars used for non-digits and non-padding. */
    n_non_digit_non_padding = spec->n_sign + spec->n_prefix + spec->n_decimal +
        spec->n_remainder;

    /* min_width can go negative, that's okay. format->width == -1 means
       we don't care. */
    if (format->fill_char == '0' && format->align == '=')
        spec->n_min_width = format->width - n_non_digit_non_padding;
    else
        spec->n_min_width = 0;

    if (spec->n_digits == 0)
        /* This case only occurs when using 'c' formatting, we need
           to special case it because the grouping code always wants
           to have at least one character. */
        spec->n_grouped_digits = 0;
    else {
        Py_UCS4 grouping_maxchar;
        spec->n_grouped_digits = _PyUnicode_InsertThousandsGrouping(
            NULL, 0,
            0, NULL,
            spec->n_digits, spec->n_min_width,
            locale->grouping, locale->thousands_sep, &grouping_maxchar);
        *maxchar = Py_MAX(*maxchar, grouping_maxchar);
    }

    /* Given the desired width and the total of digit and non-digit
       space we consume, see if we need any padding. format->width can
       be negative (meaning no padding), but this code still works in
       that case. */
    n_padding = format->width -
                        (n_non_digit_non_padding + spec->n_grouped_digits);
    if (n_padding > 0) {
        /* Some padding is needed. Determine if it's left, space, or right. */
        switch (format->align) {
        case '<':
            spec->n_rpadding = n_padding;
            break;
        case '^':
            spec->n_lpadding = n_padding / 2;
            spec->n_rpadding = n_padding - spec->n_lpadding;
            break;
        case '=':
            spec->n_spadding = n_padding;
            break;
        case '>':
            spec->n_lpadding = n_padding;
            break;
        default:
            /* Shouldn't get here, but treat it as '>' */
            spec->n_lpadding = n_padding;
            assert(0);
            break;
        }
    }

    if (spec->n_lpadding || spec->n_spadding || spec->n_rpadding)
        *maxchar = Py_MAX(*maxchar, format->fill_char);

    if (spec->n_decimal)
        *maxchar = Py_MAX(*maxchar, PyUnicode_MAX_CHAR_VALUE(locale->decimal_point));

    return spec->n_lpadding + spec->n_sign + spec->n_prefix +
        spec->n_spadding + spec->n_grouped_digits + spec->n_decimal +
        spec->n_remainder + spec->n_rpadding;
}

/* Fill in the digit parts of a numbers's string representation,
   as determined in calc_number_widths().
   Return -1 on error, or 0 on success. */
static int
fill_number(PyObject *out, Py_ssize_t pos, const NumberFieldWidths *spec,
            PyObject *digits, Py_ssize_t d_start, Py_ssize_t d_end,
            PyObject *prefix, Py_ssize_t p_start,
            Py_UCS4 fill_char,
            LocaleInfo *locale, int toupper)
{
    /* Used to keep track of digits, decimal, and remainder. */
    Py_ssize_t d_pos = d_start;
    unsigned int kind = PyUnicode_KIND(out);
    void *data = PyUnicode_DATA(out);
    Py_ssize_t r;

    if (spec->n_lpadding) {
        PyUnicode_Fill(out, pos, pos + spec->n_lpadding, fill_char);
        pos += spec->n_lpadding;
    }
    if (spec->n_sign == 1) {
        PyUnicode_WRITE(kind, data, pos++, spec->sign);
    }
    if (spec->n_prefix) {
        if (PyUnicode_CopyCharacters(out, pos,
                                     prefix, p_start,
                                     spec->n_prefix) < 0)
            return -1;
        if (toupper) {
            Py_ssize_t t;
            for (t = 0; t < spec->n_prefix; t++) {
                Py_UCS4 c = PyUnicode_READ(kind, data, pos + t);
                c = Py_TOUPPER(c);
                assert (c <= 127);
                PyUnicode_WRITE(kind, data, pos + t, c);
            }
        }
        pos += spec->n_prefix;
    }
    if (spec->n_spadding) {
        PyUnicode_Fill(out, pos, pos + spec->n_spadding, fill_char);
        pos += spec->n_spadding;
    }

    /* Only for type 'c' special case, it has no digits. */
    if (spec->n_digits != 0) {
        /* Fill the digits with InsertThousandsGrouping. */
        char *pdigits;
        if (PyUnicode_READY(digits))
            return -1;
        pdigits = PyUnicode_DATA(digits);
        if (PyUnicode_KIND(digits) < kind) {
            pdigits = _PyUnicode_AsKind(digits, kind);
            if (pdigits == NULL)
                return -1;
        }
        r = _PyUnicode_InsertThousandsGrouping(
                out, pos,
                spec->n_grouped_digits,
                pdigits + kind * d_pos,
                spec->n_digits, spec->n_min_width,
                locale->grouping, locale->thousands_sep, NULL);
        if (r == -1)
            return -1;
        assert(r == spec->n_grouped_digits);
        if (PyUnicode_KIND(digits) < kind)
            PyMem_Free(pdigits);
        d_pos += spec->n_digits;
    }
    if (toupper) {
        Py_ssize_t t;
        for (t = 0; t < spec->n_grouped_digits; t++) {
            Py_UCS4 c = PyUnicode_READ(kind, data, pos + t);
            c = Py_TOUPPER(c);
            if (c > 127) {
                PyErr_SetString(PyExc_SystemError, "non-ascii grouped digit");
                return -1;
            }
            PyUnicode_WRITE(kind, data, pos + t, c);
        }
    }
    pos += spec->n_grouped_digits;

    if (spec->n_decimal) {
        if (PyUnicode_CopyCharacters(out, pos, locale->decimal_point, 0, spec->n_decimal) < 0)
            return -1;
        pos += spec->n_decimal;
        d_pos += 1;
    }

    if (spec->n_remainder) {
        if (PyUnicode_CopyCharacters(out, pos, digits, d_pos, spec->n_remainder) < 0)
            return -1;
        pos += spec->n_remainder;
        d_pos += spec->n_remainder;
    }

    if (spec->n_rpadding) {
        PyUnicode_Fill(out, pos, pos + spec->n_rpadding, fill_char);
        pos += spec->n_rpadding;
    }
    return 0;
}

static char no_grouping[1] = {CHAR_MAX};

/* Find the decimal point character(s?), thousands_separator(s?), and
   grouping description, either for the current locale if type is
   LT_CURRENT_LOCALE, a hard-coded locale if LT_DEFAULT_LOCALE, or
   none if LT_NO_LOCALE. */
static int
get_locale_info(int type, LocaleInfo *locale_info)
{
    switch (type) {
    case LT_CURRENT_LOCALE: {
        struct lconv *locale_data = localeconv();
        locale_info->decimal_point = PyUnicode_DecodeLocale(
                                         locale_data->decimal_point,
                                         NULL);
        if (locale_info->decimal_point == NULL)
            return -1;
        locale_info->thousands_sep = PyUnicode_DecodeLocale(
                                         locale_data->thousands_sep,
                                         NULL);
        if (locale_info->thousands_sep == NULL) {
            Py_DECREF(locale_info->decimal_point);
            return -1;
        }
        locale_info->grouping = locale_data->grouping;
        break;
    }
    case LT_DEFAULT_LOCALE:
        locale_info->decimal_point = PyUnicode_FromOrdinal('.');
        locale_info->thousands_sep = PyUnicode_FromOrdinal(',');
        if (!locale_info->decimal_point || !locale_info->thousands_sep) {
            Py_XDECREF(locale_info->decimal_point);
            Py_XDECREF(locale_info->thousands_sep);
            return -1;
        }
        locale_info->grouping = "\3"; /* Group every 3 characters.  The
                                         (implicit) trailing 0 means repeat
                                         infinitely. */
        break;
    case LT_NO_LOCALE:
        locale_info->decimal_point = PyUnicode_FromOrdinal('.');
        locale_info->thousands_sep = PyUnicode_New(0, 0);
        if (!locale_info->decimal_point || !locale_info->thousands_sep) {
            Py_XDECREF(locale_info->decimal_point);
            Py_XDECREF(locale_info->thousands_sep);
            return -1;
        }
        locale_info->grouping = no_grouping;
        break;
    default:
        assert(0);
    }
    return 0;
}

static void
free_locale_info(LocaleInfo *locale_info)
{
    Py_XDECREF(locale_info->decimal_point);
    Py_XDECREF(locale_info->thousands_sep);
}

/************************************************************************/
/*********** string formatting ******************************************/
/************************************************************************/

static PyObject *
format_string_internal(PyObject *value, const InternalFormatSpec *format)
{
    Py_ssize_t lpad;
    Py_ssize_t rpad;
    Py_ssize_t total;
    Py_ssize_t pos;
    Py_ssize_t len = PyUnicode_GET_LENGTH(value);
    PyObject *result = NULL;
    Py_UCS4 maxchar = 127;

    /* sign is not allowed on strings */
    if (format->sign != '\0') {
        PyErr_SetString(PyExc_ValueError,
                        "Sign not allowed in string format specifier");
        goto done;
    }

    /* alternate is not allowed on strings */
    if (format->alternate) {
        PyErr_SetString(PyExc_ValueError,
                        "Alternate form (#) not allowed in string format "
                        "specifier");
        goto done;
    }

    /* '=' alignment not allowed on strings */
    if (format->align == '=') {
        PyErr_SetString(PyExc_ValueError,
                        "'=' alignment not allowed "
                        "in string format specifier");
        goto done;
    }

    /* if precision is specified, output no more that format.precision
       characters */
    if (format->precision >= 0 && len >= format->precision) {
        len = format->precision;
    }

    calc_padding(len, format->width, format->align, &lpad, &rpad, &total);

    if (lpad != 0 || rpad != 0)
        maxchar = Py_MAX(maxchar, format->fill_char);

    /* allocate the resulting string */
    result = PyUnicode_New(total, maxchar);
    if (result == NULL)
        goto done;

    /* Write into that space. First the padding. */
    pos = fill_padding(result, 0, len,
                       format->fill_char=='\0'?' ':format->fill_char,
                       lpad, rpad);

    /* Then the source string. */
    if (PyUnicode_CopyCharacters(result, pos, value, 0, len) < 0)
        Py_CLEAR(result);

done:
    assert(!result || _PyUnicode_CheckConsistency(result, 1));
    return result;
}


/************************************************************************/
/*********** long formatting ********************************************/
/************************************************************************/

typedef PyObject*
(*IntOrLongToString)(PyObject *value, int base);

static PyObject *
format_int_or_long_internal(PyObject *value, const InternalFormatSpec *format,
                            IntOrLongToString tostring)
{
    PyObject *result = NULL;
    Py_UCS4 maxchar = 127;
    PyObject *tmp = NULL;
    Py_ssize_t inumeric_chars;
    Py_UCS4 sign_char = '\0';
    Py_ssize_t n_digits;       /* count of digits need from the computed
                                  string */
    Py_ssize_t n_remainder = 0; /* Used only for 'c' formatting, which
                                   produces non-digits */
    Py_ssize_t n_prefix = 0;   /* Count of prefix chars, (e.g., '0x') */
    Py_ssize_t n_total;
    Py_ssize_t prefix = 0;
    NumberFieldWidths spec;
    long x;
    int err;

    /* Locale settings, either from the actual locale or
       from a hard-code pseudo-locale */
    LocaleInfo locale = STATIC_LOCALE_INFO_INIT;

    /* no precision allowed on integers */
    if (format->precision != -1) {
        PyErr_SetString(PyExc_ValueError,
                        "Precision not allowed in integer format specifier");
        goto done;
    }

    /* special case for character formatting */
    if (format->type == 'c') {
        /* error to specify a sign */
        if (format->sign != '\0') {
            PyErr_SetString(PyExc_ValueError,
                            "Sign not allowed with integer"
                            " format specifier 'c'");
            goto done;
        }

        /* taken from unicodeobject.c formatchar() */
        /* Integer input truncated to a character */
        x = PyLong_AsLong(value);
        if (x == -1 && PyErr_Occurred())
            goto done;
        if (x < 0 || x > 0x10ffff) {
            PyErr_SetString(PyExc_OverflowError,
                            "%c arg not in range(0x110000)");
            goto done;
        }
        tmp = PyUnicode_FromOrdinal(x);
        inumeric_chars = 0;
        n_digits = 1;
        maxchar = Py_MAX(maxchar, (Py_UCS4)x);

        /* As a sort-of hack, we tell calc_number_widths that we only
           have "remainder" characters. calc_number_widths thinks
           these are characters that don't get formatted, only copied
           into the output string. We do this for 'c' formatting,
           because the characters are likely to be non-digits. */
        n_remainder = 1;
    }
    else {
        int base;
        int leading_chars_to_skip = 0;  /* Number of characters added by
                                           PyNumber_ToBase that we want to
                                           skip over. */

        /* Compute the base and how many characters will be added by
           PyNumber_ToBase */
        switch (format->type) {
        case 'b':
            base = 2;
            leading_chars_to_skip = 2; /* 0b */
            break;
        case 'o':
            base = 8;
            leading_chars_to_skip = 2; /* 0o */
            break;
        case 'x':
        case 'X':
            base = 16;
            leading_chars_to_skip = 2; /* 0x */
            break;
        default:  /* shouldn't be needed, but stops a compiler warning */
        case 'd':
        case 'n':
            base = 10;
            break;
        }

        /* The number of prefix chars is the same as the leading
           chars to skip */
        if (format->alternate)
            n_prefix = leading_chars_to_skip;

        /* Do the hard part, converting to a string in a given base */
        tmp = tostring(value, base);
        if (tmp == NULL || PyUnicode_READY(tmp) == -1)
            goto done;

        inumeric_chars = 0;
        n_digits = PyUnicode_GET_LENGTH(tmp);

        prefix = inumeric_chars;

        /* Is a sign character present in the output?  If so, remember it
           and skip it */
        if (PyUnicode_READ_CHAR(tmp, inumeric_chars) == '-') {
            sign_char = '-';
            ++prefix;
            ++leading_chars_to_skip;
        }

        /* Skip over the leading chars (0x, 0b, etc.) */
        n_digits -= leading_chars_to_skip;
        inumeric_chars += leading_chars_to_skip;
    }

    /* Determine the grouping, separator, and decimal point, if any. */
    if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
                        (format->thousands_separators ?
                         LT_DEFAULT_LOCALE :
                         LT_NO_LOCALE),
                        &locale) == -1)
        goto done;

    /* Calculate how much memory we'll need. */
    n_total = calc_number_widths(&spec, n_prefix, sign_char, tmp, inumeric_chars,
                                 inumeric_chars + n_digits, n_remainder, 0,
                                 &locale, format, &maxchar);

    /* Allocate the memory. */
    result = PyUnicode_New(n_total, maxchar);
    if (!result)
        goto done;

    /* Populate the memory. */
    err = fill_number(result, 0, &spec,
                      tmp, inumeric_chars, inumeric_chars + n_digits,
                      tmp, prefix,
                      format->fill_char == '\0' ? ' ' : format->fill_char,
                      &locale, format->type == 'X');
    if (err)
        Py_CLEAR(result);

done:
    Py_XDECREF(tmp);
    free_locale_info(&locale);
    assert(!result || _PyUnicode_CheckConsistency(result, 1));
    return result;
}

/************************************************************************/
/*********** float formatting *******************************************/
/************************************************************************/

static PyObject*
strtounicode(char *charbuffer, Py_ssize_t len)
{
    return PyUnicode_FromKindAndData(PyUnicode_1BYTE_KIND, charbuffer, len);
}

/* much of this is taken from unicodeobject.c */
static PyObject *
format_float_internal(PyObject *value,
                      const InternalFormatSpec *format)
{
    char *buf = NULL;       /* buffer returned from PyOS_double_to_string */
    Py_ssize_t n_digits;
    Py_ssize_t n_remainder;
    Py_ssize_t n_total;
    int has_decimal;
    double val;
    Py_ssize_t precision = format->precision;
    Py_ssize_t default_precision = 6;
    Py_UCS4 type = format->type;
    int add_pct = 0;
    Py_ssize_t index;
    NumberFieldWidths spec;
    int flags = 0;
    PyObject *result = NULL;
    Py_UCS4 maxchar = 127;
    Py_UCS4 sign_char = '\0';
    int float_type; /* Used to see if we have a nan, inf, or regular float. */
    PyObject *unicode_tmp = NULL;
    int err;

    /* Locale settings, either from the actual locale or
       from a hard-code pseudo-locale */
    LocaleInfo locale = STATIC_LOCALE_INFO_INIT;

    if (format->alternate)
        flags |= Py_DTSF_ALT;

    if (type == '\0') {
        /* Omitted type specifier.  Behaves in the same way as repr(x)
           and str(x) if no precision is given, else like 'g', but with
           at least one digit after the decimal point. */
        flags |= Py_DTSF_ADD_DOT_0;
        type = 'r';
        default_precision = 0;
    }

    if (type == 'n')
        /* 'n' is the same as 'g', except for the locale used to
           format the result. We take care of that later. */
        type = 'g';

    val = PyFloat_AsDouble(value);
    if (val == -1.0 && PyErr_Occurred())
        goto done;

    if (type == '%') {
        type = 'f';
        val *= 100;
        add_pct = 1;
    }

    if (precision < 0)
        precision = default_precision;
    else if (type == 'r')
        type = 'g';

    /* Cast "type", because if we're in unicode we need to pass a
       8-bit char. This is safe, because we've restricted what "type"
       can be. */
    buf = PyOS_double_to_string(val, (char)type, precision, flags,
                                &float_type);
    if (buf == NULL)
        goto done;
    n_digits = strlen(buf);

    if (add_pct) {
        /* We know that buf has a trailing zero (since we just called
           strlen() on it), and we don't use that fact any more. So we
           can just write over the trailing zero. */
        buf[n_digits] = '%';
        n_digits += 1;
    }

    /* Since there is no unicode version of PyOS_double_to_string,
       just use the 8 bit version and then convert to unicode. */
    unicode_tmp = strtounicode(buf, n_digits);
    if (unicode_tmp == NULL)
        goto done;
    index = 0;

    /* Is a sign character present in the output?  If so, remember it
       and skip it */
    if (PyUnicode_READ_CHAR(unicode_tmp, index) == '-') {
        sign_char = '-';
        ++index;
        --n_digits;
    }

    /* Determine if we have any "remainder" (after the digits, might include
       decimal or exponent or both (or neither)) */
    parse_number(unicode_tmp, index, index + n_digits, &n_remainder, &has_decimal);

    /* Determine the grouping, separator, and decimal point, if any. */
    if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
                        (format->thousands_separators ?
                         LT_DEFAULT_LOCALE :
                         LT_NO_LOCALE),
                        &locale) == -1)
        goto done;

    /* Calculate how much memory we'll need. */
    n_total = calc_number_widths(&spec, 0, sign_char, unicode_tmp, index,
                                 index + n_digits, n_remainder, has_decimal,
                                 &locale, format, &maxchar);

    /* Allocate the memory. */
    result = PyUnicode_New(n_total, maxchar);
    if (result == NULL)
        goto done;

    /* Populate the memory. */
    err = fill_number(result, 0, &spec,
                      unicode_tmp, index, index + n_digits,
                      NULL, 0,
                      format->fill_char == '\0' ? ' ' : format->fill_char,
                      &locale, 0);
    if (err)
        Py_CLEAR(result);

done:
    PyMem_Free(buf);
    Py_DECREF(unicode_tmp);
    free_locale_info(&locale);
    assert(!result || _PyUnicode_CheckConsistency(result, 1));
    return result;
}

/************************************************************************/
/*********** complex formatting *****************************************/
/************************************************************************/

static PyObject *
format_complex_internal(PyObject *value,
                        const InternalFormatSpec *format)
{
    double re;
    double im;
    char *re_buf = NULL;       /* buffer returned from PyOS_double_to_string */
    char *im_buf = NULL;       /* buffer returned from PyOS_double_to_string */

    InternalFormatSpec tmp_format = *format;
    Py_ssize_t n_re_digits;
    Py_ssize_t n_im_digits;
    Py_ssize_t n_re_remainder;
    Py_ssize_t n_im_remainder;
    Py_ssize_t n_re_total;
    Py_ssize_t n_im_total;
    int re_has_decimal;
    int im_has_decimal;
    Py_ssize_t precision = format->precision;
    Py_ssize_t default_precision = 6;
    Py_UCS4 type = format->type;
    Py_ssize_t i_re;
    Py_ssize_t i_im;
    NumberFieldWidths re_spec;
    NumberFieldWidths im_spec;
    int flags = 0;
    PyObject *result = NULL;
    Py_UCS4 maxchar = 127;
    int rkind;
    void *rdata;
    Py_ssize_t index;
    Py_UCS4 re_sign_char = '\0';
    Py_UCS4 im_sign_char = '\0';
    int re_float_type; /* Used to see if we have a nan, inf, or regular float. */
    int im_float_type;
    int add_parens = 0;
    int skip_re = 0;
    Py_ssize_t lpad;
    Py_ssize_t rpad;
    Py_ssize_t total;
    PyObject *re_unicode_tmp = NULL;
    PyObject *im_unicode_tmp = NULL;
    int err;

    /* Locale settings, either from the actual locale or
       from a hard-code pseudo-locale */
    LocaleInfo locale = STATIC_LOCALE_INFO_INIT;

    /* Zero padding is not allowed. */
    if (format->fill_char == '0') {
        PyErr_SetString(PyExc_ValueError,
                        "Zero padding is not allowed in complex format "
                        "specifier");
        goto done;
    }

    /* Neither is '=' alignment . */
    if (format->align == '=') {
        PyErr_SetString(PyExc_ValueError,
                        "'=' alignment flag is not allowed in complex format "
                        "specifier");
        goto done;
    }

    re = PyComplex_RealAsDouble(value);
    if (re == -1.0 && PyErr_Occurred())
        goto done;
    im = PyComplex_ImagAsDouble(value);
    if (im == -1.0 && PyErr_Occurred())
        goto done;

    if (format->alternate)
        flags |= Py_DTSF_ALT;

    if (type == '\0') {
        /* Omitted type specifier. Should be like str(self). */
        type = 'r';
        default_precision = 0;
        if (re == 0.0 && copysign(1.0, re) == 1.0)
            skip_re = 1;
        else
            add_parens = 1;
    }

    if (type == 'n')
        /* 'n' is the same as 'g', except for the locale used to
           format the result. We take care of that later. */
        type = 'g';

    if (precision < 0)
        precision = default_precision;
    else if (type == 'r')
        type = 'g';

    /* Cast "type", because if we're in unicode we need to pass a
       8-bit char. This is safe, because we've restricted what "type"
       can be. */
    re_buf = PyOS_double_to_string(re, (char)type, precision, flags,
                                   &re_float_type);
    if (re_buf == NULL)
        goto done;
    im_buf = PyOS_double_to_string(im, (char)type, precision, flags,
                                   &im_float_type);
    if (im_buf == NULL)
        goto done;

    n_re_digits = strlen(re_buf);
    n_im_digits = strlen(im_buf);

    /* Since there is no unicode version of PyOS_double_to_string,
       just use the 8 bit version and then convert to unicode. */
    re_unicode_tmp = strtounicode(re_buf, n_re_digits);
    if (re_unicode_tmp == NULL)
        goto done;
    i_re = 0;

    im_unicode_tmp = strtounicode(im_buf, n_im_digits);
    if (im_unicode_tmp == NULL)
        goto done;
    i_im = 0;

    /* Is a sign character present in the output?  If so, remember it
       and skip it */
    if (PyUnicode_READ_CHAR(re_unicode_tmp, i_re) == '-') {
        re_sign_char = '-';
        ++i_re;
        --n_re_digits;
    }
    if (PyUnicode_READ_CHAR(im_unicode_tmp, i_im) == '-') {
        im_sign_char = '-';
        ++i_im;
        --n_im_digits;
    }

    /* Determine if we have any "remainder" (after the digits, might include
       decimal or exponent or both (or neither)) */
    parse_number(re_unicode_tmp, i_re, i_re + n_re_digits,
                 &n_re_remainder, &re_has_decimal);
    parse_number(im_unicode_tmp, i_im, i_im + n_im_digits,
                 &n_im_remainder, &im_has_decimal);

    /* Determine the grouping, separator, and decimal point, if any. */
    if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
                        (format->thousands_separators ?
                         LT_DEFAULT_LOCALE :
                         LT_NO_LOCALE),
                        &locale) == -1)
        goto done;

    /* Turn off any padding. We'll do it later after we've composed
       the numbers without padding. */
    tmp_format.fill_char = '\0';
    tmp_format.align = '<';
    tmp_format.width = -1;

    /* Calculate how much memory we'll need. */
    n_re_total = calc_number_widths(&re_spec, 0, re_sign_char, re_unicode_tmp,
                                    i_re, i_re + n_re_digits, n_re_remainder,
                                    re_has_decimal, &locale, &tmp_format,
                                    &maxchar);

    /* Same formatting, but always include a sign, unless the real part is
     * going to be omitted, in which case we use whatever sign convention was
     * requested by the original format. */
    if (!skip_re)
        tmp_format.sign = '+';
    n_im_total = calc_number_widths(&im_spec, 0, im_sign_char, im_unicode_tmp,
                                    i_im, i_im + n_im_digits, n_im_remainder,
                                    im_has_decimal, &locale, &tmp_format,
                                    &maxchar);

    if (skip_re)
        n_re_total = 0;

    /* Add 1 for the 'j', and optionally 2 for parens. */
    calc_padding(n_re_total + n_im_total + 1 + add_parens * 2,
                 format->width, format->align, &lpad, &rpad, &total);

    if (lpad || rpad)
        maxchar = Py_MAX(maxchar, format->fill_char);

    result = PyUnicode_New(total, maxchar);
    if (result == NULL)
        goto done;
    rkind = PyUnicode_KIND(result);
    rdata = PyUnicode_DATA(result);

    /* Populate the memory. First, the padding. */
    index = fill_padding(result, 0,
                         n_re_total + n_im_total + 1 + add_parens * 2,
                         format->fill_char=='\0' ? ' ' : format->fill_char,
                         lpad, rpad);

    if (add_parens)
        PyUnicode_WRITE(rkind, rdata, index++, '(');

    if (!skip_re) {
        err = fill_number(result, index, &re_spec,
                          re_unicode_tmp, i_re, i_re + n_re_digits,
                          NULL, 0,
                          0,
                          &locale, 0);
        if (err) {
            Py_CLEAR(result);
            goto done;
        }
        index += n_re_total;
    }
    err = fill_number(result, index, &im_spec,
                      im_unicode_tmp, i_im, i_im + n_im_digits,
                      NULL, 0,
                      0,
                      &locale, 0);
    if (err) {
        Py_CLEAR(result);
        goto done;
    }
    index += n_im_total;
    PyUnicode_WRITE(rkind, rdata, index++, 'j');

    if (add_parens)
        PyUnicode_WRITE(rkind, rdata, index++, ')');

done:
    PyMem_Free(re_buf);
    PyMem_Free(im_buf);
    Py_XDECREF(re_unicode_tmp);
    Py_XDECREF(im_unicode_tmp);
    free_locale_info(&locale);
    assert(!result || _PyUnicode_CheckConsistency(result, 1));
    return result;
}

/************************************************************************/
/*********** built in formatters ****************************************/
/************************************************************************/
PyObject *
_PyUnicode_FormatAdvanced(PyObject *obj,
                          PyObject *format_spec,
                          Py_ssize_t start, Py_ssize_t end)
{
    InternalFormatSpec format;
    PyObject *result;

    /* check for the special case of zero length format spec, make
       it equivalent to str(obj) */
    if (start == end)
        return PyObject_Str(obj);

    /* parse the format_spec */
    if (!parse_internal_render_format_spec(format_spec, start, end,
                                           &format, 's', '<'))
        return NULL;

    /* type conversion? */
    switch (format.type) {
    case 's':
        /* no type conversion needed, already a string.  do the formatting */
        result = format_string_internal(obj, &format);
        if (result != NULL)
            assert(_PyUnicode_CheckConsistency(result, 1));
        break;
    default:
        /* unknown */
        unknown_presentation_type(format.type, obj->ob_type->tp_name);
        result = NULL;
    }
    return result;
}

static PyObject*
format_int_or_long(PyObject* obj, PyObject* format_spec,
                   Py_ssize_t start, Py_ssize_t end,
                   IntOrLongToString tostring)
{
    PyObject *result = NULL;
    PyObject *tmp = NULL;
    InternalFormatSpec format;

    /* check for the special case of zero length format spec, make
       it equivalent to str(obj) */
    if (start == end) {
        result = PyObject_Str(obj);
        goto done;
    }

    /* parse the format_spec */
    if (!parse_internal_render_format_spec(format_spec, start, end,
                                           &format, 'd', '>'))
        goto done;

    /* type conversion? */
    switch (format.type) {
    case 'b':
    case 'c':
    case 'd':
    case 'o':
    case 'x':
    case 'X':
    case 'n':
        /* no type conversion needed, already an int (or long).  do
           the formatting */
            result = format_int_or_long_internal(obj, &format, tostring);
        break;

    case 'e':
    case 'E':
    case 'f':
    case 'F':
    case 'g':
    case 'G':
    case '%':
        /* convert to float */
        tmp = PyNumber_Float(obj);
        if (tmp == NULL)
            goto done;
        result = format_float_internal(tmp, &format);
        break;

    default:
        /* unknown */
        unknown_presentation_type(format.type, obj->ob_type->tp_name);
        goto done;
    }

done:
    Py_XDECREF(tmp);
    return result;
}

/* Need to define long_format as a function that will convert a long
   to a string.  In 3.0, _PyLong_Format has the correct signature. */
#define long_format _PyLong_Format

PyObject *
_PyLong_FormatAdvanced(PyObject *obj,
                       PyObject *format_spec,
                       Py_ssize_t start, Py_ssize_t end)
{
    return format_int_or_long(obj, format_spec, start, end,
                              long_format);
}

PyObject *
_PyFloat_FormatAdvanced(PyObject *obj,
                        PyObject *format_spec,
                        Py_ssize_t start, Py_ssize_t end)
{
    PyObject *result = NULL;
    InternalFormatSpec format;

    /* check for the special case of zero length format spec, make
       it equivalent to str(obj) */
    if (start == end) {
        result = PyObject_Str(obj);
        goto done;
    }

    /* parse the format_spec */
    if (!parse_internal_render_format_spec(format_spec, start, end,
                                           &format, '\0', '>'))
        goto done;

    /* type conversion? */
    switch (format.type) {
    case '\0': /* No format code: like 'g', but with at least one decimal. */
    case 'e':
    case 'E':
    case 'f':
    case 'F':
    case 'g':
    case 'G':
    case 'n':
    case '%':
        /* no conversion, already a float.  do the formatting */
        result = format_float_internal(obj, &format);
        break;

    default:
        /* unknown */
        unknown_presentation_type(format.type, obj->ob_type->tp_name);
        goto done;
    }

done:
    return result;
}

PyObject *
_PyComplex_FormatAdvanced(PyObject *obj,
                          PyObject *format_spec,
                          Py_ssize_t start, Py_ssize_t end)
{
    PyObject *result = NULL;
    InternalFormatSpec format;

    /* check for the special case of zero length format spec, make
       it equivalent to str(obj) */
    if (start == end) {
        result = PyObject_Str(obj);
        goto done;
    }

    /* parse the format_spec */
    if (!parse_internal_render_format_spec(format_spec, start, end,
                                           &format, '\0', '>'))
        goto done;

    /* type conversion? */
    switch (format.type) {
    case '\0': /* No format code: like 'g', but with at least one decimal. */
    case 'e':
    case 'E':
    case 'f':
    case 'F':
    case 'g':
    case 'G':
    case 'n':
        /* no conversion, already a complex.  do the formatting */
        result = format_complex_internal(obj, &format);
        break;

    default:
        /* unknown */
        unknown_presentation_type(format.type, obj->ob_type->tp_name);
        goto done;
    }

done:
    return result;
}
Tip: Filter by directory path e.g. /media app.js to search for public/media/app.js.
Tip: Use camelCasing e.g. ProjME to search for ProjectModifiedEvent.java.
Tip: Filter by extension type e.g. /repo .js to search for all .js files in the /repo directory.
Tip: Separate your search with spaces e.g. /ssh pom.xml to search for src/ssh/pom.xml.
Tip: Use ↑ and ↓ arrow keys to navigate and return to view the file.
Tip: You can also navigate files with Ctrl+j (next) and Ctrl+k (previous) and view the file with Ctrl+o.
Tip: You can also navigate files with Alt+j (next) and Alt+k (previous) and view the file with Alt+o.