Source

python-clinic / Tools / unicode / mkstringprep.py

Full commit
  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
import re, sys
from unicodedata import ucd_3_2_0 as unicodedata

if sys.maxunicode == 65535:
    raise RuntimeError("need UCS-4 Python")

def gen_category(cats):
    for i in range(0, 0x110000):
        if unicodedata.category(chr(i)) in cats:
            yield(i)

def gen_bidirectional(cats):
    for i in range(0, 0x110000):
        if unicodedata.bidirectional(chr(i)) in cats:
            yield(i)

def compact_set(l):
    single = []
    tuple = []
    prev = None
    span = 0
    for e in l:
        if prev is None:
            prev = e
            span = 0
            continue
        if prev+span+1 != e:
            if span > 2:
                tuple.append((prev,prev+span+1))
            else:
                for i in range(prev, prev+span+1):
                    single.append(i)
            prev = e
            span = 0
        else:
            span += 1
    if span:
        tuple.append((prev,prev+span+1))
    else:
        single.append(prev)
    if not single and len(tuple) == 1:
        tuple = "range(%d,%d)" % tuple[0]
    else:
        tuple = " + ".join("list(range(%d,%d))" % t for t in tuple)
    if not single:
        return "set(%s)" % tuple
    if not tuple:
        return "set(%r)" % (single,)
    return "set(%r + %s)" % (single, tuple)

############## Read the tables in the RFC #######################

with open("rfc3454.txt") as f:
    data = f.readlines()

tables = []
curname = None
for l in data:
    l = l.strip()
    if not l:
        continue
    # Skip RFC page breaks
    if l.startswith(("Hoffman & Blanchet", "RFC 3454")):
        continue
    # Find start/end lines
    m = re.match("----- (Start|End) Table ([A-Z](.[0-9])+) -----", l)
    if m:
        if m.group(1) == "Start":
            if curname:
                raise RuntimeError("Double Start", (curname, l))
            curname = m.group(2)
            table = {}
            tables.append((curname, table))
            continue
        else:
            if not curname:
                raise RuntimeError("End without start", l)
            if curname != m.group(2):
                raise RuntimeError("Unexpected end", l)
            curname = None
            continue
    if not curname:
        continue
    # Now we are in a table
    fields = l.split(";")
    if len(fields) > 1:
        # Drop comment field
        fields = fields[:-1]
    if len(fields) == 1:
        fields = fields[0].split("-")
        if len(fields) > 1:
            # range
            try:
                start, end = fields
            except ValueError:
                raise RuntimeError("Unpacking problem", l)
        else:
            start = end = fields[0]
        start = int(start, 16)
        end = int(end, 16)
        for i in range(start, end+1):
            table[i] = i
    else:
        code, value = fields
        value = value.strip()
        if value:
            value = [int(v, 16) for v in value.split(" ")]
        else:
            # table B.1
            value = None
        table[int(code, 16)] = value

########### Generate compact Python versions of the tables #############

print("""# This file is generated by mkstringprep.py. DO NOT EDIT.
\"\"\"Library that exposes various tables found in the StringPrep RFC 3454.

There are two kinds of tables: sets, for which a member test is provided,
and mappings, for which a mapping function is provided.
\"\"\"

from unicodedata import ucd_3_2_0 as unicodedata
""")

print("assert unicodedata.unidata_version == %r" % (unicodedata.unidata_version,))

# A.1 is the table of unassigned characters
# XXX Plane 15 PUA is listed as unassigned in Python.
name, table = tables[0]
del tables[0]
assert name == "A.1"
table = set(table.keys())
Cn = set(gen_category(["Cn"]))

# FDD0..FDEF are process internal codes
Cn -= set(range(0xFDD0, 0xFDF0))
# not a character
Cn -= set(range(0xFFFE, 0x110000, 0x10000))
Cn -= set(range(0xFFFF, 0x110000, 0x10000))

# assert table == Cn

print("""
def in_table_a1(code):
    if unicodedata.category(code) != 'Cn': return False
    c = ord(code)
    if 0xFDD0 <= c < 0xFDF0: return False
    return (c & 0xFFFF) not in (0xFFFE, 0xFFFF)
""")

# B.1 cannot easily be derived
name, table = tables[0]
del tables[0]
assert name == "B.1"
table = sorted(table.keys())
print("""
b1_set = """ + compact_set(table) + """
def in_table_b1(code):
    return ord(code) in b1_set
""")

# B.2 and B.3 is case folding.
# It takes CaseFolding.txt into account, which is
# not available in the Python database. Since
# B.2 is derived from B.3, we process B.3 first.
# B.3 supposedly *is* CaseFolding-3.2.0.txt.

name, table_b2 = tables[0]
del tables[0]
assert name == "B.2"

name, table_b3 = tables[0]
del tables[0]
assert name == "B.3"

# B.3 is mostly Python's .lower, except for a number
# of special cases, e.g. considering canonical forms.

b3_exceptions = {}

for k,v in table_b2.items():
    if list(map(ord, chr(k).lower())) != v:
        b3_exceptions[k] = "".join(map(chr,v))

b3 = sorted(b3_exceptions.items())

print("""
b3_exceptions = {""")
for i, kv in enumerate(b3):
    print("0x%x:%a," % kv, end=' ')
    if i % 4 == 3:
        print()
print("}")

print("""
def map_table_b3(code):
    r = b3_exceptions.get(ord(code))
    if r is not None: return r
    return code.lower()
""")

def map_table_b3(code):
    r = b3_exceptions.get(ord(code))
    if r is not None: return r
    return code.lower()

# B.2 is case folding for NFKC. This is the same as B.3,
# except where NormalizeWithKC(Fold(a)) !=
# NormalizeWithKC(Fold(NormalizeWithKC(Fold(a))))

def map_table_b2(a):
    al = map_table_b3(a)
    b = unicodedata.normalize("NFKC", al)
    bl = "".join([map_table_b3(ch) for ch in b])
    c = unicodedata.normalize("NFKC", bl)
    if b != c:
        return c
    else:
        return al

specials = {}
for k,v in table_b2.items():
    if list(map(ord, map_table_b2(chr(k)))) != v:
        specials[k] = v

# B.3 should not add any additional special cases
assert specials == {}

print("""
def map_table_b2(a):
    al = map_table_b3(a)
    b = unicodedata.normalize("NFKC", al)
    bl = "".join([map_table_b3(ch) for ch in b])
    c = unicodedata.normalize("NFKC", bl)
    if b != c:
        return c
    else:
        return al
""")

# C.1.1 is a table with a single character
name, table = tables[0]
del tables[0]
assert name == "C.1.1"
assert table == {0x20:0x20}

print("""
def in_table_c11(code):
    return code == " "
""")

# C.1.2 is the rest of all space characters
name, table = tables[0]
del tables[0]
assert name == "C.1.2"

# table = set(table.keys())
# Zs = set(gen_category(["Zs"])) - {0x20}
# assert Zs == table

print("""
def in_table_c12(code):
    return unicodedata.category(code) == "Zs" and code != " "

def in_table_c11_c12(code):
    return unicodedata.category(code) == "Zs"
""")

# C.2.1 ASCII control characters
name, table_c21 = tables[0]
del tables[0]
assert name == "C.2.1"

Cc = set(gen_category(["Cc"]))
Cc_ascii = Cc & set(range(128))
table_c21 = set(table_c21.keys())
assert Cc_ascii == table_c21

print("""
def in_table_c21(code):
    return ord(code) < 128 and unicodedata.category(code) == "Cc"
""")

# C.2.2 Non-ASCII control characters. It also includes
# a number of characters in category Cf.
name, table_c22 = tables[0]
del tables[0]
assert name == "C.2.2"

Cc_nonascii = Cc - Cc_ascii
table_c22 = set(table_c22.keys())
assert len(Cc_nonascii - table_c22) == 0

specials = list(table_c22 - Cc_nonascii)
specials.sort()

print("""c22_specials = """ + compact_set(specials) + """
def in_table_c22(code):
    c = ord(code)
    if c < 128: return False
    if unicodedata.category(code) == "Cc": return True
    return c in c22_specials

def in_table_c21_c22(code):
    return unicodedata.category(code) == "Cc" or \\
           ord(code) in c22_specials
""")

# C.3 Private use
name, table = tables[0]
del tables[0]
assert name == "C.3"

Co = set(gen_category(["Co"]))
assert set(table.keys()) == Co

print("""
def in_table_c3(code):
    return unicodedata.category(code) == "Co"
""")

# C.4 Non-character code points, xFFFE, xFFFF
# plus process internal codes
name, table = tables[0]
del tables[0]
assert name == "C.4"

nonchar = set(range(0xFDD0,0xFDF0))
nonchar.update(range(0xFFFE,0x110000,0x10000))
nonchar.update(range(0xFFFF,0x110000,0x10000))
table = set(table.keys())
assert table == nonchar

print("""
def in_table_c4(code):
    c = ord(code)
    if c < 0xFDD0: return False
    if c < 0xFDF0: return True
    return (ord(code) & 0xFFFF) in (0xFFFE, 0xFFFF)
""")

# C.5 Surrogate codes
name, table = tables[0]
del tables[0]
assert name == "C.5"

Cs = set(gen_category(["Cs"]))
assert set(table.keys()) == Cs

print("""
def in_table_c5(code):
    return unicodedata.category(code) == "Cs"
""")

# C.6 Inappropriate for plain text
name, table = tables[0]
del tables[0]
assert name == "C.6"

table = sorted(table.keys())

print("""
c6_set = """ + compact_set(table) + """
def in_table_c6(code):
    return ord(code) in c6_set
""")

# C.7 Inappropriate for canonical representation
name, table = tables[0]
del tables[0]
assert name == "C.7"

table = sorted(table.keys())

print("""
c7_set = """ + compact_set(table) + """
def in_table_c7(code):
    return ord(code) in c7_set
""")

# C.8 Change display properties or are deprecated
name, table = tables[0]
del tables[0]
assert name == "C.8"

table = sorted(table.keys())

print("""
c8_set = """ + compact_set(table) + """
def in_table_c8(code):
    return ord(code) in c8_set
""")

# C.9 Tagging characters
name, table = tables[0]
del tables[0]
assert name == "C.9"

table = sorted(table.keys())

print("""
c9_set = """ + compact_set(table) + """
def in_table_c9(code):
    return ord(code) in c9_set
""")

# D.1 Characters with bidirectional property "R" or "AL"
name, table = tables[0]
del tables[0]
assert name == "D.1"

RandAL = set(gen_bidirectional(["R","AL"]))
assert set(table.keys()) == RandAL

print("""
def in_table_d1(code):
    return unicodedata.bidirectional(code) in ("R","AL")
""")

# D.2 Characters with bidirectional property "L"
name, table = tables[0]
del tables[0]
assert name == "D.2"

L = set(gen_bidirectional(["L"]))
assert set(table.keys()) == L

print("""
def in_table_d2(code):
    return unicodedata.bidirectional(code) == "L"
""")