summaryrefslogtreecommitdiff
path: root/genUnicode.py
blob: c6e13a8549c91492e3362ebec1f8f1204646b4dc (plain)
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
#!/usr/bin/env python3
#
# Original script modified in November 2003 to take advantage of
# the character-validation range routines, and updated to the
# current Unicode information (Version 4.0.1)
#
# NOTE: there is an 'alias' facility for blocks which are not present in
#	the current release, but are needed for ABI compatibility.  This
#	must be accomplished MANUALLY!  Please see the comments below under
#     'blockAliases'
#
import sys
import string
import time

webpage = "http://www.unicode.org/Public/4.0-Update1/UCD-4.0.1.html"
sources = "Blocks-4.0.1.txt UnicodeData-4.0.1.txt"

#
# blockAliases is a small hack - it is used for mapping block names which
# were were used in the 3.1 release, but are missing or changed in the current
# release.  The format is "OldBlockName:NewBlockName1[,NewBlockName2[,...]]"
blockAliases = []
blockAliases.append("CombiningMarksforSymbols:CombiningDiacriticalMarksforSymbols")
blockAliases.append("Greek:GreekandCoptic")
blockAliases.append("PrivateUse:PrivateUseArea,SupplementaryPrivateUseArea-A," + 
	"SupplementaryPrivateUseArea-B")

# minTableSize gives the minimum number of ranges which must be present
# before a range table is produced.  If there are less than this
# number, inline comparisons are generated
minTableSize = 8

(blockfile, catfile) = sources.split()


#
# Now process the "blocks" file, reducing it to a dictionary
# indexed by blockname, containing a tuple with the applicable
# block range
#
BlockNames = {}
try:
    blocks = open(blockfile, "r")
except:
    print("Missing %s, aborting ..." % blockfile)
    sys.exit(1)

for line in blocks.readlines():
    if line[0] == '#':
        continue
    line = line.strip()
    if line == '':
        continue
    try:
        fields = line.split(';')
        range = fields[0].strip()
        (start, end) = range.split("..")
        name = fields[1].strip()
        name = name.replace(' ', '')
    except:
        print("Failed to process line: %s" % (line))
        continue
    start = "0x" + start
    end = "0x" + end
    try:
        BlockNames[name].append((start, end))
    except:
        BlockNames[name] = [(start, end)]
blocks.close()
print("Parsed %d blocks descriptions" % (len(BlockNames.keys())))

for block in blockAliases:
    alias = block.split(':')
    alist = alias[1].split(',')
    for comp in alist:
        if comp in BlockNames:
            if alias[0] not in BlockNames:
                BlockNames[alias[0]] = []
            for r in BlockNames[comp]:
                BlockNames[alias[0]].append(r)
        else:
            print("Alias %s: %s not in Blocks" % (alias[0], comp))
            continue

#
# Next process the Categories file. This is more complex, since
# the file is in code sequence, and we need to invert it.  We use
# a dictionary with index category-name, with each entry containing
# all the ranges (codepoints) of that category.  Note that category
# names comprise two parts - the general category, and the "subclass"
# within that category.  Therefore, both "general category" (which is
# the first character of the 2-character category-name) and the full
# (2-character) name are entered into this dictionary.
#
try:
    data = open(catfile, "r")
except:
    print("Missing %s, aborting ..." % catfile)
    sys.exit(1)

nbchar = 0;
Categories = {}
for line in data.readlines():
    if line[0] == '#':
        continue
    line = line.strip()
    if line == '':
        continue
    try:
        fields = line.split(';')
        point = fields[0].strip()
        value = 0
        while point != '':
            value = value * 16
            if point[0] >= '0' and point[0] <= '9':
                value = value + ord(point[0]) - ord('0')
            elif point[0] >= 'A' and point[0] <= 'F':
                value = value + 10 + ord(point[0]) - ord('A')
            elif point[0] >= 'a' and point[0] <= 'f':
                value = value + 10 + ord(point[0]) - ord('a')
            point = point[1:]
        name = fields[2]
    except:
        print("Failed to process line: %s" % (line))
        continue
    
    nbchar = nbchar + 1
    # update entry for "full name"
    try:
        Categories[name].append(value)
    except:
        try:
            Categories[name] = [value]
        except:
            print("Failed to process line: %s" % (line))
    # update "general category" name
    try:
        Categories[name[0]].append(value)
    except:
        try:
            Categories[name[0]] = [value]
        except:
            print("Failed to process line: %s" % (line))

blocks.close()
print("Parsed %d char generating %d categories" % (nbchar, len(Categories.keys())))

#
# The data is now all read.  Time to process it into a more useful form.
#
# reduce the number list into ranges
for cat in Categories.keys():
    list = Categories[cat]
    start = -1
    prev = -1
    end = -1
    ranges = []
    for val in list:
        if start == -1:
            start = val
            prev = val
            continue
        elif val == prev + 1:
            prev = val
            continue
        elif prev == start:
            ranges.append((prev, prev))
            start = val
            prev = val
            continue
        else:
            ranges.append((start, prev))
            start = val
            prev = val
            continue
    if prev == start:
        ranges.append((prev, prev))
    else:
        ranges.append((start, prev))
    Categories[cat] = ranges

#
# Assure all data is in alphabetic order, since we will be doing binary
# searches on the tables.
#
bkeys = sorted(BlockNames.keys())

ckeys = sorted(Categories.keys())

#
# Generate the resulting files
#
try:
    header = open("include/libxml/xmlunicode.h", "w")
except:
    print("Failed to open include/libxml/xmlunicode.h")
    sys.exit(1)

try:
    output = open("xmlunicode.c", "w")
except:
    print("Failed to open xmlunicode.c")
    sys.exit(1)

date = time.asctime(time.localtime(time.time()))

header.write(
"""/*
 * Summary: Unicode character APIs
 * Description: API for the Unicode character APIs
 *
 * This file is automatically generated from the
 * UCS description files of the Unicode Character Database
 * %s
 * using the genUnicode.py Python script.
 *
 * Generation date: %s
 * Sources: %s
 * Author: Daniel Veillard
 */

#ifndef __XML_UNICODE_H__
#define __XML_UNICODE_H__

#include <libxml/xmlversion.h>

#ifdef LIBXML_UNICODE_ENABLED

#ifdef __cplusplus
extern "C" {
#endif

""" % (webpage, date, sources));

output.write(
"""/*
 * xmlunicode.c: this module implements the Unicode character APIs
 *
 * This file is automatically generated from the
 * UCS description files of the Unicode Character Database
 * %s
 * using the genUnicode.py Python script.
 *
 * Generation date: %s
 * Sources: %s
 * Daniel Veillard <veillard@redhat.com>
 */

#define IN_LIBXML
#include "libxml.h"

#ifdef LIBXML_UNICODE_ENABLED

#include <string.h>
#include <libxml/xmlversion.h>
#include <libxml/xmlunicode.h>
#include <libxml/chvalid.h>

typedef int (xmlIntFunc)(int);	/* just to keep one's mind untwisted */

typedef struct {
    const char *rangename;
    xmlIntFunc *func;
} xmlUnicodeRange;

typedef struct {
    const xmlUnicodeRange *table;
    int		    numentries;
} xmlUnicodeNameTable;


static xmlIntFunc *xmlUnicodeLookup(const xmlUnicodeNameTable *tptr, const char *tname);

static const xmlUnicodeRange xmlUnicodeBlocks[] = {
""" % (webpage, date, sources));

flag = 0
for block in bkeys:
    name = block.replace('-', '')
    if flag:
        output.write(',\n')
    else:
        flag = 1
    output.write('  {"%s", xmlUCSIs%s}' % (block, name))
output.write('};\n\n')

output.write('static const xmlUnicodeRange xmlUnicodeCats[] = {\n')
flag = 0;
for name in ckeys:
    if flag:
        output.write(',\n')
    else:
        flag = 1
    output.write('  {"%s", xmlUCSIsCat%s}' % (name, name))
output.write('};\n\n')

#
# For any categories with more than minTableSize ranges we generate
# a range table suitable for xmlCharInRange
#
for name in ckeys:
  if len(Categories[name]) > minTableSize:
    numshort = 0
    numlong = 0
    ranges = Categories[name]
    sptr = "NULL"
    lptr = "NULL"
    for range in ranges:
      (low, high) = range
      if high < 0x10000:
        if numshort == 0:
          pline = "static const xmlChSRange xml%sS[] = {" % name
          sptr = "xml%sS" % name
        else:
          pline += ","
        numshort += 1
      else:
        if numlong == 0:
          if numshort > 0:
            output.write(pline + " };\n")
          pline = "static const xmlChLRange xml%sL[] = {" % name
          lptr = "xml%sL" % name
        else:
          pline += ","
        numlong += 1
      if len(pline) > 60:
        output.write(pline + "\n")
        pline = "    "
      elif pline[-1:] == ",":
        pline += " "
      pline += "{%s, %s}" % (hex(low), hex(high))
    output.write(pline + " };\nstatic const xmlChRangeGroup xml%sG = {%s,%s,%s,%s};\n\n"
         % (name, numshort, numlong, sptr, lptr))


output.write(
"""static const xmlUnicodeNameTable xmlUnicodeBlockTbl = {xmlUnicodeBlocks, %s};
static const xmlUnicodeNameTable xmlUnicodeCatTbl = {xmlUnicodeCats, %s};

/**
 * xmlUnicodeLookup:
 * @tptr: pointer to the name table
 * @name: name to be found
 *
 * binary table lookup for user-supplied name
 *
 * Returns pointer to range function if found, otherwise NULL
 */
static xmlIntFunc
*xmlUnicodeLookup(const xmlUnicodeNameTable *tptr, const char *tname) {
    int low, high, mid, cmp;
    const xmlUnicodeRange *sptr;

    if ((tptr == NULL) || (tname == NULL)) return(NULL);

    low = 0;
    high = tptr->numentries - 1;
    sptr = tptr->table;
    while (low <= high) {
	mid = (low + high) / 2;
	if ((cmp=strcmp(tname, sptr[mid].rangename)) == 0)
	    return (sptr[mid].func);
	if (cmp < 0)
	    high = mid - 1;
	else
	    low = mid + 1;
    }
    return (NULL);
}

""" % (len(BlockNames), len(Categories)) )

for block in bkeys:
    name = block.replace('-', '')
    header.write("XMLPUBFUN int xmlUCSIs%s\t(int code);\n" % name)
    output.write("/**\n * xmlUCSIs%s:\n * @code: UCS code point\n" % (name))
    output.write(" *\n * Check whether the character is part of %s UCS Block\n"%
                 (block))
    output.write(" *\n * Returns 1 if true 0 otherwise\n */\n");
    output.write("int\nxmlUCSIs%s(int code) {\n    return(" % name)
    flag = 0
    for (start, end) in BlockNames[block]:
        if flag:
            output.write(" ||\n           ")
        else:
            flag = 1
        output.write("((code >= %s) && (code <= %s))" % (start, end))
    output.write(");\n}\n\n")

header.write("\nXMLPUBFUN int xmlUCSIsBlock\t(int code, const char *block);\n\n")
output.write(
"""/**
 * xmlUCSIsBlock:
 * @code: UCS code point
 * @block: UCS block name
 *
 * Check whether the character is part of the UCS Block
 *
 * Returns 1 if true, 0 if false and -1 on unknown block
 */
int
xmlUCSIsBlock(int code, const char *block) {
    xmlIntFunc *func;

    func = xmlUnicodeLookup(&xmlUnicodeBlockTbl, block);
    if (func == NULL)
	return (-1);
    return (func(code));
}

""")

for name in ckeys:
    ranges = Categories[name]
    header.write("XMLPUBFUN int xmlUCSIsCat%s\t(int code);\n" % name)
    output.write("/**\n * xmlUCSIsCat%s:\n * @code: UCS code point\n" % (name))
    output.write(" *\n * Check whether the character is part of %s UCS Category\n"%
                 (name))
    output.write(" *\n * Returns 1 if true 0 otherwise\n */\n");
    output.write("int\nxmlUCSIsCat%s(int code) {\n" % name)
    if len(Categories[name]) > minTableSize:
        output.write("    return(xmlCharInRange((unsigned int)code, &xml%sG)"
            % name)
    else:
        start = 1
        for range in ranges:
            (begin, end) = range;
            if start:
                output.write("    return(");
                start = 0
            else:
                output.write(" ||\n           ");
            if (begin == end):
                output.write("(code == %s)" % (hex(begin)))
            else:
                output.write("((code >= %s) && (code <= %s))" % (
                         hex(begin), hex(end)))
    output.write(");\n}\n\n")

header.write("\nXMLPUBFUN int xmlUCSIsCat\t(int code, const char *cat);\n")
output.write(
"""/**
 * xmlUCSIsCat:
 * @code: UCS code point
 * @cat: UCS Category name
 *
 * Check whether the character is part of the UCS Category
 *
 * Returns 1 if true, 0 if false and -1 on unknown category
 */
int
xmlUCSIsCat(int code, const char *cat) {
    xmlIntFunc *func;

    func = xmlUnicodeLookup(&xmlUnicodeCatTbl, cat);
    if (func == NULL)
	return (-1);
    return (func(code));
}

#endif /* LIBXML_UNICODE_ENABLED */
""")

header.write("""
#ifdef __cplusplus
}
#endif

#endif /* LIBXML_UNICODE_ENABLED */

#endif /* __XML_UNICODE_H__ */
""");

header.close()
output.close()