| changeset 28068 | 9ece901f7a19 |
| parent 28067 | 69a02b1e947c |
| child 28069 | b2d24c2898f9 |
| 28067:69a02b1e947c | 28068:9ece901f7a19 |
|---|---|
5 # This software may be used and distributed according to the terms of the |
5 # This software may be used and distributed according to the terms of the |
6 # GNU General Public License version 2 or any later version. |
6 # GNU General Public License version 2 or any later version. |
7 |
7 |
8 from __future__ import absolute_import |
8 from __future__ import absolute_import |
9 |
9 |
10 import array |
|
10 import locale |
11 import locale |
11 import os |
12 import os |
12 import unicodedata |
13 import unicodedata |
13 |
14 |
14 from . import ( |
15 from . import ( |
378 upper = 1 |
379 upper = 1 |
379 other = 0 |
380 other = 0 |
380 |
381 |
381 _jsonmap = [] |
382 _jsonmap = [] |
382 _jsonmap.extend("\\u%04x" % x for x in xrange(32)) |
383 _jsonmap.extend("\\u%04x" % x for x in xrange(32)) |
383 _jsonmap.extend(chr(x) for x in xrange(32, 256)) |
384 _jsonmap.extend(chr(x) for x in xrange(32, 127)) |
384 _jsonmap[0x7f] = '\\u007f' |
385 _jsonmap.append('\\u007f') |
385 _jsonmap[0x09] = '\\t' |
386 _jsonmap[0x09] = '\\t' |
386 _jsonmap[0x0a] = '\\n' |
387 _jsonmap[0x0a] = '\\n' |
387 _jsonmap[0x22] = '\\"' |
388 _jsonmap[0x22] = '\\"' |
388 _jsonmap[0x5c] = '\\\\' |
389 _jsonmap[0x5c] = '\\\\' |
389 _jsonmap[0x08] = '\\b' |
390 _jsonmap[0x08] = '\\b' |
390 _jsonmap[0x0c] = '\\f' |
391 _jsonmap[0x0c] = '\\f' |
391 _jsonmap[0x0d] = '\\r' |
392 _jsonmap[0x0d] = '\\r' |
392 |
393 _paranoidjsonmap = _jsonmap[:] |
393 def jsonescape(s): |
394 _jsonmap.extend(chr(x) for x in xrange(128, 256)) |
395 |
|
396 def jsonescape(s, paranoid=False): |
|
394 '''returns a string suitable for JSON |
397 '''returns a string suitable for JSON |
395 |
398 |
396 JSON is problematic for us because it doesn't support non-Unicode |
399 JSON is problematic for us because it doesn't support non-Unicode |
397 bytes. To deal with this, we take the following approach: |
400 bytes. To deal with this, we take the following approach: |
398 |
401 |
413 'a weird byte: \\xed\\xb3\\x9d' |
416 'a weird byte: \\xed\\xb3\\x9d' |
414 >>> jsonescape('utf-8: caf\\xc3\\xa9') |
417 >>> jsonescape('utf-8: caf\\xc3\\xa9') |
415 'utf-8: caf\\xc3\\xa9' |
418 'utf-8: caf\\xc3\\xa9' |
416 >>> jsonescape('') |
419 >>> jsonescape('') |
417 '' |
420 '' |
421 |
|
422 If paranoid, non-ascii characters are also escaped. This is suitable for |
|
423 web output. |
|
424 |
|
425 >>> jsonescape('escape boundary: \\x7e \\x7f \\xc2\\x80', paranoid=True) |
|
426 'escape boundary: ~ \\\\u007f \\\\u0080' |
|
427 >>> jsonescape('a weird byte: \\xdd', paranoid=True) |
|
428 'a weird byte: \\\\udcdd' |
|
429 >>> jsonescape('utf-8: caf\\xc3\\xa9', paranoid=True) |
|
430 'utf-8: caf\\\\u00e9' |
|
431 >>> jsonescape('non-BMP: \\xf0\\x9d\\x84\\x9e', paranoid=True) |
|
432 'non-BMP: \\\\ud834\\\\udd1e' |
|
418 ''' |
433 ''' |
419 |
434 |
420 return ''.join(_jsonmap[x] for x in bytearray(toutf8b(s))) |
435 if paranoid: |
436 jm = _paranoidjsonmap |
|
437 else: |
|
438 jm = _jsonmap |
|
439 |
|
440 u8chars = toutf8b(s) |
|
441 try: |
|
442 return ''.join(jm[x] for x in bytearray(u8chars)) # fast path |
|
443 except IndexError: |
|
444 pass |
|
445 # non-BMP char is represented as UTF-16 surrogate pair |
|
446 u16codes = array.array('H', u8chars.decode('utf-8').encode('utf-16')) |
|
447 u16codes.pop(0) # drop BOM |
|
448 return ''.join(jm[x] if x < 128 else '\\u%04x' % x for x in u16codes) |
|
421 |
449 |
422 _utf8len = [0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4] |
450 _utf8len = [0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4] |
423 |
451 |
424 def getutf8char(s, pos): |
452 def getutf8char(s, pos): |
425 '''get the next full utf-8 character in the given string, starting at pos |
453 '''get the next full utf-8 character in the given string, starting at pos |