pygments.rb 0.2.13 → 0.3.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (64) hide show
  1. data/.gitignore +1 -0
  2. data/README.md +45 -19
  3. data/Rakefile +21 -11
  4. data/bench.rb +15 -48
  5. data/cache-lexers.rb +8 -0
  6. data/lexers +0 -0
  7. data/lib/pygments.rb +3 -6
  8. data/lib/pygments/mentos.py +343 -0
  9. data/lib/pygments/popen.rb +383 -0
  10. data/lib/pygments/version.rb +1 -1
  11. data/pygments.rb.gemspec +5 -4
  12. data/test/test_data.c +2581 -0
  13. data/test/test_data.py +514 -0
  14. data/test/test_data_generated +2582 -0
  15. data/test/test_pygments.rb +208 -84
  16. data/vendor/pygments-main/pygments/lexers/_mapping.py +1 -1
  17. data/vendor/pygments-main/pygments/lexers/shell.py +1 -1
  18. data/vendor/simplejson/.gitignore +10 -0
  19. data/vendor/simplejson/.travis.yml +5 -0
  20. data/vendor/simplejson/CHANGES.txt +291 -0
  21. data/vendor/simplejson/LICENSE.txt +19 -0
  22. data/vendor/simplejson/MANIFEST.in +5 -0
  23. data/vendor/simplejson/README.rst +19 -0
  24. data/vendor/simplejson/conf.py +179 -0
  25. data/vendor/simplejson/index.rst +628 -0
  26. data/vendor/simplejson/scripts/make_docs.py +18 -0
  27. data/vendor/simplejson/setup.py +104 -0
  28. data/vendor/simplejson/simplejson/__init__.py +510 -0
  29. data/vendor/simplejson/simplejson/_speedups.c +2745 -0
  30. data/vendor/simplejson/simplejson/decoder.py +425 -0
  31. data/vendor/simplejson/simplejson/encoder.py +567 -0
  32. data/vendor/simplejson/simplejson/ordered_dict.py +119 -0
  33. data/vendor/simplejson/simplejson/scanner.py +77 -0
  34. data/vendor/simplejson/simplejson/tests/__init__.py +67 -0
  35. data/vendor/simplejson/simplejson/tests/test_bigint_as_string.py +55 -0
  36. data/vendor/simplejson/simplejson/tests/test_check_circular.py +30 -0
  37. data/vendor/simplejson/simplejson/tests/test_decimal.py +66 -0
  38. data/vendor/simplejson/simplejson/tests/test_decode.py +83 -0
  39. data/vendor/simplejson/simplejson/tests/test_default.py +9 -0
  40. data/vendor/simplejson/simplejson/tests/test_dump.py +67 -0
  41. data/vendor/simplejson/simplejson/tests/test_encode_basestring_ascii.py +46 -0
  42. data/vendor/simplejson/simplejson/tests/test_encode_for_html.py +32 -0
  43. data/vendor/simplejson/simplejson/tests/test_errors.py +34 -0
  44. data/vendor/simplejson/simplejson/tests/test_fail.py +91 -0
  45. data/vendor/simplejson/simplejson/tests/test_float.py +19 -0
  46. data/vendor/simplejson/simplejson/tests/test_indent.py +86 -0
  47. data/vendor/simplejson/simplejson/tests/test_item_sort_key.py +20 -0
  48. data/vendor/simplejson/simplejson/tests/test_namedtuple.py +121 -0
  49. data/vendor/simplejson/simplejson/tests/test_pass1.py +76 -0
  50. data/vendor/simplejson/simplejson/tests/test_pass2.py +14 -0
  51. data/vendor/simplejson/simplejson/tests/test_pass3.py +20 -0
  52. data/vendor/simplejson/simplejson/tests/test_recursion.py +67 -0
  53. data/vendor/simplejson/simplejson/tests/test_scanstring.py +117 -0
  54. data/vendor/simplejson/simplejson/tests/test_separators.py +42 -0
  55. data/vendor/simplejson/simplejson/tests/test_speedups.py +20 -0
  56. data/vendor/simplejson/simplejson/tests/test_tuple.py +49 -0
  57. data/vendor/simplejson/simplejson/tests/test_unicode.py +109 -0
  58. data/vendor/simplejson/simplejson/tool.py +39 -0
  59. metadata +80 -22
  60. data/ext/extconf.rb +0 -14
  61. data/ext/pygments.c +0 -466
  62. data/lib/pygments/c.rb +0 -54
  63. data/lib/pygments/ffi.rb +0 -155
  64. data/vendor/.gitignore +0 -1
@@ -0,0 +1,425 @@
1
+ """Implementation of JSONDecoder
2
+ """
3
+ import re
4
+ import sys
5
+ import struct
6
+
7
+ from simplejson.scanner import make_scanner
8
+ def _import_c_scanstring():
9
+ try:
10
+ from simplejson._speedups import scanstring
11
+ return scanstring
12
+ except ImportError:
13
+ return None
14
+ c_scanstring = _import_c_scanstring()
15
+
16
+ __all__ = ['JSONDecoder']
17
+
18
+ FLAGS = re.VERBOSE | re.MULTILINE | re.DOTALL
19
+
20
+ def _floatconstants():
21
+ _BYTES = '7FF80000000000007FF0000000000000'.decode('hex')
22
+ # The struct module in Python 2.4 would get frexp() out of range here
23
+ # when an endian is specified in the format string. Fixed in Python 2.5+
24
+ if sys.byteorder != 'big':
25
+ _BYTES = _BYTES[:8][::-1] + _BYTES[8:][::-1]
26
+ nan, inf = struct.unpack('dd', _BYTES)
27
+ return nan, inf, -inf
28
+
29
+ NaN, PosInf, NegInf = _floatconstants()
30
+
31
+
32
+ class JSONDecodeError(ValueError):
33
+ """Subclass of ValueError with the following additional properties:
34
+
35
+ msg: The unformatted error message
36
+ doc: The JSON document being parsed
37
+ pos: The start index of doc where parsing failed
38
+ end: The end index of doc where parsing failed (may be None)
39
+ lineno: The line corresponding to pos
40
+ colno: The column corresponding to pos
41
+ endlineno: The line corresponding to end (may be None)
42
+ endcolno: The column corresponding to end (may be None)
43
+
44
+ """
45
+ def __init__(self, msg, doc, pos, end=None):
46
+ ValueError.__init__(self, errmsg(msg, doc, pos, end=end))
47
+ self.msg = msg
48
+ self.doc = doc
49
+ self.pos = pos
50
+ self.end = end
51
+ self.lineno, self.colno = linecol(doc, pos)
52
+ if end is not None:
53
+ self.endlineno, self.endcolno = linecol(doc, end)
54
+ else:
55
+ self.endlineno, self.endcolno = None, None
56
+
57
+
58
+ def linecol(doc, pos):
59
+ lineno = doc.count('\n', 0, pos) + 1
60
+ if lineno == 1:
61
+ colno = pos
62
+ else:
63
+ colno = pos - doc.rindex('\n', 0, pos)
64
+ return lineno, colno
65
+
66
+
67
+ def errmsg(msg, doc, pos, end=None):
68
+ # Note that this function is called from _speedups
69
+ lineno, colno = linecol(doc, pos)
70
+ if end is None:
71
+ #fmt = '{0}: line {1} column {2} (char {3})'
72
+ #return fmt.format(msg, lineno, colno, pos)
73
+ fmt = '%s: line %d column %d (char %d)'
74
+ return fmt % (msg, lineno, colno, pos)
75
+ endlineno, endcolno = linecol(doc, end)
76
+ #fmt = '{0}: line {1} column {2} - line {3} column {4} (char {5} - {6})'
77
+ #return fmt.format(msg, lineno, colno, endlineno, endcolno, pos, end)
78
+ fmt = '%s: line %d column %d - line %d column %d (char %d - %d)'
79
+ return fmt % (msg, lineno, colno, endlineno, endcolno, pos, end)
80
+
81
+
82
+ _CONSTANTS = {
83
+ '-Infinity': NegInf,
84
+ 'Infinity': PosInf,
85
+ 'NaN': NaN,
86
+ }
87
+
88
+ STRINGCHUNK = re.compile(r'(.*?)(["\\\x00-\x1f])', FLAGS)
89
+ BACKSLASH = {
90
+ '"': u'"', '\\': u'\\', '/': u'/',
91
+ 'b': u'\b', 'f': u'\f', 'n': u'\n', 'r': u'\r', 't': u'\t',
92
+ }
93
+
94
+ DEFAULT_ENCODING = "utf-8"
95
+
96
+ def py_scanstring(s, end, encoding=None, strict=True,
97
+ _b=BACKSLASH, _m=STRINGCHUNK.match):
98
+ """Scan the string s for a JSON string. End is the index of the
99
+ character in s after the quote that started the JSON string.
100
+ Unescapes all valid JSON string escape sequences and raises ValueError
101
+ on attempt to decode an invalid string. If strict is False then literal
102
+ control characters are allowed in the string.
103
+
104
+ Returns a tuple of the decoded string and the index of the character in s
105
+ after the end quote."""
106
+ if encoding is None:
107
+ encoding = DEFAULT_ENCODING
108
+ chunks = []
109
+ _append = chunks.append
110
+ begin = end - 1
111
+ while 1:
112
+ chunk = _m(s, end)
113
+ if chunk is None:
114
+ raise JSONDecodeError(
115
+ "Unterminated string starting at", s, begin)
116
+ end = chunk.end()
117
+ content, terminator = chunk.groups()
118
+ # Content is contains zero or more unescaped string characters
119
+ if content:
120
+ if not isinstance(content, unicode):
121
+ content = unicode(content, encoding)
122
+ _append(content)
123
+ # Terminator is the end of string, a literal control character,
124
+ # or a backslash denoting that an escape sequence follows
125
+ if terminator == '"':
126
+ break
127
+ elif terminator != '\\':
128
+ if strict:
129
+ msg = "Invalid control character %r at" % (terminator,)
130
+ #msg = "Invalid control character {0!r} at".format(terminator)
131
+ raise JSONDecodeError(msg, s, end)
132
+ else:
133
+ _append(terminator)
134
+ continue
135
+ try:
136
+ esc = s[end]
137
+ except IndexError:
138
+ raise JSONDecodeError(
139
+ "Unterminated string starting at", s, begin)
140
+ # If not a unicode escape sequence, must be in the lookup table
141
+ if esc != 'u':
142
+ try:
143
+ char = _b[esc]
144
+ except KeyError:
145
+ msg = "Invalid \\escape: " + repr(esc)
146
+ raise JSONDecodeError(msg, s, end)
147
+ end += 1
148
+ else:
149
+ # Unicode escape sequence
150
+ esc = s[end + 1:end + 5]
151
+ next_end = end + 5
152
+ if len(esc) != 4:
153
+ msg = "Invalid \\uXXXX escape"
154
+ raise JSONDecodeError(msg, s, end)
155
+ uni = int(esc, 16)
156
+ # Check for surrogate pair on UCS-4 systems
157
+ if 0xd800 <= uni <= 0xdbff and sys.maxunicode > 65535:
158
+ msg = "Invalid \\uXXXX\\uXXXX surrogate pair"
159
+ if not s[end + 5:end + 7] == '\\u':
160
+ raise JSONDecodeError(msg, s, end)
161
+ esc2 = s[end + 7:end + 11]
162
+ if len(esc2) != 4:
163
+ raise JSONDecodeError(msg, s, end)
164
+ uni2 = int(esc2, 16)
165
+ uni = 0x10000 + (((uni - 0xd800) << 10) | (uni2 - 0xdc00))
166
+ next_end += 6
167
+ char = unichr(uni)
168
+ end = next_end
169
+ # Append the unescaped character
170
+ _append(char)
171
+ return u''.join(chunks), end
172
+
173
+
174
+ # Use speedup if available
175
+ scanstring = c_scanstring or py_scanstring
176
+
177
+ WHITESPACE = re.compile(r'[ \t\n\r]*', FLAGS)
178
+ WHITESPACE_STR = ' \t\n\r'
179
+
180
+ def JSONObject((s, end), encoding, strict, scan_once, object_hook,
181
+ object_pairs_hook, memo=None,
182
+ _w=WHITESPACE.match, _ws=WHITESPACE_STR):
183
+ # Backwards compatibility
184
+ if memo is None:
185
+ memo = {}
186
+ memo_get = memo.setdefault
187
+ pairs = []
188
+ # Use a slice to prevent IndexError from being raised, the following
189
+ # check will raise a more specific ValueError if the string is empty
190
+ nextchar = s[end:end + 1]
191
+ # Normally we expect nextchar == '"'
192
+ if nextchar != '"':
193
+ if nextchar in _ws:
194
+ end = _w(s, end).end()
195
+ nextchar = s[end:end + 1]
196
+ # Trivial empty object
197
+ if nextchar == '}':
198
+ if object_pairs_hook is not None:
199
+ result = object_pairs_hook(pairs)
200
+ return result, end + 1
201
+ pairs = {}
202
+ if object_hook is not None:
203
+ pairs = object_hook(pairs)
204
+ return pairs, end + 1
205
+ elif nextchar != '"':
206
+ raise JSONDecodeError(
207
+ "Expecting property name enclosed in double quotes",
208
+ s, end)
209
+ end += 1
210
+ while True:
211
+ key, end = scanstring(s, end, encoding, strict)
212
+ key = memo_get(key, key)
213
+
214
+ # To skip some function call overhead we optimize the fast paths where
215
+ # the JSON key separator is ": " or just ":".
216
+ if s[end:end + 1] != ':':
217
+ end = _w(s, end).end()
218
+ if s[end:end + 1] != ':':
219
+ raise JSONDecodeError("Expecting ':' delimiter", s, end)
220
+
221
+ end += 1
222
+
223
+ try:
224
+ if s[end] in _ws:
225
+ end += 1
226
+ if s[end] in _ws:
227
+ end = _w(s, end + 1).end()
228
+ except IndexError:
229
+ pass
230
+
231
+ try:
232
+ value, end = scan_once(s, end)
233
+ except StopIteration:
234
+ raise JSONDecodeError("Expecting object", s, end)
235
+ pairs.append((key, value))
236
+
237
+ try:
238
+ nextchar = s[end]
239
+ if nextchar in _ws:
240
+ end = _w(s, end + 1).end()
241
+ nextchar = s[end]
242
+ except IndexError:
243
+ nextchar = ''
244
+ end += 1
245
+
246
+ if nextchar == '}':
247
+ break
248
+ elif nextchar != ',':
249
+ raise JSONDecodeError("Expecting ',' delimiter", s, end - 1)
250
+
251
+ try:
252
+ nextchar = s[end]
253
+ if nextchar in _ws:
254
+ end += 1
255
+ nextchar = s[end]
256
+ if nextchar in _ws:
257
+ end = _w(s, end + 1).end()
258
+ nextchar = s[end]
259
+ except IndexError:
260
+ nextchar = ''
261
+
262
+ end += 1
263
+ if nextchar != '"':
264
+ raise JSONDecodeError(
265
+ "Expecting property name enclosed in double quotes",
266
+ s, end - 1)
267
+
268
+ if object_pairs_hook is not None:
269
+ result = object_pairs_hook(pairs)
270
+ return result, end
271
+ pairs = dict(pairs)
272
+ if object_hook is not None:
273
+ pairs = object_hook(pairs)
274
+ return pairs, end
275
+
276
+ def JSONArray((s, end), scan_once, _w=WHITESPACE.match, _ws=WHITESPACE_STR):
277
+ values = []
278
+ nextchar = s[end:end + 1]
279
+ if nextchar in _ws:
280
+ end = _w(s, end + 1).end()
281
+ nextchar = s[end:end + 1]
282
+ # Look-ahead for trivial empty array
283
+ if nextchar == ']':
284
+ return values, end + 1
285
+ _append = values.append
286
+ while True:
287
+ try:
288
+ value, end = scan_once(s, end)
289
+ except StopIteration:
290
+ raise JSONDecodeError("Expecting object", s, end)
291
+ _append(value)
292
+ nextchar = s[end:end + 1]
293
+ if nextchar in _ws:
294
+ end = _w(s, end + 1).end()
295
+ nextchar = s[end:end + 1]
296
+ end += 1
297
+ if nextchar == ']':
298
+ break
299
+ elif nextchar != ',':
300
+ raise JSONDecodeError("Expecting ',' delimiter", s, end)
301
+
302
+ try:
303
+ if s[end] in _ws:
304
+ end += 1
305
+ if s[end] in _ws:
306
+ end = _w(s, end + 1).end()
307
+ except IndexError:
308
+ pass
309
+
310
+ return values, end
311
+
312
+ class JSONDecoder(object):
313
+ """Simple JSON <http://json.org> decoder
314
+
315
+ Performs the following translations in decoding by default:
316
+
317
+ +---------------+-------------------+
318
+ | JSON | Python |
319
+ +===============+===================+
320
+ | object | dict |
321
+ +---------------+-------------------+
322
+ | array | list |
323
+ +---------------+-------------------+
324
+ | string | unicode |
325
+ +---------------+-------------------+
326
+ | number (int) | int, long |
327
+ +---------------+-------------------+
328
+ | number (real) | float |
329
+ +---------------+-------------------+
330
+ | true | True |
331
+ +---------------+-------------------+
332
+ | false | False |
333
+ +---------------+-------------------+
334
+ | null | None |
335
+ +---------------+-------------------+
336
+
337
+ It also understands ``NaN``, ``Infinity``, and ``-Infinity`` as
338
+ their corresponding ``float`` values, which is outside the JSON spec.
339
+
340
+ """
341
+
342
+ def __init__(self, encoding=None, object_hook=None, parse_float=None,
343
+ parse_int=None, parse_constant=None, strict=True,
344
+ object_pairs_hook=None):
345
+ """
346
+ *encoding* determines the encoding used to interpret any
347
+ :class:`str` objects decoded by this instance (``'utf-8'`` by
348
+ default). It has no effect when decoding :class:`unicode` objects.
349
+
350
+ Note that currently only encodings that are a superset of ASCII work,
351
+ strings of other encodings should be passed in as :class:`unicode`.
352
+
353
+ *object_hook*, if specified, will be called with the result of every
354
+ JSON object decoded and its return value will be used in place of the
355
+ given :class:`dict`. This can be used to provide custom
356
+ deserializations (e.g. to support JSON-RPC class hinting).
357
+
358
+ *object_pairs_hook* is an optional function that will be called with
359
+ the result of any object literal decode with an ordered list of pairs.
360
+ The return value of *object_pairs_hook* will be used instead of the
361
+ :class:`dict`. This feature can be used to implement custom decoders
362
+ that rely on the order that the key and value pairs are decoded (for
363
+ example, :func:`collections.OrderedDict` will remember the order of
364
+ insertion). If *object_hook* is also defined, the *object_pairs_hook*
365
+ takes priority.
366
+
367
+ *parse_float*, if specified, will be called with the string of every
368
+ JSON float to be decoded. By default, this is equivalent to
369
+ ``float(num_str)``. This can be used to use another datatype or parser
370
+ for JSON floats (e.g. :class:`decimal.Decimal`).
371
+
372
+ *parse_int*, if specified, will be called with the string of every
373
+ JSON int to be decoded. By default, this is equivalent to
374
+ ``int(num_str)``. This can be used to use another datatype or parser
375
+ for JSON integers (e.g. :class:`float`).
376
+
377
+ *parse_constant*, if specified, will be called with one of the
378
+ following strings: ``'-Infinity'``, ``'Infinity'``, ``'NaN'``. This
379
+ can be used to raise an exception if invalid JSON numbers are
380
+ encountered.
381
+
382
+ *strict* controls the parser's behavior when it encounters an
383
+ invalid control character in a string. The default setting of
384
+ ``True`` means that unescaped control characters are parse errors, if
385
+ ``False`` then control characters will be allowed in strings.
386
+
387
+ """
388
+ self.encoding = encoding
389
+ self.object_hook = object_hook
390
+ self.object_pairs_hook = object_pairs_hook
391
+ self.parse_float = parse_float or float
392
+ self.parse_int = parse_int or int
393
+ self.parse_constant = parse_constant or _CONSTANTS.__getitem__
394
+ self.strict = strict
395
+ self.parse_object = JSONObject
396
+ self.parse_array = JSONArray
397
+ self.parse_string = scanstring
398
+ self.memo = {}
399
+ self.scan_once = make_scanner(self)
400
+
401
+ def decode(self, s, _w=WHITESPACE.match):
402
+ """Return the Python representation of ``s`` (a ``str`` or ``unicode``
403
+ instance containing a JSON document)
404
+
405
+ """
406
+ obj, end = self.raw_decode(s, idx=_w(s, 0).end())
407
+ end = _w(s, end).end()
408
+ if end != len(s):
409
+ raise JSONDecodeError("Extra data", s, end, len(s))
410
+ return obj
411
+
412
+ def raw_decode(self, s, idx=0):
413
+ """Decode a JSON document from ``s`` (a ``str`` or ``unicode``
414
+ beginning with a JSON document) and return a 2-tuple of the Python
415
+ representation and the index in ``s`` where the document ended.
416
+
417
+ This can be used to decode a JSON document from a string that may
418
+ have extraneous data at the end.
419
+
420
+ """
421
+ try:
422
+ obj, end = self.scan_once(s, idx)
423
+ except StopIteration:
424
+ raise JSONDecodeError("No JSON object could be decoded", s, idx)
425
+ return obj, end
@@ -0,0 +1,567 @@
1
+ """Implementation of JSONEncoder
2
+ """
3
+ import re
4
+ from decimal import Decimal
5
+
6
+ def _import_speedups():
7
+ try:
8
+ from simplejson import _speedups
9
+ return _speedups.encode_basestring_ascii, _speedups.make_encoder
10
+ except ImportError:
11
+ return None, None
12
+ c_encode_basestring_ascii, c_make_encoder = _import_speedups()
13
+
14
+ from simplejson.decoder import PosInf
15
+
16
+ ESCAPE = re.compile(ur'[\x00-\x1f\\"\b\f\n\r\t\u2028\u2029]')
17
+ ESCAPE_ASCII = re.compile(r'([\\"]|[^\ -~])')
18
+ HAS_UTF8 = re.compile(r'[\x80-\xff]')
19
+ ESCAPE_DCT = {
20
+ '\\': '\\\\',
21
+ '"': '\\"',
22
+ '\b': '\\b',
23
+ '\f': '\\f',
24
+ '\n': '\\n',
25
+ '\r': '\\r',
26
+ '\t': '\\t',
27
+ u'\u2028': '\\u2028',
28
+ u'\u2029': '\\u2029',
29
+ }
30
+ for i in range(0x20):
31
+ #ESCAPE_DCT.setdefault(chr(i), '\\u{0:04x}'.format(i))
32
+ ESCAPE_DCT.setdefault(chr(i), '\\u%04x' % (i,))
33
+
34
+ FLOAT_REPR = repr
35
+
36
+ def encode_basestring(s):
37
+ """Return a JSON representation of a Python string
38
+
39
+ """
40
+ if isinstance(s, str) and HAS_UTF8.search(s) is not None:
41
+ s = s.decode('utf-8')
42
+ def replace(match):
43
+ return ESCAPE_DCT[match.group(0)]
44
+ return u'"' + ESCAPE.sub(replace, s) + u'"'
45
+
46
+
47
+ def py_encode_basestring_ascii(s):
48
+ """Return an ASCII-only JSON representation of a Python string
49
+
50
+ """
51
+ if isinstance(s, str) and HAS_UTF8.search(s) is not None:
52
+ s = s.decode('utf-8')
53
+ def replace(match):
54
+ s = match.group(0)
55
+ try:
56
+ return ESCAPE_DCT[s]
57
+ except KeyError:
58
+ n = ord(s)
59
+ if n < 0x10000:
60
+ #return '\\u{0:04x}'.format(n)
61
+ return '\\u%04x' % (n,)
62
+ else:
63
+ # surrogate pair
64
+ n -= 0x10000
65
+ s1 = 0xd800 | ((n >> 10) & 0x3ff)
66
+ s2 = 0xdc00 | (n & 0x3ff)
67
+ #return '\\u{0:04x}\\u{1:04x}'.format(s1, s2)
68
+ return '\\u%04x\\u%04x' % (s1, s2)
69
+ return '"' + str(ESCAPE_ASCII.sub(replace, s)) + '"'
70
+
71
+
72
+ encode_basestring_ascii = (
73
+ c_encode_basestring_ascii or py_encode_basestring_ascii)
74
+
75
+ class JSONEncoder(object):
76
+ """Extensible JSON <http://json.org> encoder for Python data structures.
77
+
78
+ Supports the following objects and types by default:
79
+
80
+ +-------------------+---------------+
81
+ | Python | JSON |
82
+ +===================+===============+
83
+ | dict, namedtuple | object |
84
+ +-------------------+---------------+
85
+ | list, tuple | array |
86
+ +-------------------+---------------+
87
+ | str, unicode | string |
88
+ +-------------------+---------------+
89
+ | int, long, float | number |
90
+ +-------------------+---------------+
91
+ | True | true |
92
+ +-------------------+---------------+
93
+ | False | false |
94
+ +-------------------+---------------+
95
+ | None | null |
96
+ +-------------------+---------------+
97
+
98
+ To extend this to recognize other objects, subclass and implement a
99
+ ``.default()`` method with another method that returns a serializable
100
+ object for ``o`` if possible, otherwise it should call the superclass
101
+ implementation (to raise ``TypeError``).
102
+
103
+ """
104
+ item_separator = ', '
105
+ key_separator = ': '
106
+ def __init__(self, skipkeys=False, ensure_ascii=True,
107
+ check_circular=True, allow_nan=True, sort_keys=False,
108
+ indent=None, separators=None, encoding='utf-8', default=None,
109
+ use_decimal=True, namedtuple_as_object=True,
110
+ tuple_as_array=True, bigint_as_string=False,
111
+ item_sort_key=None):
112
+ """Constructor for JSONEncoder, with sensible defaults.
113
+
114
+ If skipkeys is false, then it is a TypeError to attempt
115
+ encoding of keys that are not str, int, long, float or None. If
116
+ skipkeys is True, such items are simply skipped.
117
+
118
+ If ensure_ascii is true, the output is guaranteed to be str
119
+ objects with all incoming unicode characters escaped. If
120
+ ensure_ascii is false, the output will be unicode object.
121
+
122
+ If check_circular is true, then lists, dicts, and custom encoded
123
+ objects will be checked for circular references during encoding to
124
+ prevent an infinite recursion (which would cause an OverflowError).
125
+ Otherwise, no such check takes place.
126
+
127
+ If allow_nan is true, then NaN, Infinity, and -Infinity will be
128
+ encoded as such. This behavior is not JSON specification compliant,
129
+ but is consistent with most JavaScript based encoders and decoders.
130
+ Otherwise, it will be a ValueError to encode such floats.
131
+
132
+ If sort_keys is true, then the output of dictionaries will be
133
+ sorted by key; this is useful for regression tests to ensure
134
+ that JSON serializations can be compared on a day-to-day basis.
135
+
136
+ If indent is a string, then JSON array elements and object members
137
+ will be pretty-printed with a newline followed by that string repeated
138
+ for each level of nesting. ``None`` (the default) selects the most compact
139
+ representation without any newlines. For backwards compatibility with
140
+ versions of simplejson earlier than 2.1.0, an integer is also accepted
141
+ and is converted to a string with that many spaces.
142
+
143
+ If specified, separators should be a (item_separator, key_separator)
144
+ tuple. The default is (', ', ': '). To get the most compact JSON
145
+ representation you should specify (',', ':') to eliminate whitespace.
146
+
147
+ If specified, default is a function that gets called for objects
148
+ that can't otherwise be serialized. It should return a JSON encodable
149
+ version of the object or raise a ``TypeError``.
150
+
151
+ If encoding is not None, then all input strings will be
152
+ transformed into unicode using that encoding prior to JSON-encoding.
153
+ The default is UTF-8.
154
+
155
+ If use_decimal is true (not the default), ``decimal.Decimal`` will
156
+ be supported directly by the encoder. For the inverse, decode JSON
157
+ with ``parse_float=decimal.Decimal``.
158
+
159
+ If namedtuple_as_object is true (the default), objects with
160
+ ``_asdict()`` methods will be encoded as JSON objects.
161
+
162
+ If tuple_as_array is true (the default), tuple (and subclasses) will
163
+ be encoded as JSON arrays.
164
+
165
+ If bigint_as_string is true (not the default), ints 2**53 and higher
166
+ or lower than -2**53 will be encoded as strings. This is to avoid the
167
+ rounding that happens in Javascript otherwise.
168
+
169
+ If specified, item_sort_key is a callable used to sort the items in
170
+ each dictionary. This is useful if you want to sort items other than
171
+ in alphabetical order by key.
172
+ """
173
+
174
+ self.skipkeys = skipkeys
175
+ self.ensure_ascii = ensure_ascii
176
+ self.check_circular = check_circular
177
+ self.allow_nan = allow_nan
178
+ self.sort_keys = sort_keys
179
+ self.use_decimal = use_decimal
180
+ self.namedtuple_as_object = namedtuple_as_object
181
+ self.tuple_as_array = tuple_as_array
182
+ self.bigint_as_string = bigint_as_string
183
+ self.item_sort_key = item_sort_key
184
+ if indent is not None and not isinstance(indent, basestring):
185
+ indent = indent * ' '
186
+ self.indent = indent
187
+ if separators is not None:
188
+ self.item_separator, self.key_separator = separators
189
+ elif indent is not None:
190
+ self.item_separator = ','
191
+ if default is not None:
192
+ self.default = default
193
+ self.encoding = encoding
194
+
195
+ def default(self, o):
196
+ """Implement this method in a subclass such that it returns
197
+ a serializable object for ``o``, or calls the base implementation
198
+ (to raise a ``TypeError``).
199
+
200
+ For example, to support arbitrary iterators, you could
201
+ implement default like this::
202
+
203
+ def default(self, o):
204
+ try:
205
+ iterable = iter(o)
206
+ except TypeError:
207
+ pass
208
+ else:
209
+ return list(iterable)
210
+ return JSONEncoder.default(self, o)
211
+
212
+ """
213
+ raise TypeError(repr(o) + " is not JSON serializable")
214
+
215
+ def encode(self, o):
216
+ """Return a JSON string representation of a Python data structure.
217
+
218
+ >>> from simplejson import JSONEncoder
219
+ >>> JSONEncoder().encode({"foo": ["bar", "baz"]})
220
+ '{"foo": ["bar", "baz"]}'
221
+
222
+ """
223
+ # This is for extremely simple cases and benchmarks.
224
+ if isinstance(o, basestring):
225
+ if isinstance(o, str):
226
+ _encoding = self.encoding
227
+ if (_encoding is not None
228
+ and not (_encoding == 'utf-8')):
229
+ o = o.decode(_encoding)
230
+ if self.ensure_ascii:
231
+ return encode_basestring_ascii(o)
232
+ else:
233
+ return encode_basestring(o)
234
+ # This doesn't pass the iterator directly to ''.join() because the
235
+ # exceptions aren't as detailed. The list call should be roughly
236
+ # equivalent to the PySequence_Fast that ''.join() would do.
237
+ chunks = self.iterencode(o, _one_shot=True)
238
+ if not isinstance(chunks, (list, tuple)):
239
+ chunks = list(chunks)
240
+ if self.ensure_ascii:
241
+ return ''.join(chunks)
242
+ else:
243
+ return u''.join(chunks)
244
+
245
+ def iterencode(self, o, _one_shot=False):
246
+ """Encode the given object and yield each string
247
+ representation as available.
248
+
249
+ For example::
250
+
251
+ for chunk in JSONEncoder().iterencode(bigobject):
252
+ mysocket.write(chunk)
253
+
254
+ """
255
+ if self.check_circular:
256
+ markers = {}
257
+ else:
258
+ markers = None
259
+ if self.ensure_ascii:
260
+ _encoder = encode_basestring_ascii
261
+ else:
262
+ _encoder = encode_basestring
263
+ if self.encoding != 'utf-8':
264
+ def _encoder(o, _orig_encoder=_encoder, _encoding=self.encoding):
265
+ if isinstance(o, str):
266
+ o = o.decode(_encoding)
267
+ return _orig_encoder(o)
268
+
269
+ def floatstr(o, allow_nan=self.allow_nan,
270
+ _repr=FLOAT_REPR, _inf=PosInf, _neginf=-PosInf):
271
+ # Check for specials. Note that this type of test is processor
272
+ # and/or platform-specific, so do tests which don't depend on
273
+ # the internals.
274
+
275
+ if o != o:
276
+ text = 'NaN'
277
+ elif o == _inf:
278
+ text = 'Infinity'
279
+ elif o == _neginf:
280
+ text = '-Infinity'
281
+ else:
282
+ return _repr(o)
283
+
284
+ if not allow_nan:
285
+ raise ValueError(
286
+ "Out of range float values are not JSON compliant: " +
287
+ repr(o))
288
+
289
+ return text
290
+
291
+
292
+ key_memo = {}
293
+ if (_one_shot and c_make_encoder is not None
294
+ and self.indent is None):
295
+ _iterencode = c_make_encoder(
296
+ markers, self.default, _encoder, self.indent,
297
+ self.key_separator, self.item_separator, self.sort_keys,
298
+ self.skipkeys, self.allow_nan, key_memo, self.use_decimal,
299
+ self.namedtuple_as_object, self.tuple_as_array,
300
+ self.bigint_as_string, self.item_sort_key,
301
+ Decimal)
302
+ else:
303
+ _iterencode = _make_iterencode(
304
+ markers, self.default, _encoder, self.indent, floatstr,
305
+ self.key_separator, self.item_separator, self.sort_keys,
306
+ self.skipkeys, _one_shot, self.use_decimal,
307
+ self.namedtuple_as_object, self.tuple_as_array,
308
+ self.bigint_as_string, self.item_sort_key,
309
+ Decimal=Decimal)
310
+ try:
311
+ return _iterencode(o, 0)
312
+ finally:
313
+ key_memo.clear()
314
+
315
+
316
+ class JSONEncoderForHTML(JSONEncoder):
317
+ """An encoder that produces JSON safe to embed in HTML.
318
+
319
+ To embed JSON content in, say, a script tag on a web page, the
320
+ characters &, < and > should be escaped. They cannot be escaped
321
+ with the usual entities (e.g. &amp;) because they are not expanded
322
+ within <script> tags.
323
+ """
324
+
325
+ def encode(self, o):
326
+ # Override JSONEncoder.encode because it has hacks for
327
+ # performance that make things more complicated.
328
+ chunks = self.iterencode(o, True)
329
+ if self.ensure_ascii:
330
+ return ''.join(chunks)
331
+ else:
332
+ return u''.join(chunks)
333
+
334
+ def iterencode(self, o, _one_shot=False):
335
+ chunks = super(JSONEncoderForHTML, self).iterencode(o, _one_shot)
336
+ for chunk in chunks:
337
+ chunk = chunk.replace('&', '\\u0026')
338
+ chunk = chunk.replace('<', '\\u003c')
339
+ chunk = chunk.replace('>', '\\u003e')
340
+ yield chunk
341
+
342
+
343
+ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
344
+ _key_separator, _item_separator, _sort_keys, _skipkeys, _one_shot,
345
+ _use_decimal, _namedtuple_as_object, _tuple_as_array,
346
+ _bigint_as_string, _item_sort_key,
347
+ ## HACK: hand-optimized bytecode; turn globals into locals
348
+ False=False,
349
+ True=True,
350
+ ValueError=ValueError,
351
+ basestring=basestring,
352
+ Decimal=Decimal,
353
+ dict=dict,
354
+ float=float,
355
+ id=id,
356
+ int=int,
357
+ isinstance=isinstance,
358
+ list=list,
359
+ long=long,
360
+ str=str,
361
+ tuple=tuple,
362
+ ):
363
+ if _item_sort_key and not callable(_item_sort_key):
364
+ raise TypeError("item_sort_key must be None or callable")
365
+
366
+ def _iterencode_list(lst, _current_indent_level):
367
+ if not lst:
368
+ yield '[]'
369
+ return
370
+ if markers is not None:
371
+ markerid = id(lst)
372
+ if markerid in markers:
373
+ raise ValueError("Circular reference detected")
374
+ markers[markerid] = lst
375
+ buf = '['
376
+ if _indent is not None:
377
+ _current_indent_level += 1
378
+ newline_indent = '\n' + (_indent * _current_indent_level)
379
+ separator = _item_separator + newline_indent
380
+ buf += newline_indent
381
+ else:
382
+ newline_indent = None
383
+ separator = _item_separator
384
+ first = True
385
+ for value in lst:
386
+ if first:
387
+ first = False
388
+ else:
389
+ buf = separator
390
+ if isinstance(value, basestring):
391
+ yield buf + _encoder(value)
392
+ elif value is None:
393
+ yield buf + 'null'
394
+ elif value is True:
395
+ yield buf + 'true'
396
+ elif value is False:
397
+ yield buf + 'false'
398
+ elif isinstance(value, (int, long)):
399
+ yield ((buf + str(value))
400
+ if (not _bigint_as_string or
401
+ (-1 << 53) < value < (1 << 53))
402
+ else (buf + '"' + str(value) + '"'))
403
+ elif isinstance(value, float):
404
+ yield buf + _floatstr(value)
405
+ elif _use_decimal and isinstance(value, Decimal):
406
+ yield buf + str(value)
407
+ else:
408
+ yield buf
409
+ if isinstance(value, list):
410
+ chunks = _iterencode_list(value, _current_indent_level)
411
+ else:
412
+ _asdict = _namedtuple_as_object and getattr(value, '_asdict', None)
413
+ if _asdict and callable(_asdict):
414
+ chunks = _iterencode_dict(_asdict(),
415
+ _current_indent_level)
416
+ elif _tuple_as_array and isinstance(value, tuple):
417
+ chunks = _iterencode_list(value, _current_indent_level)
418
+ elif isinstance(value, dict):
419
+ chunks = _iterencode_dict(value, _current_indent_level)
420
+ else:
421
+ chunks = _iterencode(value, _current_indent_level)
422
+ for chunk in chunks:
423
+ yield chunk
424
+ if newline_indent is not None:
425
+ _current_indent_level -= 1
426
+ yield '\n' + (_indent * _current_indent_level)
427
+ yield ']'
428
+ if markers is not None:
429
+ del markers[markerid]
430
+
431
+ def _iterencode_dict(dct, _current_indent_level):
432
+ if not dct:
433
+ yield '{}'
434
+ return
435
+ if markers is not None:
436
+ markerid = id(dct)
437
+ if markerid in markers:
438
+ raise ValueError("Circular reference detected")
439
+ markers[markerid] = dct
440
+ yield '{'
441
+ if _indent is not None:
442
+ _current_indent_level += 1
443
+ newline_indent = '\n' + (_indent * _current_indent_level)
444
+ item_separator = _item_separator + newline_indent
445
+ yield newline_indent
446
+ else:
447
+ newline_indent = None
448
+ item_separator = _item_separator
449
+ first = True
450
+ if _item_sort_key:
451
+ items = dct.items()
452
+ items.sort(key=_item_sort_key)
453
+ elif _sort_keys:
454
+ items = dct.items()
455
+ items.sort(key=lambda kv: kv[0])
456
+ else:
457
+ items = dct.iteritems()
458
+ for key, value in items:
459
+ if isinstance(key, basestring):
460
+ pass
461
+ # JavaScript is weakly typed for these, so it makes sense to
462
+ # also allow them. Many encoders seem to do something like this.
463
+ elif isinstance(key, float):
464
+ key = _floatstr(key)
465
+ elif key is True:
466
+ key = 'true'
467
+ elif key is False:
468
+ key = 'false'
469
+ elif key is None:
470
+ key = 'null'
471
+ elif isinstance(key, (int, long)):
472
+ key = str(key)
473
+ elif _skipkeys:
474
+ continue
475
+ else:
476
+ raise TypeError("key " + repr(key) + " is not a string")
477
+ if first:
478
+ first = False
479
+ else:
480
+ yield item_separator
481
+ yield _encoder(key)
482
+ yield _key_separator
483
+ if isinstance(value, basestring):
484
+ yield _encoder(value)
485
+ elif value is None:
486
+ yield 'null'
487
+ elif value is True:
488
+ yield 'true'
489
+ elif value is False:
490
+ yield 'false'
491
+ elif isinstance(value, (int, long)):
492
+ yield (str(value)
493
+ if (not _bigint_as_string or
494
+ (-1 << 53) < value < (1 << 53))
495
+ else ('"' + str(value) + '"'))
496
+ elif isinstance(value, float):
497
+ yield _floatstr(value)
498
+ elif _use_decimal and isinstance(value, Decimal):
499
+ yield str(value)
500
+ else:
501
+ if isinstance(value, list):
502
+ chunks = _iterencode_list(value, _current_indent_level)
503
+ else:
504
+ _asdict = _namedtuple_as_object and getattr(value, '_asdict', None)
505
+ if _asdict and callable(_asdict):
506
+ chunks = _iterencode_dict(_asdict(),
507
+ _current_indent_level)
508
+ elif _tuple_as_array and isinstance(value, tuple):
509
+ chunks = _iterencode_list(value, _current_indent_level)
510
+ elif isinstance(value, dict):
511
+ chunks = _iterencode_dict(value, _current_indent_level)
512
+ else:
513
+ chunks = _iterencode(value, _current_indent_level)
514
+ for chunk in chunks:
515
+ yield chunk
516
+ if newline_indent is not None:
517
+ _current_indent_level -= 1
518
+ yield '\n' + (_indent * _current_indent_level)
519
+ yield '}'
520
+ if markers is not None:
521
+ del markers[markerid]
522
+
523
+ def _iterencode(o, _current_indent_level):
524
+ if isinstance(o, basestring):
525
+ yield _encoder(o)
526
+ elif o is None:
527
+ yield 'null'
528
+ elif o is True:
529
+ yield 'true'
530
+ elif o is False:
531
+ yield 'false'
532
+ elif isinstance(o, (int, long)):
533
+ yield (str(o)
534
+ if (not _bigint_as_string or
535
+ (-1 << 53) < o < (1 << 53))
536
+ else ('"' + str(o) + '"'))
537
+ elif isinstance(o, float):
538
+ yield _floatstr(o)
539
+ elif isinstance(o, list):
540
+ for chunk in _iterencode_list(o, _current_indent_level):
541
+ yield chunk
542
+ else:
543
+ _asdict = _namedtuple_as_object and getattr(o, '_asdict', None)
544
+ if _asdict and callable(_asdict):
545
+ for chunk in _iterencode_dict(_asdict(), _current_indent_level):
546
+ yield chunk
547
+ elif (_tuple_as_array and isinstance(o, tuple)):
548
+ for chunk in _iterencode_list(o, _current_indent_level):
549
+ yield chunk
550
+ elif isinstance(o, dict):
551
+ for chunk in _iterencode_dict(o, _current_indent_level):
552
+ yield chunk
553
+ elif _use_decimal and isinstance(o, Decimal):
554
+ yield str(o)
555
+ else:
556
+ if markers is not None:
557
+ markerid = id(o)
558
+ if markerid in markers:
559
+ raise ValueError("Circular reference detected")
560
+ markers[markerid] = o
561
+ o = _default(o)
562
+ for chunk in _iterencode(o, _current_indent_level):
563
+ yield chunk
564
+ if markers is not None:
565
+ del markers[markerid]
566
+
567
+ return _iterencode