Mercurial > cpython-lru_cache-weakref

annotate Lib/tokenize.py @ 58765:2ba7677bfa95 legacy-trunk

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
testCopysign was defined twice in test_math; combine the definitions
author Mark Dickinson <dickinsm@gmail.com>
date Sat, 06 Feb 2010 23:11:25 +0000
parents e00c78869b3c
children b36cb4602e21
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
5173
ff4f8fa8a273 Fixed doc string, added __version__, fixed 1 bug.
Guido van Rossum <guido@python.org>
parents: 5172
diff changeset
1 """Tokenization help for Python programs.
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
2
18351
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
3 generate_tokens(readline) is a generator that breaks a stream of
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
4 text into Python tokens. It accepts a readline-like method which is called
18351
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
5 repeatedly to get the next line of input (or "" for EOF). It generates
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
6 5-tuples with these members:
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
7
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
8 the token type (see token.py)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
9 the token (a string)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
10 the starting (row, column) indices of the token (a 2-tuple of ints)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
11 the ending (row, column) indices of the token (a 2-tuple of ints)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
12 the original line (string)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
13
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
14 It is designed to match the working of the Python tokenizer exactly, except
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
15 that it produces COMMENT tokens for comments and gives type OP for all
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
16 operators
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
17
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
18 Older entry points
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
19 tokenize_loop(readline, tokeneater)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
20 tokenize(readline, tokeneater=printtoken)
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
21 are the same, except instead of generating tokens, tokeneater is a callback
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
22 function to which the 5 fields described above are passed as 5 arguments,
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
23 each time a new token is found."""
5173
ff4f8fa8a273 Fixed doc string, added __version__, fixed 1 bug.
Guido van Rossum <guido@python.org>
parents: 5172
diff changeset
24
16899
320fa50a1a7d Add __author__ and __credits__ variables.
Ka-Ping Yee <ping@zesty.ca>
parents: 16889
diff changeset
25 __author__ = 'Ka-Ping Yee <ping@lfw.org>'
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
26 __credits__ = ('GvR, ESR, Tim Peters, Thomas Wouters, Fred Drake, '
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
27 'Skip Montanaro, Raymond Hettinger')
5173
ff4f8fa8a273 Fixed doc string, added __version__, fixed 1 bug.
Guido van Rossum <guido@python.org>
parents: 5172
diff changeset
28
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
29 import string, re
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
30 from token import *
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
31
16889
0feb63490093 final round of __all__ lists (I hope) - skipped urllib2 because Moshe may be
Skip Montanaro <skip@pobox.com>
parents: 16514
diff changeset
32 import token
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
33 __all__ = [x for x in dir(token) if not x.startswith("_")]
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
34 __all__ += ["COMMENT", "tokenize", "generate_tokens", "NL", "untokenize"]
22585
93daa4fde087 Cleanup x so it is not left in module
Neal Norwitz <nnorwitz@gmail.com>
parents: 19621
diff changeset
35 del x
16889
0feb63490093 final round of __all__ lists (I hope) - skipped urllib2 because Moshe may be
Skip Montanaro <skip@pobox.com>
parents: 16514
diff changeset
36 del token
0feb63490093 final round of __all__ lists (I hope) - skipped urllib2 because Moshe may be
Skip Montanaro <skip@pobox.com>
parents: 16514
diff changeset
37
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
38 COMMENT = N_TOKENS
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
39 tok_name[COMMENT] = 'COMMENT'
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
40 NL = N_TOKENS + 1
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
41 tok_name[NL] = 'NL'
16889
0feb63490093 final round of __all__ lists (I hope) - skipped urllib2 because Moshe may be
Skip Montanaro <skip@pobox.com>
parents: 16514
diff changeset
42 N_TOKENS += 2
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
43
16514
90c600a819e9 String method conversion.
Eric S. Raymond <esr@thyrsus.com>
parents: 15898
diff changeset
44 def group(*choices): return '(' + '|'.join(choices) + ')'
28089
14f65d661084 Get rid of many apply() calls.
Guido van Rossum <guido@python.org>
parents: 26032
diff changeset
45 def any(*choices): return group(*choices) + '*'
14f65d661084 Get rid of many apply() calls.
Guido van Rossum <guido@python.org>
parents: 26032
diff changeset
46 def maybe(*choices): return group(*choices) + '?'
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
47
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
48 Whitespace = r'[ \f\t]*'
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
49 Comment = r'#[^\r\n]*'
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
50 Ignore = Whitespace + any(r'\\\r?\n' + Whitespace) + maybe(Comment)
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
51 Name = r'[a-zA-Z_]\w*'
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
52
44780
61a11cc95632 Fix #1679: "0x" was taken as a valid integer literal.
Georg Brandl <georg@python.org>
parents: 44747
diff changeset
53 Hexnumber = r'0[xX][\da-fA-F]+[lL]?'
45801
b5cb7a678f38 Added PEP 3127 support to tokenize (with tests); added PEP 3127 to NEWS.
Eric Smith <eric@trueblade.com>
parents: 44780
diff changeset
54 Octnumber = r'(0[oO][0-7]+)|(0[0-7]*)[lL]?'
b5cb7a678f38 Added PEP 3127 support to tokenize (with tests); added PEP 3127 to NEWS.
Eric Smith <eric@trueblade.com>
parents: 44780
diff changeset
55 Binnumber = r'0[bB][01]+[lL]?'
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
56 Decnumber = r'[1-9]\d*[lL]?'
45801
b5cb7a678f38 Added PEP 3127 support to tokenize (with tests); added PEP 3127 to NEWS.
Eric Smith <eric@trueblade.com>
parents: 44780
diff changeset
57 Intnumber = group(Hexnumber, Binnumber, Octnumber, Decnumber)
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
58 Exponent = r'[eE][-+]?\d+'
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
59 Pointfloat = group(r'\d+\.\d*', r'\.\d+') + maybe(Exponent)
19621
0bec699783d0 SF patch #455966: Allow leading 0 in float/imag literals.
Tim Peters <tim.peters@gmail.com>
parents: 19052
diff changeset
60 Expfloat = r'\d+' + Exponent
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
61 Floatnumber = group(Pointfloat, Expfloat)
19621
0bec699783d0 SF patch #455966: Allow leading 0 in float/imag literals.
Tim Peters <tim.peters@gmail.com>
parents: 19052
diff changeset
62 Imagnumber = group(r'\d+[jJ]', Floatnumber + r'[jJ]')
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
63 Number = group(Imagnumber, Floatnumber, Intnumber)
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
64
15080
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
65 # Tail end of ' string.
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
66 Single = r"[^'\\]*(?:\\.[^'\\]*)*'"
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
67 # Tail end of " string.
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
68 Double = r'[^"\\]*(?:\\.[^"\\]*)*"'
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
69 # Tail end of ''' string.
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
70 Single3 = r"[^'\\]*(?:(?:\\.|'(?!''))[^'\\]*)*'''"
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
71 # Tail end of """ string.
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
72 Double3 = r'[^"\\]*(?:(?:\\.|"(?!""))[^"\\]*)*"""'
15898
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
73 Triple = group("[uU]?[rR]?'''", '[uU]?[rR]?"""')
15080
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
74 # Single-line ' or " string.
15898
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
75 String = group(r"[uU]?[rR]?'[^\n'\\]*(?:\\.[^\n'\\]*)*'",
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
76 r'[uU]?[rR]?"[^\n"\\]*(?:\\.[^\n"\\]*)*"')
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
77
15080
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
78 # Because of leftmost-then-longest match semantics, be sure to put the
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
79 # longest operators first (e.g., if = came before ==, == would get
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
80 # recognized as two instances of =).
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
81 Operator = group(r"\*\*=?", r">>=?", r"<<=?", r"<>", r"!=",
19052
3969fcbf9500 Add new tokens // and //=, in support of PEP 238.
Guido van Rossum <guido@python.org>
parents: 18781
diff changeset
82 r"//=?",
15080
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
83 r"[+\-*/%&|^=<>]=?",
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
84 r"~")
14166
e9c2f830f9ba Update for augmented assignment, tested & approved by Guido.
Thomas Wouters <thomas@python.org>
parents: 14021
diff changeset
85
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
86 Bracket = '[][(){}]'
32888
c7e03cf8333e PEP-0318, @decorator-style. In Guido's words:
Anthony Baxter <anthonybaxter@gmail.com>
parents: 28089
diff changeset
87 Special = group(r'\r?\n', r'[:;.,`@]')
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
88 Funny = group(Operator, Bracket, Special)
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
89
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
90 PlainToken = group(Number, Funny, String, Name)
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
91 Token = Ignore + PlainToken
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
92
15080
0e0ea56e4be8 Possible fix for Skip's bug 116136 (sre recursion limit hit in tokenize.py).
Tim Peters <tim.peters@gmail.com>
parents: 14166
diff changeset
93 # First (or only) line of ' or " string.
15898
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
94 ContStr = group(r"[uU]?[rR]?'[^\n'\\]*(?:\\.[^\n'\\]*)*" +
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
95 group("'", r'\\\r?\n'),
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
96 r'[uU]?[rR]?"[^\n"\\]*(?:\\.[^\n"\\]*)*' +
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
97 group('"', r'\\\r?\n'))
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
98 PseudoExtras = group(r'\\\r?\n', Comment, Triple)
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
99 PseudoToken = Whitespace + group(PseudoExtras, Number, Funny, ContStr, Name)
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
100
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
101 tokenprog, pseudoprog, single3prog, double3prog = map(
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
102 re.compile, (Token, PseudoToken, Single3, Double3))
6532
d369629e8c27 New, fixed version with proper r"..." and R"..." support from Ka-Ping.
Guido van Rossum <guido@python.org>
parents: 6531
diff changeset
103 endprogs = {"'": re.compile(Single), '"': re.compile(Double),
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
104 "'''": single3prog, '"""': double3prog,
6532
d369629e8c27 New, fixed version with proper r"..." and R"..." support from Ka-Ping.
Guido van Rossum <guido@python.org>
parents: 6531
diff changeset
105 "r'''": single3prog, 'r"""': double3prog,
15898
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
106 "u'''": single3prog, 'u"""': double3prog,
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
107 "ur'''": single3prog, 'ur"""': double3prog,
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
108 "R'''": single3prog, 'R"""': double3prog,
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
109 "U'''": single3prog, 'U"""': double3prog,
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
110 "uR'''": single3prog, 'uR"""': double3prog,
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
111 "Ur'''": single3prog, 'Ur"""': double3prog,
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
112 "UR'''": single3prog, 'UR"""': double3prog,
44747
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
113 "b'''": single3prog, 'b"""': double3prog,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
114 "br'''": single3prog, 'br"""': double3prog,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
115 "B'''": single3prog, 'B"""': double3prog,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
116 "bR'''": single3prog, 'bR"""': double3prog,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
117 "Br'''": single3prog, 'Br"""': double3prog,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
118 "BR'''": single3prog, 'BR"""': double3prog,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
119 'r': None, 'R': None, 'u': None, 'U': None,
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
120 'b': None, 'B': None}
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
121
25016
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
122 triple_quoted = {}
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
123 for t in ("'''", '"""',
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
124 "r'''", 'r"""', "R'''", 'R"""',
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
125 "u'''", 'u"""', "U'''", 'U"""',
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
126 "ur'''", 'ur"""', "Ur'''", 'Ur"""',
44747
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
127 "uR'''", 'uR"""', "UR'''", 'UR"""',
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
128 "b'''", 'b"""', "B'''", 'B"""',
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
129 "br'''", 'br"""', "Br'''", 'Br"""',
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
130 "bR'''", 'bR"""', "BR'''", 'BR"""'):
25016
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
131 triple_quoted[t] = t
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
132 single_quoted = {}
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
133 for t in ("'", '"',
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
134 "r'", 'r"', "R'", 'R"',
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
135 "u'", 'u"', "U'", 'U"',
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
136 "ur'", 'ur"', "Ur'", 'Ur"',
44747
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
137 "uR'", 'uR"', "UR'", 'UR"',
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
138 "b'", 'b"', "B'", 'B"',
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
139 "br'", 'br"', "Br'", 'Br"',
8da9ebf73026 Added bytes and b'' as aliases for str and ''
Christian Heimes <christian@cheimes.de>
parents: 40414
diff changeset
140 "bR'", 'bR"', "BR'", 'BR"' ):
25016
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
141 single_quoted[t] = t
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
142
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
143 tabsize = 8
14021
5b34b70ecf22 Convert some old-style string exceptions to class exceptions.
Fred Drake <fdrake@acm.org>
parents: 7839
diff changeset
144
17178
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
145 class TokenError(Exception): pass
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
146
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
147 class StopTokenizing(Exception): pass
14021
5b34b70ecf22 Convert some old-style string exceptions to class exceptions.
Fred Drake <fdrake@acm.org>
parents: 7839
diff changeset
148
48132
4b1ba50d3bdf Remove a tuple unpacking in a parameter list to remove a SyntaxWarning raised
Brett Cannon <bcannon@gmail.com>
parents: 47480
diff changeset
149 def printtoken(type, token, srow_scol, erow_ecol, line): # for testing
4b1ba50d3bdf Remove a tuple unpacking in a parameter list to remove a SyntaxWarning raised
Brett Cannon <bcannon@gmail.com>
parents: 47480
diff changeset
150 srow, scol = srow_scol
4b1ba50d3bdf Remove a tuple unpacking in a parameter list to remove a SyntaxWarning raised
Brett Cannon <bcannon@gmail.com>
parents: 47480
diff changeset
151 erow, ecol = erow_ecol
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
152 print "%d,%d-%d,%d:\t%s\t%s" % \
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
153 (srow, scol, erow, ecol, tok_name[type], repr(token))
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
154
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
155 def tokenize(readline, tokeneater=printtoken):
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
156 """
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
157 The tokenize() function accepts two parameters: one representing the
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
158 input stream, and one providing an output mechanism for tokenize().
23521
a90faa5f9417 Whitespace normalization.
Tim Peters <tim.peters@gmail.com>
parents: 23455
diff changeset
159
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
160 The first parameter, readline, must be a callable object which provides
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
161 the same interface as the readline() method of built-in file objects.
23521
a90faa5f9417 Whitespace normalization.
Tim Peters <tim.peters@gmail.com>
parents: 23455
diff changeset
162 Each call to the function should return one line of input as a string.
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
163
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
164 The second parameter, tokeneater, must also be a callable object. It is
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
165 called once for each token, with five arguments, corresponding to the
23521
a90faa5f9417 Whitespace normalization.
Tim Peters <tim.peters@gmail.com>
parents: 23455
diff changeset
166 tuples generated by generate_tokens().
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
167 """
17178
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
168 try:
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
169 tokenize_loop(readline, tokeneater)
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
170 except StopTokenizing:
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
171 pass
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
172
18351
44a87c300f2b Turns out Neil didn't intend for *all* of his gen-branch work to get
Tim Peters <tim.peters@gmail.com>
parents: 18148
diff changeset
173 # backwards compatible interface
17178
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
174 def tokenize_loop(readline, tokeneater):
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
175 for token_info in generate_tokens(readline):
28089
14f65d661084 Get rid of many apply() calls.
Guido van Rossum <guido@python.org>
parents: 26032
diff changeset
176 tokeneater(*token_info)
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
177
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
178 class Untokenizer:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
179
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
180 def __init__(self):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
181 self.tokens = []
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
182 self.prev_row = 1
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
183 self.prev_col = 0
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
184
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
185 def add_whitespace(self, start):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
186 row, col = start
39641
f45451819491 Replace dead code with an assert.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39640
diff changeset
187 assert row <= self.prev_row
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
188 col_offset = col - self.prev_col
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
189 if col_offset:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
190 self.tokens.append(" " * col_offset)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
191
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
192 def untokenize(self, iterable):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
193 for t in iterable:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
194 if len(t) == 2:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
195 self.compat(t, iterable)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
196 break
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
197 tok_type, token, start, end, line = t
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
198 self.add_whitespace(start)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
199 self.tokens.append(token)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
200 self.prev_row, self.prev_col = end
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
201 if tok_type in (NEWLINE, NL):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
202 self.prev_row += 1
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
203 self.prev_col = 0
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
204 return "".join(self.tokens)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
205
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
206 def compat(self, token, iterable):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
207 startline = False
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
208 indents = []
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
209 toks_append = self.tokens.append
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
210 toknum, tokval = token
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
211 if toknum in (NAME, NUMBER):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
212 tokval += ' '
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
213 if toknum in (NEWLINE, NL):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
214 startline = True
46142
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
215 prevstring = False
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
216 for tok in iterable:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
217 toknum, tokval = tok[:2]
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
218
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
219 if toknum in (NAME, NUMBER):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
220 tokval += ' '
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
221
46142
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
222 # Insert a space between two consecutive strings
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
223 if toknum == STRING:
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
224 if prevstring:
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
225 tokval = ' ' + tokval
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
226 prevstring = True
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
227 else:
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
228 prevstring = False
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
229
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
230 if toknum == INDENT:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
231 indents.append(tokval)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
232 continue
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
233 elif toknum == DEDENT:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
234 indents.pop()
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
235 continue
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
236 elif toknum in (NEWLINE, NL):
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
237 startline = True
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
238 elif startline and indents:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
239 toks_append(indents[-1])
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
240 startline = False
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
241 toks_append(tokval)
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
242
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
243 def untokenize(iterable):
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
244 """Transform tokens back into Python source code.
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
245
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
246 Each element returned by the iterable must be a token sequence
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
247 with at least two elements, a token number and token value. If
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
248 only two tokens are passed, the resulting output is poor.
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
249
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
250 Round-trip invariant for full input:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
251 Untokenized source will match input source exactly
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
252
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
253 Round-trip invariant for limited intput:
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
254 # Output text will tokenize the back to the input
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
255 t1 = [tok[:2] for tok in generate_tokens(f.readline)]
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
256 newcode = untokenize(t1)
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
257 readline = iter(newcode.splitlines(1)).next
46142
31ba3fa1585b Issue2495: tokenize.untokenize did not insert space between two consecutive string literals:
Amaury Forgeot d'Arc <amauryfa@gmail.com>
parents: 45801
diff changeset
258 t2 = [tok[:2] for tok in generate_tokens(readline)]
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
259 assert t1 == t2
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
260 """
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
261 ut = Untokenizer()
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
262 return ut.untokenize(iterable)
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
263
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
264 def generate_tokens(readline):
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
265 """
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
266 The generate_tokens() generator requires one argment, readline, which
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
267 must be a callable object which provides the same interface as the
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
268 readline() method of built-in file objects. Each call to the function
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
269 should return one line of input as a string. Alternately, readline
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
270 can be a callable function terminating with StopIteration:
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
271 readline = open(myfile).next # Example of alternate readline
23521
a90faa5f9417 Whitespace normalization.
Tim Peters <tim.peters@gmail.com>
parents: 23455
diff changeset
272
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
273 The generator produces 5-tuples with these members: the token type; the
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
274 token string; a 2-tuple (srow, scol) of ints specifying the row and
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
275 column where the token begins in the source; a 2-tuple (erow, ecol) of
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
276 ints specifying the row and column where the token ends in the source;
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
277 and the line on which the token was found. The line passed is the
23521
a90faa5f9417 Whitespace normalization.
Tim Peters <tim.peters@gmail.com>
parents: 23455
diff changeset
278 logical line; continuation lines are included.
23455
68c776ba5ea5 Added docstrings excerpted from Python Library Reference.
Raymond Hettinger <python@rcn.com>
parents: 22719
diff changeset
279 """
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
280 lnum = parenlev = continued = 0
18781
01ae962df885 Use string.ascii_letters instead of string.letters (SF bug #226706).
Fred Drake <fdrake@acm.org>
parents: 18609
diff changeset
281 namechars, numchars = string.ascii_letters + '_', '0123456789'
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
282 contstr, needcont = '', 0
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
283 contline = None
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
284 indents = [0]
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
285
47480
2a496f1eb9f1 revert 63965 for preformance reasons
Benjamin Peterson <benjamin@python.org>
parents: 47478
diff changeset
286 while 1: # loop over lines in stream
35058
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
287 try:
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
288 line = readline()
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
289 except StopIteration:
8076424d455d Add untokenize() function to allow full round-trip tokenization.
Raymond Hettinger <python@rcn.com>
parents: 32888
diff changeset
290 line = ''
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
291 lnum += 1
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
292 pos, max = 0, len(line)
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
293
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
294 if contstr: # continued string
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
295 if not line:
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
296 raise TokenError, ("EOF in multi-line string", strstart)
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
297 endmatch = endprog.match(line)
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
298 if endmatch:
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
299 pos = end = endmatch.end(0)
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
300 yield (STRING, contstr + line[:end],
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
301 strstart, (lnum, end), contline + line)
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
302 contstr, needcont = '', 0
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
303 contline = None
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
304 elif needcont and line[-2:] != '\\\n' and line[-3:] != '\\\r\n':
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
305 yield (ERRORTOKEN, contstr + line,
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
306 strstart, (lnum, len(line)), contline)
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
307 contstr = ''
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
308 contline = None
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
309 continue
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
310 else:
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
311 contstr = contstr + line
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
312 contline = contline + line
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
313 continue
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
314
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
315 elif parenlev == 0 and not continued: # new statement
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
316 if not line: break
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
317 column = 0
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
318 while pos < max: # measure leading whitespace
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
319 if line[pos] == ' ':
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
320 column += 1
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
321 elif line[pos] == '\t':
56535
45b62725377a use floor division and add a test that exercises the tabsize codepath
Benjamin Peterson <benjamin@python.org>
parents: 56534
diff changeset
322 column = (column//tabsize + 1)*tabsize
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
323 elif line[pos] == '\f':
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
324 column = 0
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
325 else:
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
326 break
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
327 pos += 1
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
328 if pos == max:
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
329 break
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
330
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
331 if line[pos] in '#\r\n': # skip comments or blank lines
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
332 if line[pos] == '#':
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
333 comment_token = line[pos:].rstrip('\r\n')
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
334 nl_pos = pos + len(comment_token)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
335 yield (COMMENT, comment_token,
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
336 (lnum, pos), (lnum, pos + len(comment_token)), line)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
337 yield (NL, line[nl_pos:],
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
338 (lnum, nl_pos), (lnum, len(line)), line)
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
339 else:
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
340 yield ((NL, COMMENT)[line[pos] == '#'], line[pos:],
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
341 (lnum, pos), (lnum, len(line)), line)
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
342 continue
558
cfb53322c9d1 Initial revision
Guido van Rossum <guido@python.org>
parents:
diff changeset
343
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
344 if column > indents[-1]: # count indents or dedents
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
345 indents.append(column)
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
346 yield (INDENT, line[:pos], (lnum, 0), (lnum, pos), line)
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
347 while column < indents[-1]:
35119
04868959f1fe SF bug #1224621: tokenize module does not detect inconsistent dedents
Raymond Hettinger <python@rcn.com>
parents: 35058
diff changeset
348 if column not in indents:
04868959f1fe SF bug #1224621: tokenize module does not detect inconsistent dedents
Raymond Hettinger <python@rcn.com>
parents: 35058
diff changeset
349 raise IndentationError(
39495
14ec9cfd9760 Make tabnanny recognize IndentationErrors raised by tokenize.
Georg Brandl <georg@python.org>
parents: 37314
diff changeset
350 "unindent does not match any outer indentation level",
14ec9cfd9760 Make tabnanny recognize IndentationErrors raised by tokenize.
Georg Brandl <georg@python.org>
parents: 37314
diff changeset
351 ("<tokenize>", lnum, pos, line))
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
352 indents = indents[:-1]
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
353 yield (DEDENT, '', (lnum, pos), (lnum, pos), line)
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
354
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
355 else: # continued statement
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
356 if not line:
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
357 raise TokenError, ("EOF in multi-line statement", (lnum, 0))
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
358 continued = 0
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
359
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
360 while pos < max:
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
361 pseudomatch = pseudoprog.match(line, pos)
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
362 if pseudomatch: # scan for tokens
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
363 start, end = pseudomatch.span(1)
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
364 spos, epos, pos = (lnum, start), (lnum, end), end
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
365 token, initial = line[start:end], line[start]
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
366
17178
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
367 if initial in numchars or \
707f95d05e69 Provide a StopTokenizing exception for conveniently exiting the loop.
Ka-Ping Yee <ping@zesty.ca>
parents: 16901
diff changeset
368 (initial == '.' and token != '.'): # ordinary number
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
369 yield (NUMBER, token, spos, epos, line)
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
370 elif initial in '\r\n':
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
371 yield (NL if parenlev > 0 else NEWLINE,
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
372 token, spos, epos, line)
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
373 elif initial == '#':
39640
5cce028874bc Bug fixes large and small for tokenize.
Jeremy Hylton <jeremy@alum.mit.edu>
parents: 39495
diff changeset
374 assert not token.endswith("\n")
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
375 yield (COMMENT, token, spos, epos, line)
25016
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
376 elif token in triple_quoted:
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
377 endprog = endprogs[token]
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
378 endmatch = endprog.match(line, pos)
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
379 if endmatch: # all on one line
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
380 pos = endmatch.end(0)
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
381 token = line[start:pos]
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
382 yield (STRING, token, spos, (lnum, pos), line)
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
383 else:
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
384 strstart = (lnum, start) # multiple lines
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
385 contstr = line[start:]
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
386 contline = line
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
387 break
25016
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
388 elif initial in single_quoted or \
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
389 token[:2] in single_quoted or \
0ebec83eb3dd Speed up the most egregious "if token in (long tuple)" cases by using
Guido van Rossum <guido@python.org>
parents: 23521
diff changeset
390 token[:3] in single_quoted:
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
391 if token[-1] == '\n': # continued string
5266
1efc4273fdb7 Ka-Ping's muich improved version of March 26, 1997:
Guido van Rossum <guido@python.org>
parents: 5178
diff changeset
392 strstart = (lnum, start)
15898
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
393 endprog = (endprogs[initial] or endprogs[token[1]] or
8366fcfb6090 Add tokenizer support and tests for u'', U"", uR'', Ur"", etc.
Ka-Ping Yee <ping@zesty.ca>
parents: 15873
diff changeset
394 endprogs[token[2]])
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
395 contstr, needcont = line[start:], 1
7839
d4f5321b9d9e Differentiate between NEWLINE token (an official newline) and NL token
Guido van Rossum <guido@python.org>
parents: 6532
diff changeset
396 contline = line
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
397 break
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
398 else: # ordinary string
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
399 yield (STRING, token, spos, epos, line)
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
400 elif initial in namechars: # ordinary name
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
401 yield (NAME, token, spos, epos, line)
6531
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
402 elif initial == '\\': # continued stmt
e497a7fc0aa8 Redone (by Ka-Ping) using the new re module, and adding recognition
Guido van Rossum <guido@python.org>
parents: 5721
diff changeset
403 continued = 1
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
404 else:
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
405 if initial in '([{':
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
406 parenlev += 1
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
407 elif initial in ')]}':
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
408 parenlev -= 1
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
409 yield (OP, token, spos, epos, line)
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
410 else:
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
411 yield (ERRORTOKEN, line[pos],
5277
7df5a0d1f523 Ping's latest. Fixes triple quoted strings ending in odd
Guido van Rossum <guido@python.org>
parents: 5266
diff changeset
412 (lnum, pos), (lnum, pos+1), line)
56537
e00c78869b3c some cleanups
Benjamin Peterson <benjamin@python.org>
parents: 56535
diff changeset
413 pos += 1
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
414
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
415 for indent in indents[1:]: # pop remaining indent levels
18148
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
416 yield (DEDENT, '', (lnum, 0), (lnum, 0), '')
7bc473b2c654 Merging the gen-branch into the main line, at Guido's direction. Yay!
Tim Peters <tim.peters@gmail.com>
parents: 17178
diff changeset
417 yield (ENDMARKER, '', (lnum, 0), (lnum, 0), '')
5172
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
418
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
419 if __name__ == '__main__': # testing
31b0e5f0badf Ka-Ping's version.
Guido van Rossum <guido@python.org>
parents: 1407
diff changeset
420 import sys
56534
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
421 if len(sys.argv) > 1:
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
422 tokenize(open(sys.argv[1]).readline)
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
423 else:
fa6188c47b63 pep8ify if blocks
Benjamin Peterson <benjamin@python.org>
parents: 48132
diff changeset
424 tokenize(sys.stdin.readline)