summaryrefslogtreecommitdiff
path: root/pygments/util.py
blob: 45070063710cd77322ee60d09e5d47e298a351f4 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
# -*- coding: utf-8 -*-
"""
    pygments.util
    ~~~~~~~~~~~~~

    Utility functions.

    :copyright: Copyright 2006-2017 by the Pygments team, see AUTHORS.
    :license: BSD, see LICENSE for details.
"""

import re
import sys


split_path_re = re.compile(r'[/\\ ]')
doctype_lookup_re = re.compile(r'''
    (<\?.*?\?>)?\s*
    <!DOCTYPE\s+(
     [a-zA-Z_][a-zA-Z0-9]*
     (?: \s+      # optional in HTML5
     [a-zA-Z_][a-zA-Z0-9]*\s+
     "[^"]*")?
     )
     [^>]*>
''', re.DOTALL | re.MULTILINE | re.VERBOSE)
tag_re = re.compile(r'<(.+?)(\s.*?)?>.*?</.+?>',
                    re.UNICODE | re.IGNORECASE | re.DOTALL | re.MULTILINE)
xml_decl_re = re.compile(r'\s*<\?xml[^>]*\?>', re.I)


class ClassNotFound(ValueError):
    """Raised if one of the lookup functions didn't find a matching class."""


class OptionError(Exception):
    pass


def get_choice_opt(options, optname, allowed, default=None, normcase=False):
    string = options.get(optname, default)
    if normcase:
        string = string.lower()
    if string not in allowed:
        raise OptionError('Value for option %s must be one of %s' %
                          (optname, ', '.join(map(str, allowed))))
    return string


def get_bool_opt(options, optname, default=None):
    string = options.get(optname, default)
    if isinstance(string, bool):
        return string
    elif isinstance(string, int):
        return bool(string)
    elif not isinstance(string, string_types):
        raise OptionError('Invalid type %r for option %s; use '
                          '1/0, yes/no, true/false, on/off' % (
                              string, optname))
    elif string.lower() in ('1', 'yes', 'true', 'on'):
        return True
    elif string.lower() in ('0', 'no', 'false', 'off'):
        return False
    else:
        raise OptionError('Invalid value %r for option %s; use '
                          '1/0, yes/no, true/false, on/off' % (
                              string, optname))


def get_int_opt(options, optname, default=None):
    string = options.get(optname, default)
    try:
        return int(string)
    except TypeError:
        raise OptionError('Invalid type %r for option %s; you '
                          'must give an integer value' % (
                              string, optname))
    except ValueError:
        raise OptionError('Invalid value %r for option %s; you '
                          'must give an integer value' % (
                              string, optname))


def get_list_opt(options, optname, default=None):
    val = options.get(optname, default)
    if isinstance(val, string_types):
        return val.split()
    elif isinstance(val, (list, tuple)):
        return list(val)
    else:
        raise OptionError('Invalid type %r for option %s; you '
                          'must give a list value' % (
                              val, optname))


def docstring_headline(obj):
    if not obj.__doc__:
        return ''
    res = []
    for line in obj.__doc__.strip().splitlines():
        if line.strip():
            res.append(" " + line.strip())
        else:
            break
    return ''.join(res).lstrip()


def make_analysator(f):
    """Return a static text analyser function that returns float values."""
    def text_analyse(text):
        try:
            rv = f(text)
        except Exception:
            return 0.0
        if not rv:
            return 0.0
        try:
            return min(1.0, max(0.0, float(rv)))
        except (ValueError, TypeError):
            return 0.0
    text_analyse.__doc__ = f.__doc__
    return staticmethod(text_analyse)


def shebang_matches(text, regex):
    r"""Check if the given regular expression matches the last part of the
    shebang if one exists.

        >>> from pygments.util import shebang_matches
        >>> shebang_matches('#!/usr/bin/env python', r'python(2\.\d)?')
        True
        >>> shebang_matches('#!/usr/bin/python2.4', r'python(2\.\d)?')
        True
        >>> shebang_matches('#!/usr/bin/python-ruby', r'python(2\.\d)?')
        False
        >>> shebang_matches('#!/usr/bin/python/ruby', r'python(2\.\d)?')
        False
        >>> shebang_matches('#!/usr/bin/startsomethingwith python',
        ...                 r'python(2\.\d)?')
        True

    It also checks for common windows executable file extensions::

        >>> shebang_matches('#!C:\\Python2.4\\Python.exe', r'python(2\.\d)?')
        True

    Parameters (``'-f'`` or ``'--foo'`` are ignored so ``'perl'`` does
    the same as ``'perl -e'``)

    Note that this method automatically searches the whole string (eg:
    the regular expression is wrapped in ``'^$'``)
    """
    index = text.find('\n')
    if index >= 0:
        first_line = text[:index].lower()
    else:
        first_line = text.lower()
    if first_line.startswith('#!'):
        try:
            found = [x for x in split_path_re.split(first_line[2:].strip())
                     if x and not x.startswith('-')][-1]
        except IndexError:
            return False
        regex = re.compile(r'^%s(\.(exe|cmd|bat|bin))?$' % regex, re.IGNORECASE)
        if regex.search(found) is not None:
            return True
    return False


def doctype_matches(text, regex):
    """Check if the doctype matches a regular expression (if present).

    Note that this method only checks the first part of a DOCTYPE.
    eg: 'html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"'
    """
    m = doctype_lookup_re.match(text)
    if m is None:
        return False
    doctype = m.group(2)
    return re.compile(regex, re.I).match(doctype.strip()) is not None


def html_doctype_matches(text):
    """Check if the file looks like it has a html doctype."""
    return doctype_matches(text, r'html')


_looks_like_xml_cache = {}


def looks_like_xml(text):
    """Check if a doctype exists or if we have some tags."""
    if xml_decl_re.match(text):
        return True
    key = hash(text)
    try:
        return _looks_like_xml_cache[key]
    except KeyError:
        m = doctype_lookup_re.match(text)
        if m is not None:
            return True
        rv = tag_re.search(text[:1000]) is not None
        _looks_like_xml_cache[key] = rv
        return rv


# Python narrow build compatibility

def _surrogatepair(c):
    # Given a unicode character code
    # with length greater than 16 bits,
    # return the two 16 bit surrogate pair.
    # From example D28 of:
    # http://www.unicode.org/book/ch03.pdf
    return (0xd7c0 + (c >> 10), (0xdc00 + (c & 0x3ff)))


def unirange(a, b):
    """Returns a regular expression string to match the given non-BMP range."""
    if b < a:
        raise ValueError("Bad character range")
    if a < 0x10000 or b < 0x10000:
        raise ValueError("unirange is only defined for non-BMP ranges")

    if sys.maxunicode > 0xffff:
        # wide build
        return u'[%s-%s]' % (unichr(a), unichr(b))
    else:
        # narrow build stores surrogates, and the 're' module handles them
        # (incorrectly) as characters.  Since there is still ordering among
        # these characters, expand the range to one that it understands.  Some
        # background in http://bugs.python.org/issue3665 and
        # http://bugs.python.org/issue12749
        #
        # Additionally, the lower constants are using unichr rather than
        # literals because jython [which uses the wide path] can't load this
        # file if they are literals.
        ah, al = _surrogatepair(a)
        bh, bl = _surrogatepair(b)
        if ah == bh:
            return u'(?:%s[%s-%s])' % (unichr(ah), unichr(al), unichr(bl))
        else:
            buf = []
            buf.append(u'%s[%s-%s]' %
                       (unichr(ah), unichr(al),
                        ah == bh and unichr(bl) or unichr(0xdfff)))
            if ah - bh > 1:
                buf.append(u'[%s-%s][%s-%s]' %
                           unichr(ah+1), unichr(bh-1), unichr(0xdc00), unichr(0xdfff))
            if ah != bh:
                buf.append(u'%s[%s-%s]' %
                           (unichr(bh), unichr(0xdc00), unichr(bl)))

            return u'(?:' + u'|'.join(buf) + u')'


def format_lines(var_name, seq, raw=False, indent_level=0):
    """Formats a sequence of strings for output."""
    lines = []
    base_indent = ' ' * indent_level * 4
    inner_indent = ' ' * (indent_level + 1) * 4
    lines.append(base_indent + var_name + ' = (')
    if raw:
        # These should be preformatted reprs of, say, tuples.
        for i in seq:
            lines.append(inner_indent + i + ',')
    else:
        for i in seq:
            # Force use of single quotes
            r = repr(i + '"')
            lines.append(inner_indent + r[:-2] + r[-1] + ',')
    lines.append(base_indent + ')')
    return '\n'.join(lines)


def duplicates_removed(it, already_seen=()):
    """
    Returns a list with duplicates removed from the iterable `it`.

    Order is preserved.
    """
    lst = []
    seen = set()
    for i in it:
        if i in seen or i in already_seen:
            continue
        lst.append(i)
        seen.add(i)
    return lst


class Future(object):
    """Generic class to defer some work.

    Handled specially in RegexLexerMeta, to support regex string construction at
    first use.
    """
    def get(self):
        raise NotImplementedError


def guess_decode(text):
    """Decode *text* with guessed encoding.

    First try UTF-8; this should fail for non-UTF-8 encodings.
    Then try the preferred locale encoding.
    Fall back to latin-1, which always works.
    """
    try:
        text = text.decode('utf-8')
        return text, 'utf-8'
    except UnicodeDecodeError:
        try:
            import locale
            prefencoding = locale.getpreferredencoding()
            text = text.decode()
            return text, prefencoding
        except (UnicodeDecodeError, LookupError):
            text = text.decode('latin1')
            return text, 'latin1'


def guess_decode_from_terminal(text, term):
    """Decode *text* coming from terminal *term*.

    First try the terminal encoding, if given.
    Then try UTF-8.  Then try the preferred locale encoding.
    Fall back to latin-1, which always works.
    """
    if getattr(term, 'encoding', None):
        try:
            text = text.decode(term.encoding)
        except UnicodeDecodeError:
            pass
        else:
            return text, term.encoding
    return guess_decode(text)


def terminal_encoding(term):
    """Return our best guess of encoding for the given *term*."""
    if getattr(term, 'encoding', None):
        return term.encoding
    import locale
    return locale.getpreferredencoding()


# Python 2/3 compatibility

if sys.version_info < (3, 0):
    unichr = unichr
    xrange = xrange
    string_types = (str, unicode)
    text_type = unicode
    u_prefix = 'u'
    iteritems = dict.iteritems
    itervalues = dict.itervalues
    import StringIO
    import cStringIO
    # unfortunately, io.StringIO in Python 2 doesn't accept str at all
    StringIO = StringIO.StringIO
    BytesIO = cStringIO.StringIO
else:
    unichr = chr
    xrange = range
    string_types = (str,)
    text_type = str
    u_prefix = ''
    iteritems = dict.items
    itervalues = dict.values
    from io import StringIO, BytesIO, TextIOWrapper

    class UnclosingTextIOWrapper(TextIOWrapper):
        # Don't close underlying buffer on destruction.
        def close(self):
            self.flush()


def add_metaclass(metaclass):
    """Class decorator for creating a class with a metaclass."""
    def wrapper(cls):
        orig_vars = cls.__dict__.copy()
        orig_vars.pop('__dict__', None)
        orig_vars.pop('__weakref__', None)
        for slots_var in orig_vars.get('__slots__', ()):
            orig_vars.pop(slots_var)
        return metaclass(cls.__name__, cls.__bases__, orig_vars)
    return wrapper