view MoinMoin/support/pygments/ @ 5564:74fa6f714526

upgraded pygments to 1.2.2
author Thomas Waldmann <tw AT waldmann-edv DOT de>
date Sun, 21 Feb 2010 23:52:54 +0100
parents 0970ed47d2cd
children f76914e77229
line wrap: on
line source
# -*- coding: utf-8 -*-

    Utility functions.

    :copyright: Copyright 2006-2010 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'''(?smx)
tag_re = re.compile(r'<(.+?)(\s.*?)?>.*?</.+?>(?uism)')

class ClassNotFound(ValueError):
    If one of the get_*_by_* functions didn't find a matching class.

class OptionError(Exception):

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, basestring):
        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
        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)
        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, basestring):
        return val.split()
    elif isinstance(val, (list, tuple)):
        return list(val)
        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())
    return ''.join(res).lstrip()

def make_analysator(f):
    Return a static text analysation function that
    returns float values.
    def text_analyse(text):
        rv = f(text)
        if not rv:
            return 0.0
        return min(1.0, max(0.0, float(rv)))
    text_analyse.__doc__ = f.__doc__
    return staticmethod(text_analyse)

def shebang_matches(text, regex):
    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)?')
        >>> shebang_matches('#!/usr/bin/python2.4', r'python(2\.\d)?')
        >>> shebang_matches('#!/usr/bin/python-ruby', r'python(2\.\d)?')
        >>> shebang_matches('#!/usr/bin/python/ruby', r'python(2\.\d)?')
        >>> shebang_matches('#!/usr/bin/startsomethingwith python',
        ...                 r'python(2\.\d)?')

    It also checks for common windows executable file extensions::

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

    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 ``'^$'``)
    index = text.find('\n')
    if index >= 0:
        first_line = text[:index].lower()
        first_line = text.lower()
    if first_line.startswith('#!'):
            found = [x for x in split_path_re.split(first_line[2:].strip())
                     if x and not x.startswith('-')][-1]
        except IndexError:
            return False
        regex = re.compile('^%s(\.(exe|cmd|bat|bin))?$' % regex, re.IGNORECASE)
        if 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 =
    return re.compile(regex).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\s+PUBLIC\s+"-//W3C//DTD X?HTML.*')

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

# Python 2/3 compatibility

if sys.version_info < (3,0):
    b = bytes = str
    u_prefix = 'u'
    import StringIO, cStringIO
    BytesIO = cStringIO.StringIO
    StringIO = StringIO.StringIO
    import builtins
    bytes = builtins.bytes
    u_prefix = ''
    def b(s):
        if isinstance(s, str):
            return bytes(map(ord, s))
        elif isinstance(s, bytes):
            return s
            raise TypeError("Invalid argument %r for b()" % (s,))
    import io
    BytesIO = io.BytesIO
    StringIO = io.StringIO