def load_lexer_from_file(filename, lexername="CustomLexer", **options):
"""Load a lexer from a file.
This method expects a file located relative to the current working
directory, which contains a Lexer class. By default, it expects the
Lexer to be name CustomLexer; you can specify your own class name
as the second argument to this function.
Users should be very careful with the input, because this method
is equivalent to running eval on the input file.
Raises ClassNotFound if there are any problems importing the Lexer.
.. versionadded:: 2.2
"""
try:
# This empty dict will contain the namespace for the exec'd file
custom_namespace = {}
with open(filename, 'rb') as f:
exec(f.read(), custom_namespace)
# Retrieve the class `lexername` from that namespace
if lexername not in custom_namespace:
raise ClassNotFound('no valid %s class found in %s' %
(lexername, filename))
lexer_class = custom_namespace[lexername]
# And finally instantiate it with the options
return lexer_class(**options)
except IOError as err:
raise ClassNotFound('cannot read %s' % filename)
except ClassNotFound as err:
raise
except Exception as err:
raise ClassNotFound('error when loading custom lexer: %s' % err)
def guess_lexer(_text, **options):
"""Guess a lexer by strong distinctions in the text (eg, shebang)."""
if not isinstance(_text, text_type):
inencoding = options.get('inencoding', options.get('encoding'))
if inencoding:
_text = _text.decode(inencoding or 'utf8')
else:
_text, _ = guess_decode(_text)
# try to get a vim modeline first
ft = get_filetype_from_buffer(_text)
if ft is not None:
try:
return get_lexer_by_name(ft, **options)
except ClassNotFound:
pass
best_lexer = [0.0, None]
for lexer in _iter_lexerclasses():
rv = lexer.analyse_text(_text)
if rv == 1.0:
return lexer(**options)
if rv > best_lexer[0]:
best_lexer[:] = (rv, lexer)
if not best_lexer[0] or best_lexer[1] is None:
raise ClassNotFound('no lexer matching the text found')
return best_lexer[1](**options)
def get_formatter_by_name(_alias, **options):
"""Lookup and instantiate a formatter by alias.
Raises ClassNotFound if not found.
"""
cls = find_formatter_class(_alias)
if cls is None:
raise ClassNotFound("no formatter found for name %r" % _alias)
return cls(**options)
def get_lexer_by_name(_alias, **options):
"""Get a lexer by an alias.
Raises ClassNotFound if not found.
"""
if not _alias:
raise ClassNotFound('no lexer for alias %r found' % _alias)
# lookup builtin lexers
for module_name, name, aliases, _, _ in itervalues(LEXERS):
if _alias.lower() in aliases:
if name not in _lexer_cache:
_load_lexers(module_name)
return _lexer_cache[name](**options)
# continue with lexers from setuptools entrypoints
for cls in find_plugin_lexers():
if _alias.lower() in cls.aliases:
return cls(**options)
raise ClassNotFound('no lexer for alias %r found' % _alias)
def find_lexer_class_by_name(_alias):
"""Lookup a lexer class by alias.
Like `get_lexer_by_name`, but does not instantiate the class.
.. versionadded:: 2.2
"""
if not _alias:
raise ClassNotFound('no lexer for alias %r found' % _alias)
# lookup builtin lexers
for module_name, name, aliases, _, _ in itervalues(LEXERS):
if _alias.lower() in aliases:
if name not in _lexer_cache:
_load_lexers(module_name)
return _lexer_cache[name]
# continue with lexers from setuptools entrypoints
for cls in find_plugin_lexers():
if _alias.lower() in cls.aliases:
return cls
raise ClassNotFound('no lexer for alias %r found' % _alias)
def get_filter_by_name(filtername, **options):
"""Return an instantiated filter.
Options are passed to the filter initializer if wanted.
Raise a ClassNotFound if not found.
"""
cls = find_filter_class(filtername)
if cls:
return cls(**options)
else:
raise ClassNotFound('filter %r not found' % filtername)
def get_lexer_for_filename(_fn, code=None, **options):
"""Get a lexer for a filename.
If multiple lexers match the filename pattern, use ``analyse_text()`` to
figure out which one is more appropriate.
Raises ClassNotFound if not found.
"""
res = find_lexer_class_for_filename(_fn, code)
if not res:
raise ClassNotFound('no lexer for filename %r found' % _fn)
return res(**options)
def guess_lexer_for_filename(_fn, _text, **options):
"""
Lookup all lexers that handle those filenames primary (``filenames``)
or secondary (``alias_filenames``). Then run a text analysis for those
lexers and choose the best result.
usage::
>>> from typecode._vendor.pygments.lexers import guess_lexer_for_filename
>>> guess_lexer_for_filename('hello.html', '<%= @foo %>').__class__.__name__
'RhtmlLexer'
>>> guess_lexer_for_filename('hello.html', '<h1>{{ title|e }}</h1>').__class__.__name__
'HtmlDjangoLexer'
>>> guess_lexer_for_filename('style.css', 'a { color: <?= $link ?> }').__class__.__name__
'CssPhpLexer'
"""
fn = basename(_fn)
primary = {}
matching_lexers = set()
for lexer in _iter_lexerclasses():
for filename in lexer.filenames:
if _fn_matches(fn, filename):
matching_lexers.add(lexer)
primary[lexer] = True
for filename in lexer.alias_filenames:
if _fn_matches(fn, filename):
matching_lexers.add(lexer)
primary[lexer] = False
if not matching_lexers:
raise ClassNotFound('no lexer for filename %r found' % fn)
if len(matching_lexers) == 1:
return matching_lexers.pop()(**options)
result = []
for lexer in matching_lexers:
rv = lexer.analyse_text(_text)
if rv == 1.0:
return lexer(**options)
result.append((rv, lexer))
def type_sort(t):
# sort by:
# - analyse score
# - is primary filename pattern?
# - priority
# - last resort: class name
return (t[0], primary[t[1]], t[1].priority, t[1].__name__)
result.sort(key=type_sort)
return result[-1][1](**options)
def get_lexer_for_mimetype(_mime, **options):
"""Get a lexer for a mimetype.
Raises ClassNotFound if not found.
"""
for modname, name, _, _, mimetypes in itervalues(LEXERS):
if _mime in mimetypes:
if name not in _lexer_cache:
_load_lexers(modname)
return _lexer_cache[name](**options)
for cls in find_plugin_lexers():
if _mime in cls.mimetypes:
return cls(**options)
raise ClassNotFound('no lexer for mimetype %r found' % _mime)
def get_style_by_name(name):
if name in STYLE_MAP:
mod, cls = STYLE_MAP[name].split('::')
builtin = "yes"
else:
for found_name, style in find_plugin_styles():
if name == found_name:
return style
# perhaps it got dropped into our styles package
builtin = ""
mod = name
cls = name.title() + "Style"
try:
mod = __import__('pygments.styles.' + mod, None, None, [cls])
except ImportError:
raise ClassNotFound("Could not find style module %r" % mod +
(builtin and ", though it should be builtin") +
".")
try:
return getattr(mod, cls)
except AttributeError:
raise ClassNotFound("Could not find style class %r in style module." %
cls)
def get_formatter_for_filename(fn, **options):
"""Lookup and instantiate a formatter by filename pattern.
Raises ClassNotFound if not found.
"""
fn = basename(fn)
for modname, name, _, filenames, _ in itervalues(FORMATTERS):
for filename in filenames:
if _fn_matches(fn, filename):
if name not in _formatter_cache:
_load_formatters(modname)
return _formatter_cache[name](**options)
for cls in find_plugin_formatters():
for filename in cls.filenames:
if _fn_matches(fn, filename):
return cls(**options)
raise ClassNotFound("no formatter found for file name %r" % fn)
