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 guess_lexer(_text, conf_threshold=0.01, mime_mul=0.09, **options):
"""Guess a lexer by strong distinctions in the text (eg, shebang)."""
# 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)
#MIME has an unusually high rv when providing it with any kind of text
if 'MIME' in str(lexer):
rv *= mime_mul
if rv == 1.0:
return lexer(**options)
if rv > best_lexer[0]:
best_lexer[:] = (rv, lexer)
if not best_lexer[
0] or best_lexer[0] < conf_threshold or best_lexer[1] is None:
raise ClassNotFound('no lexer matching the text found')
return best_lexer[1](**options)
def guess_lexer_using_modeline(text):
"""Guess lexer for given text using Vim modeline.
Returns a tuple of (lexer, accuracy).
"""
lexer, accuracy = None, None
file_type = None
try:
file_type = get_filetype_from_buffer(text)
except:
pass
if file_type is not None:
try:
lexer = get_lexer_by_name(file_type)
except ClassNotFound:
pass
if lexer is not None:
try:
accuracy = lexer.analyse_text(text)
except:
pass
return lexer, accuracy
def test_modelines():
for buf in [
'vi: ft=python' + '\n' * 8, 'vi: ft=python' + '\n' * 8,
'\n\n\n\nvi=8: syntax=python' + '\n' * 8,
'\n' * 8 + 'ex: filetype=python',
'\n' * 8 + 'vim: some,other,syn=python\n\n\n\n'
]:
assert modeline.get_filetype_from_buffer(buf) == 'python'
def guess_lexer(_text, **options):
"""Guess a lexer by strong distinctions in the text (eg, shebang)."""
# 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 guess_lexer(_text, **options):
"""Guess a lexer by strong distinctions in the text (eg, shebang)."""
# 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 verify(buf):
assert modeline.get_filetype_from_buffer(buf) == 'python'
def verify(buf):
assert modeline.get_filetype_from_buffer(buf) == 'python'
