def extract_code_strings(filepath):
import pygments
from pygments import lexers
from pygments.token import Token
comments = []
code_words = set()
lex = lexers.find_lexer_class_for_filename(filepath)
if lex is None:
return comments, code_words
slineno = 1
with open(filepath, encoding='utf-8') as fh:
source = fh.read()
for ty, ttext in lex.get_tokens(source):
if ty in {
Token.Literal.String, Token.Literal.String.Double,
Token.Literal.String.Single
}:
comments.append(Comment(filepath, ttext, slineno, 'STRING'))
else:
for match in re_vars.finditer(ttext):
code_words.add(match.group(0))
# Ugh - not nice or fast.
slineno += ttext.count("\n")
return comments, code_words
def configure_syntax_highlighting(self, extension=None):
if self.highlighter:
self.language.emit(self.DEFAULT_LANGUAGE)
self.highlighter = None
# Removing the highlighter does not refresh current text
# Therefore, it is a necessity to call the set_text method
self.modes.remove(modes.PygmentsSyntaxHighlighter)
self.set_text(self.get_text())
if extension:
lexer = find_lexer_class_for_filename(extension)
if not lexer:
return
self.language.emit(lexer().name)
self.highlighter = modes.PygmentsSyntaxHighlighter(self.document(), lexer=lexer())
self.highlighter.pygments_style = self.THEME
self.modes.append(self.highlighter)
def pygments_lexer(self):
language = self.data.get('pygments_language')
lexer = None
if language is not None:
try:
lexer = lexers.get_lexer_by_name(language)
except ClassNotFound:
lexer = None
if lexer is None and self.gist_file_name is not None:
try:
lexer = lexers.find_lexer_class_for_filename(
self.gist_file_name, )
except ClassNotFound:
lexer = None
else:
lexer = lexer()
if lexer is None:
lexer = lexers.TextLexer()
return lexer
def main(fn, lexer=None, options={}):
if lexer is not None:
lxcls = get_lexer_by_name(lexer).__class__
else:
lxcls = find_lexer_class_for_filename(os.path.basename(fn))
if lxcls is None:
name, rest = fn.split('_', 1)
lxcls = find_lexer_class(name)
if lxcls is None:
raise AssertionError('no lexer found for file %r' % fn)
print('Using lexer: %s (%s.%s)' % (lxcls.name, lxcls.__module__,
lxcls.__name__))
debug_lexer = False
# if profile:
# # does not work for e.g. ExtendedRegexLexers
# if lxcls.__bases__ == (RegexLexer,):
# # yes we can! (change the metaclass)
# lxcls.__class__ = ProfilingRegexLexerMeta
# lxcls.__bases__ = (ProfilingRegexLexer,)
# lxcls._prof_sort_index = profsort
# else:
# if lxcls.__bases__ == (RegexLexer,):
# lxcls.__bases__ = (DebuggingRegexLexer,)
# debug_lexer = True
# elif lxcls.__bases__ == (DebuggingRegexLexer,):
# # already debugged before
# debug_lexer = True
# else:
# # HACK: ExtendedRegexLexer subclasses will only partially work here.
# lxcls.__bases__ = (DebuggingRegexLexer,)
# debug_lexer = True
lx = lxcls(**options)
lno = 1
if fn == '-':
text = sys.stdin.read()
else:
with open(fn, 'rb') as fp:
text = fp.read().decode('utf-8')
text = text.strip('\n') + '\n'
tokens = []
states = []
def show_token(tok, state):
reprs = list(map(repr, tok))
print(' ' + reprs[1] + ' ' + ' ' * (29-len(reprs[1])) + reprs[0], end=' ')
if debug_lexer:
print(' ' + ' ' * (29-len(reprs[0])) + ' : '.join(state) if state else '', end=' ')
print()
for type, val in lx.get_tokens(text):
lno += val.count('\n')
if type == Error and not ignerror:
print('Error parsing', fn, 'on line', lno)
if not showall:
print('Previous tokens' + (debug_lexer and ' and states' or '') + ':')
for i in range(max(len(tokens) - num, 0), len(tokens)):
if debug_lexer:
show_token(tokens[i], states[i])
else:
show_token(tokens[i], None)
print('Error token:')
l = len(repr(val))
print(' ' + repr(val), end=' ')
if debug_lexer and hasattr(lx, 'ctx'):
print(' ' * (60-l) + ' : '.join(lx.ctx.stack), end=' ')
print()
print()
return 1
tokens.append((type, val))
if debug_lexer:
if hasattr(lx, 'ctx'):
states.append(lx.ctx.stack[:])
else:
states.append(None)
if showall:
show_token((type, val), states[-1] if debug_lexer else None)
return 0
def main_inner(popts, args, usage):
opts = {}
O_opts = []
P_opts = []
F_opts = []
for opt, arg in popts:
if opt == '-O':
O_opts.append(arg)
elif opt == '-P':
P_opts.append(arg)
elif opt == '-F':
F_opts.append(arg)
opts[opt] = arg
if opts.pop('-h', None) is not None:
print(usage)
return 0
if opts.pop('-V', None) is not None:
print('Pygments version %s, (c) 2006-2017 by Georg Brandl.' % __version__)
return 0
# handle ``pygmentize -L``
L_opt = opts.pop('-L', None)
if L_opt is not None:
if opts:
print(usage, file=sys.stderr)
return 2
# print version
main(['', '-V'])
if not args:
args = ['lexer', 'formatter', 'filter', 'style']
for arg in args:
_print_list(arg.rstrip('s'))
return 0
# handle ``pygmentize -H``
H_opt = opts.pop('-H', None)
if H_opt is not None:
if opts or len(args) != 2:
print(usage, file=sys.stderr)
return 2
what, name = args # pylint: disable=unbalanced-tuple-unpacking
if what not in ('lexer', 'formatter', 'filter'):
print(usage, file=sys.stderr)
return 2
return _print_help(what, name)
# parse -O options
parsed_opts = _parse_options(O_opts)
opts.pop('-O', None)
# parse -P options
for p_opt in P_opts:
try:
name, value = p_opt.split('=', 1)
except ValueError:
parsed_opts[p_opt] = True
else:
parsed_opts[name] = value
opts.pop('-P', None)
# encodings
inencoding = parsed_opts.get('inencoding', parsed_opts.get('encoding'))
outencoding = parsed_opts.get('outencoding', parsed_opts.get('encoding'))
# handle ``pygmentize -N``
infn = opts.pop('-N', None)
if infn is not None:
lexer = find_lexer_class_for_filename(infn)
if lexer is None:
lexer = TextLexer
print(lexer.aliases[0])
return 0
# handle ``pygmentize -S``
S_opt = opts.pop('-S', None)
a_opt = opts.pop('-a', None)
if S_opt is not None:
f_opt = opts.pop('-f', None)
if not f_opt:
print(usage, file=sys.stderr)
return 2
if opts or args:
print(usage, file=sys.stderr)
return 2
try:
parsed_opts['style'] = S_opt
fmter = get_formatter_by_name(f_opt, **parsed_opts)
except ClassNotFound as err:
print(err, file=sys.stderr)
return 1
print(fmter.get_style_defs(a_opt or ''))
return 0
# if no -S is given, -a is not allowed
if a_opt is not None:
print(usage, file=sys.stderr)
return 2
# parse -F options
F_opts = _parse_filters(F_opts)
opts.pop('-F', None)
allow_custom_lexer_formatter = False
# -x: allow custom (eXternal) lexers and formatters
if opts.pop('-x', None) is not None:
allow_custom_lexer_formatter = True
# select lexer
lexer = None
# given by name?
lexername = opts.pop('-l', None)
if lexername:
# custom lexer, located relative to user's cwd
if allow_custom_lexer_formatter and '.py' in lexername:
try:
if ':' in lexername:
filename, name = lexername.rsplit(':', 1)
lexer = load_lexer_from_file(filename, name,
**parsed_opts)
else:
lexer = load_lexer_from_file(lexername, **parsed_opts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
else:
try:
lexer = get_lexer_by_name(lexername, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
# read input code
code = None
if args:
if len(args) > 1:
print(usage, file=sys.stderr)
return 2
if '-s' in opts:
print('Error: -s option not usable when input file specified',
file=sys.stderr)
return 2
infn = args[0]
try:
with open(infn, 'rb') as infp:
code = infp.read()
except Exception as err:
print('Error: cannot read infile:', err, file=sys.stderr)
return 1
if not inencoding:
code, inencoding = guess_decode(code)
# do we have to guess the lexer?
if not lexer:
try:
lexer = get_lexer_for_filename(infn, code, **parsed_opts)
except ClassNotFound as err:
if '-g' in opts:
try:
lexer = guess_lexer(code, **parsed_opts)
except ClassNotFound:
lexer = TextLexer(**parsed_opts)
else:
print('Error:', err, file=sys.stderr)
return 1
except OptionError as err:
print('Error:', err, file=sys.stderr)
return 1
elif '-s' not in opts: # treat stdin as full file (-s support is later)
# read code from terminal, always in binary mode since we want to
# decode ourselves and be tolerant with it
if sys.version_info > (3,):
# Python 3: we have to use .buffer to get a binary stream
code = sys.stdin.buffer.read()
else:
code = sys.stdin.read()
if not inencoding:
code, inencoding = guess_decode_from_terminal(code, sys.stdin)
# else the lexer will do the decoding
if not lexer:
try:
lexer = guess_lexer(code, **parsed_opts)
except ClassNotFound:
lexer = TextLexer(**parsed_opts)
else: # -s option needs a lexer with -l
if not lexer:
print('Error: when using -s a lexer has to be selected with -l',
file=sys.stderr)
return 2
# process filters
for fname, fopts in F_opts:
try:
lexer.add_filter(fname, **fopts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
# select formatter
outfn = opts.pop('-o', None)
fmter = opts.pop('-f', None)
if fmter:
# custom formatter, located relative to user's cwd
if allow_custom_lexer_formatter and '.py' in fmter:
try:
if ':' in fmter:
file, fmtername = fmter.rsplit(':', 1)
fmter = load_formatter_from_file(file, fmtername,
**parsed_opts)
else:
fmter = load_formatter_from_file(fmter, **parsed_opts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
else:
try:
fmter = get_formatter_by_name(fmter, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
if outfn:
if not fmter:
try:
fmter = get_formatter_for_filename(outfn, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
try:
outfile = open(outfn, 'wb')
except Exception as err:
print('Error: cannot open outfile:', err, file=sys.stderr)
return 1
else:
if not fmter:
fmter = TerminalFormatter(**parsed_opts)
if sys.version_info > (3,):
# Python 3: we have to use .buffer to get a binary stream
outfile = sys.stdout.buffer
else:
outfile = sys.stdout
# determine output encoding if not explicitly selected
if not outencoding:
if outfn:
# output file? use lexer encoding for now (can still be None)
fmter.encoding = inencoding
else:
# else use terminal encoding
fmter.encoding = terminal_encoding(sys.stdout)
# provide coloring under Windows, if possible
if not outfn and sys.platform in ('win32', 'cygwin') and \
fmter.name in ('Terminal', 'Terminal256'): # pragma: no cover
# unfortunately colorama doesn't support binary streams on Py3
if sys.version_info > (3,):
from pygments.util import UnclosingTextIOWrapper
outfile = UnclosingTextIOWrapper(outfile, encoding=fmter.encoding)
fmter.encoding = None
try:
import colorama.initialise
except ImportError:
pass
else:
outfile = colorama.initialise.wrap_stream(
outfile, convert=None, strip=None, autoreset=False, wrap=True)
# When using the LaTeX formatter and the option `escapeinside` is
# specified, we need a special lexer which collects escaped text
# before running the chosen language lexer.
escapeinside = parsed_opts.get('escapeinside', '')
if len(escapeinside) == 2 and isinstance(fmter, LatexFormatter):
left = escapeinside[0]
right = escapeinside[1]
lexer = LatexEmbeddedLexer(left, right, lexer)
# ... and do it!
if '-s' not in opts:
# process whole input as per normal...
highlight(code, lexer, fmter, outfile)
return 0
else:
# line by line processing of stdin (eg: for 'tail -f')...
try:
while 1:
if sys.version_info > (3,):
# Python 3: we have to use .buffer to get a binary stream
line = sys.stdin.buffer.readline()
else:
line = sys.stdin.readline()
if not line:
break
if not inencoding:
line = guess_decode_from_terminal(line, sys.stdin)[0]
highlight(line, lexer, fmter, outfile)
if hasattr(outfile, 'flush'):
outfile.flush()
return 0
except KeyboardInterrupt: # pragma: no cover
return 0
def main(fn, lexer=None, options={}):
if lexer is not None:
lxcls = get_lexer_by_name(lexer).__class__
else:
lxcls = find_lexer_class_for_filename(os.path.basename(fn))
if lxcls is None:
name, rest = fn.split('_', 1)
lxcls = find_lexer_class(name)
if lxcls is None:
raise AssertionError('no lexer found for file %r' % fn)
print('Using lexer: %s (%s.%s)' %
(lxcls.name, lxcls.__module__, lxcls.__name__))
debug_lexer = False
# if profile:
# # does not work for e.g. ExtendedRegexLexers
# if lxcls.__bases__ == (RegexLexer,):
# # yes we can! (change the metaclass)
# lxcls.__class__ = ProfilingRegexLexerMeta
# lxcls.__bases__ = (ProfilingRegexLexer,)
# lxcls._prof_sort_index = profsort
# else:
# if lxcls.__bases__ == (RegexLexer,):
# lxcls.__bases__ = (DebuggingRegexLexer,)
# debug_lexer = True
# elif lxcls.__bases__ == (DebuggingRegexLexer,):
# # already debugged before
# debug_lexer = True
# else:
# # HACK: ExtendedRegexLexer subclasses will only partially work here.
# lxcls.__bases__ = (DebuggingRegexLexer,)
# debug_lexer = True
lx = lxcls(**options)
lno = 1
if fn == '-':
text = sys.stdin.read()
else:
with open(fn, 'rb') as fp:
text = fp.read().decode('utf-8')
text = text.strip('\n') + '\n'
tokens = []
states = []
def show_token(tok, state):
reprs = list(map(repr, tok))
print(' ' + reprs[1] + ' ' + ' ' * (29 - len(reprs[1])) + reprs[0],
end=' ')
if debug_lexer:
print(' ' + ' ' * (29 - len(reprs[0])) +
' : '.join(state) if state else '',
end=' ')
print()
for type, val in lx.get_tokens(text):
lno += val.count('\n')
if type == Error and not ignerror:
print('Error parsing', fn, 'on line', lno)
if not showall:
print('Previous tokens' +
(debug_lexer and ' and states' or '') + ':')
for i in range(max(len(tokens) - num, 0), len(tokens)):
if debug_lexer:
show_token(tokens[i], states[i])
else:
show_token(tokens[i], None)
print('Error token:')
l = len(repr(val))
print(' ' + repr(val), end=' ')
if debug_lexer and hasattr(lx, 'ctx'):
print(' ' * (60 - l) + ' : '.join(lx.ctx.stack), end=' ')
print()
print()
return 1
tokens.append((type, val))
if debug_lexer:
if hasattr(lx, 'ctx'):
states.append(lx.ctx.stack[:])
else:
states.append(None)
if showall:
show_token((type, val), states[-1] if debug_lexer else None)
return 0
def main_inner(parser, argns):
if argns.help:
parser.print_help()
return 0
if argns.V:
print('Pygments version %s, (c) 2006-2021 by Georg Brandl, Matthäus '
'Chajdas and contributors.' % __version__)
return 0
def is_only_option(opt):
return not any(v for (k, v) in vars(argns).items() if k != opt)
# handle ``pygmentize -L``
if argns.L is not None:
if not is_only_option('L'):
parser.print_help(sys.stderr)
return 2
# print version
main(['', '-V'])
allowed_types = {'lexer', 'formatter', 'filter', 'style'}
largs = [arg.rstrip('s') for arg in argns.L]
if any(arg not in allowed_types for arg in largs):
parser.print_help(sys.stderr)
return 0
if not largs:
largs = allowed_types
for arg in largs:
_print_list(arg)
return 0
# handle ``pygmentize -H``
if argns.H:
if not is_only_option('H'):
parser.print_help(sys.stderr)
return 2
what, name = argns.H
if what not in ('lexer', 'formatter', 'filter'):
parser.print_help(sys.stderr)
return 2
return _print_help(what, name)
# parse -O options
parsed_opts = _parse_options(argns.O or [])
# parse -P options
for p_opt in argns.P or []:
try:
name, value = p_opt.split('=', 1)
except ValueError:
parsed_opts[p_opt] = True
else:
parsed_opts[name] = value
# encodings
inencoding = parsed_opts.get('inencoding', parsed_opts.get('encoding'))
outencoding = parsed_opts.get('outencoding', parsed_opts.get('encoding'))
# handle ``pygmentize -N``
if argns.N:
lexer = find_lexer_class_for_filename(argns.N)
if lexer is None:
lexer = TextLexer
print(lexer.aliases[0])
return 0
# handle ``pygmentize -C``
if argns.C:
inp = sys.stdin.buffer.read()
try:
lexer = guess_lexer(inp, inencoding=inencoding)
except ClassNotFound:
lexer = TextLexer
print(lexer.aliases[0])
return 0
# handle ``pygmentize -S``
S_opt = argns.S
a_opt = argns.a
if S_opt is not None:
f_opt = argns.f
if not f_opt:
parser.print_help(sys.stderr)
return 2
if argns.l or argns.INPUTFILE:
parser.print_help(sys.stderr)
return 2
try:
parsed_opts['style'] = S_opt
fmter = get_formatter_by_name(f_opt, **parsed_opts)
except ClassNotFound as err:
print(err, file=sys.stderr)
return 1
print(fmter.get_style_defs(a_opt or ''))
return 0
# if no -S is given, -a is not allowed
if argns.a is not None:
parser.print_help(sys.stderr)
return 2
# parse -F options
F_opts = _parse_filters(argns.F or [])
# -x: allow custom (eXternal) lexers and formatters
allow_custom_lexer_formatter = bool(argns.x)
# select lexer
lexer = None
# given by name?
lexername = argns.l
if lexername:
# custom lexer, located relative to user's cwd
if allow_custom_lexer_formatter and '.py' in lexername:
try:
filename = None
name = None
if ':' in lexername:
filename, name = lexername.rsplit(':', 1)
if '.py' in name:
# This can happen on Windows: If the lexername is
# C:\lexer.py -- return to normal load path in that case
name = None
if filename and name:
lexer = load_lexer_from_file(filename, name, **parsed_opts)
else:
lexer = load_lexer_from_file(lexername, **parsed_opts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
else:
try:
lexer = get_lexer_by_name(lexername, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
# read input code
code = None
if argns.INPUTFILE:
if argns.s:
print('Error: -s option not usable when input file specified',
file=sys.stderr)
return 2
infn = argns.INPUTFILE
try:
with open(infn, 'rb') as infp:
code = infp.read()
except Exception as err:
print('Error: cannot read infile:', err, file=sys.stderr)
return 1
if not inencoding:
code, inencoding = guess_decode(code)
# do we have to guess the lexer?
if not lexer:
try:
lexer = get_lexer_for_filename(infn, code, **parsed_opts)
except ClassNotFound as err:
if argns.g:
try:
lexer = guess_lexer(code, **parsed_opts)
except ClassNotFound:
lexer = TextLexer(**parsed_opts)
else:
print('Error:', err, file=sys.stderr)
return 1
except OptionError as err:
print('Error:', err, file=sys.stderr)
return 1
elif not argns.s: # treat stdin as full file (-s support is later)
# read code from terminal, always in binary mode since we want to
# decode ourselves and be tolerant with it
code = sys.stdin.buffer.read() # use .buffer to get a binary stream
if not inencoding:
code, inencoding = guess_decode_from_terminal(code, sys.stdin)
# else the lexer will do the decoding
if not lexer:
try:
lexer = guess_lexer(code, **parsed_opts)
except ClassNotFound:
lexer = TextLexer(**parsed_opts)
else: # -s option needs a lexer with -l
if not lexer:
print('Error: when using -s a lexer has to be selected with -l',
file=sys.stderr)
return 2
# process filters
for fname, fopts in F_opts:
try:
lexer.add_filter(fname, **fopts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
# select formatter
outfn = argns.o
fmter = argns.f
if fmter:
# custom formatter, located relative to user's cwd
if allow_custom_lexer_formatter and '.py' in fmter:
try:
filename = None
name = None
if ':' in fmter:
# Same logic as above for custom lexer
filename, name = fmter.rsplit(':', 1)
if '.py' in name:
name = None
if filename and name:
fmter = load_formatter_from_file(filename, name,
**parsed_opts)
else:
fmter = load_formatter_from_file(fmter, **parsed_opts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
else:
try:
fmter = get_formatter_by_name(fmter, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
if outfn:
if not fmter:
try:
fmter = get_formatter_for_filename(outfn, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
try:
outfile = open(outfn, 'wb')
except Exception as err:
print('Error: cannot open outfile:', err, file=sys.stderr)
return 1
else:
if not fmter:
if '256' in os.environ.get('TERM', ''):
fmter = Terminal256Formatter(**parsed_opts)
else:
fmter = TerminalFormatter(**parsed_opts)
outfile = sys.stdout.buffer
# determine output encoding if not explicitly selected
if not outencoding:
if outfn:
# output file? use lexer encoding for now (can still be None)
fmter.encoding = inencoding
else:
# else use terminal encoding
fmter.encoding = terminal_encoding(sys.stdout)
# provide coloring under Windows, if possible
if not outfn and sys.platform in ('win32', 'cygwin') and \
fmter.name in ('Terminal', 'Terminal256'): # pragma: no cover
# unfortunately colorama doesn't support binary streams on Py3
outfile = UnclosingTextIOWrapper(outfile, encoding=fmter.encoding)
fmter.encoding = None
try:
import colorama.initialise
except ImportError:
pass
else:
outfile = colorama.initialise.wrap_stream(outfile,
convert=None,
strip=None,
autoreset=False,
wrap=True)
# When using the LaTeX formatter and the option `escapeinside` is
# specified, we need a special lexer which collects escaped text
# before running the chosen language lexer.
escapeinside = parsed_opts.get('escapeinside', '')
if len(escapeinside) == 2 and isinstance(fmter, LatexFormatter):
left = escapeinside[0]
right = escapeinside[1]
lexer = LatexEmbeddedLexer(left, right, lexer)
# ... and do it!
if not argns.s:
# process whole input as per normal...
try:
highlight(code, lexer, fmter, outfile)
finally:
if outfn:
outfile.close()
return 0
else:
# line by line processing of stdin (eg: for 'tail -f')...
try:
while 1:
line = sys.stdin.buffer.readline()
if not line:
break
if not inencoding:
line = guess_decode_from_terminal(line, sys.stdin)[0]
highlight(line, lexer, fmter, outfile)
if hasattr(outfile, 'flush'):
outfile.flush()
return 0
except KeyboardInterrupt: # pragma: no cover
return 0
finally:
if outfn:
outfile.close()
def main_inner(popts, args, usage):
opts = {}
O_opts = []
P_opts = []
F_opts = []
for opt, arg in popts:
if opt == '-O':
O_opts.append(arg)
elif opt == '-P':
P_opts.append(arg)
elif opt == '-F':
F_opts.append(arg)
opts[opt] = arg
if opts.pop('-h', None) is not None:
print(usage)
return 0
if opts.pop('-V', None) is not None:
print('Pygments version %s, (c) 2006-2017 by Georg Brandl.' %
__version__)
return 0
# handle ``pygmentize -L``
L_opt = opts.pop('-L', None)
if L_opt is not None:
if opts:
print(usage, file=sys.stderr)
return 2
# print version
main(['', '-V'])
if not args:
args = ['lexer', 'formatter', 'filter', 'style']
for arg in args:
_print_list(arg.rstrip('s'))
return 0
# handle ``pygmentize -H``
H_opt = opts.pop('-H', None)
if H_opt is not None:
if opts or len(args) != 2:
print(usage, file=sys.stderr)
return 2
what, name = args # pylint: disable=unbalanced-tuple-unpacking
if what not in ('lexer', 'formatter', 'filter'):
print(usage, file=sys.stderr)
return 2
return _print_help(what, name)
# parse -O options
parsed_opts = _parse_options(O_opts)
opts.pop('-O', None)
# parse -P options
for p_opt in P_opts:
try:
name, value = p_opt.split('=', 1)
except ValueError:
parsed_opts[p_opt] = True
else:
parsed_opts[name] = value
opts.pop('-P', None)
# encodings
inencoding = parsed_opts.get('inencoding', parsed_opts.get('encoding'))
outencoding = parsed_opts.get('outencoding', parsed_opts.get('encoding'))
# handle ``pygmentize -N``
infn = opts.pop('-N', None)
if infn is not None:
lexer = find_lexer_class_for_filename(infn)
if lexer is None:
lexer = TextLexer
print(lexer.aliases[0])
return 0
# handle ``pygmentize -S``
S_opt = opts.pop('-S', None)
a_opt = opts.pop('-a', None)
if S_opt is not None:
f_opt = opts.pop('-f', None)
if not f_opt:
print(usage, file=sys.stderr)
return 2
if opts or args:
print(usage, file=sys.stderr)
return 2
try:
parsed_opts['style'] = S_opt
fmter = get_formatter_by_name(f_opt, **parsed_opts)
except ClassNotFound as err:
print(err, file=sys.stderr)
return 1
print(fmter.get_style_defs(a_opt or ''))
return 0
# if no -S is given, -a is not allowed
if a_opt is not None:
print(usage, file=sys.stderr)
return 2
# parse -F options
F_opts = _parse_filters(F_opts)
opts.pop('-F', None)
allow_custom_lexer_formatter = False
# -x: allow custom (eXternal) lexers and formatters
if opts.pop('-x', None) is not None:
allow_custom_lexer_formatter = True
# select lexer
lexer = None
# given by name?
lexername = opts.pop('-l', None)
if lexername:
# custom lexer, located relative to user's cwd
if allow_custom_lexer_formatter and '.py' in lexername:
try:
if ':' in lexername:
filename, name = lexername.rsplit(':', 1)
lexer = load_lexer_from_file(filename, name, **parsed_opts)
else:
lexer = load_lexer_from_file(lexername, **parsed_opts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
else:
try:
lexer = get_lexer_by_name(lexername, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
# read input code
code = None
if args:
if len(args) > 1:
print(usage, file=sys.stderr)
return 2
if '-s' in opts:
print('Error: -s option not usable when input file specified',
file=sys.stderr)
return 2
infn = args[0]
try:
with open(infn, 'rb') as infp:
code = infp.read()
except Exception as err:
print('Error: cannot read infile:', err, file=sys.stderr)
return 1
if not inencoding:
code, inencoding = guess_decode(code)
# do we have to guess the lexer?
if not lexer:
try:
lexer = get_lexer_for_filename(infn, code, **parsed_opts)
except ClassNotFound as err:
if '-g' in opts:
try:
lexer = guess_lexer(code, **parsed_opts)
except ClassNotFound:
lexer = TextLexer(**parsed_opts)
else:
print('Error:', err, file=sys.stderr)
return 1
except OptionError as err:
print('Error:', err, file=sys.stderr)
return 1
elif '-s' not in opts: # treat stdin as full file (-s support is later)
# read code from terminal, always in binary mode since we want to
# decode ourselves and be tolerant with it
if sys.version_info > (3, ):
# Python 3: we have to use .buffer to get a binary stream
code = sys.stdin.buffer.read()
else:
code = sys.stdin.read()
if not inencoding:
code, inencoding = guess_decode_from_terminal(code, sys.stdin)
# else the lexer will do the decoding
if not lexer:
try:
lexer = guess_lexer(code, **parsed_opts)
except ClassNotFound:
lexer = TextLexer(**parsed_opts)
else: # -s option needs a lexer with -l
if not lexer:
print('Error: when using -s a lexer has to be selected with -l',
file=sys.stderr)
return 2
# process filters
for fname, fopts in F_opts:
try:
lexer.add_filter(fname, **fopts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
# select formatter
outfn = opts.pop('-o', None)
fmter = opts.pop('-f', None)
if fmter:
# custom formatter, located relative to user's cwd
if allow_custom_lexer_formatter and '.py' in fmter:
try:
if ':' in fmter:
file, fmtername = fmter.rsplit(':', 1)
fmter = load_formatter_from_file(file, fmtername,
**parsed_opts)
else:
fmter = load_formatter_from_file(fmter, **parsed_opts)
except ClassNotFound as err:
print('Error:', err, file=sys.stderr)
return 1
else:
try:
fmter = get_formatter_by_name(fmter, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
if outfn:
if not fmter:
try:
fmter = get_formatter_for_filename(outfn, **parsed_opts)
except (OptionError, ClassNotFound) as err:
print('Error:', err, file=sys.stderr)
return 1
try:
outfile = open(outfn, 'wb')
except Exception as err:
print('Error: cannot open outfile:', err, file=sys.stderr)
return 1
else:
if not fmter:
if '256' in os.environ.get('TERM', ''):
fmter = Terminal256Formatter(**parsed_opts)
else:
fmter = TerminalFormatter(**parsed_opts)
if sys.version_info > (3, ):
# Python 3: we have to use .buffer to get a binary stream
outfile = sys.stdout.buffer
else:
outfile = sys.stdout
# determine output encoding if not explicitly selected
if not outencoding:
if outfn:
# output file? use lexer encoding for now (can still be None)
fmter.encoding = inencoding
else:
# else use terminal encoding
fmter.encoding = terminal_encoding(sys.stdout)
# provide coloring under Windows, if possible
if not outfn and sys.platform in ('win32', 'cygwin') and \
fmter.name in ('Terminal', 'Terminal256'): # pragma: no cover
# unfortunately colorama doesn't support binary streams on Py3
if sys.version_info > (3, ):
from pygments.util import UnclosingTextIOWrapper
outfile = UnclosingTextIOWrapper(outfile, encoding=fmter.encoding)
fmter.encoding = None
try:
import colorama.initialise
except ImportError:
pass
else:
outfile = colorama.initialise.wrap_stream(outfile,
convert=None,
strip=None,
autoreset=False,
wrap=True)
# When using the LaTeX formatter and the option `escapeinside` is
# specified, we need a special lexer which collects escaped text
# before running the chosen language lexer.
escapeinside = parsed_opts.get('escapeinside', '')
if len(escapeinside) == 2 and isinstance(fmter, LatexFormatter):
left = escapeinside[0]
right = escapeinside[1]
lexer = LatexEmbeddedLexer(left, right, lexer)
# ... and do it!
if '-s' not in opts:
# process whole input as per normal...
highlight(code, lexer, fmter, outfile)
return 0
else:
# line by line processing of stdin (eg: for 'tail -f')...
try:
while 1:
if sys.version_info > (3, ):
# Python 3: we have to use .buffer to get a binary stream
line = sys.stdin.buffer.readline()
else:
line = sys.stdin.readline()
if not line:
break
if not inencoding:
line = guess_decode_from_terminal(line, sys.stdin)[0]
highlight(line, lexer, fmter, outfile)
if hasattr(outfile, 'flush'):
outfile.flush()
return 0
except KeyboardInterrupt: # pragma: no cover
return 0
def main_inner(popts, args, usage):
opts = {}
O_opts = []
P_opts = []
F_opts = []
for opt, arg in popts:
if opt == '-O':
O_opts.append(arg)
elif opt == '-P':
P_opts.append(arg)
elif opt == '-F':
F_opts.append(arg)
opts[opt] = arg
if opts.pop('-h', None) is not None:
print(usage)
return 0
if opts.pop('-V', None) is not None:
print('Pygments version %s, (c) 2006-2019 by Georg Brandl.' %
__version__)
return 0
# handle ``pygmentize -L``
L_opt = opts.pop('-L', None)
if L_opt is not None:
if opts:
print >> sys.stderr, usage
return 2
# print version
main(['', '-V'])
if not args:
args = ['lexer', 'formatter', 'filter', 'style']
for arg in args:
_print_list(arg.rstrip('s'))
return 0
# handle ``pygmentize -H``
H_opt = opts.pop('-H', None)
if H_opt is not None:
if opts or len(args) != 2:
print >> sys.stderr, usage
return 2
what, name = args # pylint: disable=unbalanced-tuple-unpacking
if what not in ('lexer', 'formatter', 'filter'):
print >> sys.stderr, usage,
return 2
return _print_help(what, name)
# parse -O options
parsed_opts = _parse_options(O_opts)
opts.pop('-O', None)
# parse -P options
for p_opt in P_opts:
try:
name, value = p_opt.split('=', 1)
except ValueError:
parsed_opts[p_opt] = True
else:
parsed_opts[name] = value
opts.pop('-P', None)
# encodings
inencoding = parsed_opts.get('inencoding', parsed_opts.get('encoding'))
outencoding = parsed_opts.get('outencoding', parsed_opts.get('encoding'))
# handle ``pygmentize -N``
infn = opts.pop('-N', None)
if infn is not None:
lexer = find_lexer_class_for_filename(infn)
if lexer is None:
lexer = TextLexer
print(lexer.aliases[0])
return 0
# handle ``pygmentize -S``
S_opt = opts.pop('-S', None)
a_opt = opts.pop('-a', None)
if S_opt is not None:
f_opt = opts.pop('-f', None)
if not f_opt:
print >> sys.stderr, usage
return 2
if opts or args:
print >> sys.stderr, usage
return 2
try:
parsed_opts['style'] = S_opt
fmter = get_formatter_by_name(f_opt, **parsed_opts)
except ClassNotFound, err:
print >> sys.stderr, err
return 1
print(fmter.get_style_defs(a_opt or ''))
return 0
