def lexer(self) -> Optional[Lexer]:
"""The lexer for this syntax, or None if no lexer was found.
Tries to find the lexer by name if a string was passed to the constructor.
"""
if isinstance(self._lexer, Lexer):
return self._lexer
try:
return get_lexer_by_name(
self._lexer,
stripnl=False,
ensurenl=True,
tabsize=self.tab_size,
)
except ClassNotFound:
return None
def _print_help(what, name):
try:
if what == 'lexer':
cls = get_lexer_by_name(name)
print("Help on the %s lexer:" % cls.name)
print(dedent(cls.__doc__))
elif what == 'formatter':
cls = find_formatter_class(name)
print("Help on the %s formatter:" % cls.name)
print(dedent(cls.__doc__))
elif what == 'filter':
cls = find_filter_class(name)
print("Help on the %s filter:" % name)
print(dedent(cls.__doc__))
return 0
except (AttributeError, ValueError):
print("%s not found!" % what, file=sys.stderr)
return 1
def guess_lexer(cls, path: str, code: Optional[str] = None) -> str:
"""Guess the alias of the Pygments lexer to use based on a path and an optional string of code.
If code is supplied, it will use a combination of the code and the filename to determine the
best lexer to use. For example, if the file is ``index.html`` and the file contains Django
templating syntax, then "html+django" will be returned. If the file is ``index.html``, and no
templating language is used, the "html" lexer will be used. If no string of code
is supplied, the lexer will be chosen based on the file extension..
Args:
path (AnyStr): The path to the file containing the code you wish to know the lexer for.
code (str, optional): Optional string of code that will be used as a fallback if no lexer
is found for the supplied path.
Returns:
str: The name of the Pygments lexer that best matches the supplied path/code.
"""
lexer: Optional[Lexer] = None
lexer_name = "default"
if code:
try:
lexer = guess_lexer_for_filename(path, code)
except ClassNotFound:
pass
if not lexer:
try:
_, ext = os.path.splitext(path)
if ext:
extension = ext.lstrip(".").lower()
lexer = get_lexer_by_name(extension)
except ClassNotFound:
pass
if lexer:
if lexer.aliases:
lexer_name = lexer.aliases[0]
else:
lexer_name = lexer.name
return lexer_name
def main_inner(parser, argns):
if argns.help:
parser.print_help()
return 0
if argns.V:
print('Pygments version %s, (c) 2006-2022 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:
arg_set = set()
for k, v in vars(argns).items():
if v:
arg_set.add(k)
arg_set.discard('L')
arg_set.discard('json')
if arg_set:
parser.print_help(sys.stderr)
return 2
# print version
if not argns.json:
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
if not argns.json:
for arg in largs:
_print_list(arg)
else:
_print_list_as_json(largs)
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 pip._vendor.colorama.initialise as 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 from_path(
cls,
path: str,
encoding: str = "utf-8",
theme: Union[str, SyntaxTheme] = DEFAULT_THEME,
dedent: bool = False,
line_numbers: bool = False,
line_range: Optional[Tuple[int, int]] = None,
start_line: int = 1,
highlight_lines: Optional[Set[int]] = None,
code_width: Optional[int] = None,
tab_size: int = 4,
word_wrap: bool = False,
background_color: Optional[str] = None,
indent_guides: bool = False,
) -> "Syntax":
"""Construct a Syntax object from a file.
Args:
path (str): Path to file to highlight.
encoding (str): Encoding of file.
theme (str, optional): Color theme, aka Pygments style (see https://pygments.org/docs/styles/#getting-a-list-of-available-styles). Defaults to "emacs".
dedent (bool, optional): Enable stripping of initial whitespace. Defaults to True.
line_numbers (bool, optional): Enable rendering of line numbers. Defaults to False.
start_line (int, optional): Starting number for line numbers. Defaults to 1.
line_range (Tuple[int, int], optional): If given should be a tuple of the start and end line to render.
highlight_lines (Set[int]): A set of line numbers to highlight.
code_width: Width of code to render (not including line numbers), or ``None`` to use all available width.
tab_size (int, optional): Size of tabs. Defaults to 4.
word_wrap (bool, optional): Enable word wrapping of code.
background_color (str, optional): Optional background color, or None to use theme color. Defaults to None.
indent_guides (bool, optional): Show indent guides. Defaults to False.
Returns:
[Syntax]: A Syntax object that may be printed to the console
"""
with open(path, "rt", encoding=encoding) as code_file:
code = code_file.read()
lexer = None
lexer_name = "default"
try:
_, ext = os.path.splitext(path)
if ext:
extension = ext.lstrip(".").lower()
lexer = get_lexer_by_name(extension)
lexer_name = lexer.name
except ClassNotFound:
pass
if lexer is None:
try:
lexer_name = guess_lexer_for_filename(path, code).name
except ClassNotFound:
pass
return cls(
code,
lexer_name,
theme=theme,
dedent=dedent,
line_numbers=line_numbers,
line_range=line_range,
start_line=start_line,
highlight_lines=highlight_lines,
code_width=code_width,
tab_size=tab_size,
word_wrap=word_wrap,
background_color=background_color,
indent_guides=indent_guides,
)
def highlight(self,
code: str,
line_range: Optional[Tuple[int, int]] = None) -> Text:
"""Highlight code and return a Text instance.
Args:
code (str): Code to highlight.
line_range(Tuple[int, int], optional): Optional line range to highlight.
Returns:
Text: A text instance containing highlighted syntax.
"""
base_style = self._get_base_style()
justify: JustifyMethod = (
"default" if base_style.transparent_background else "left")
text = Text(
justify=justify,
style=base_style,
tab_size=self.tab_size,
no_wrap=not self.word_wrap,
)
_get_theme_style = self._theme.get_style_for_token
try:
lexer = get_lexer_by_name(
self.lexer_name,
stripnl=False,
ensurenl=True,
tabsize=self.tab_size,
)
except ClassNotFound:
text.append(code)
else:
if line_range:
# More complicated path to only stylize a portion of the code
# This speeds up further operations as there are less spans to process
line_start, line_end = line_range
def line_tokenize() -> Iterable[Tuple[Any, str]]:
"""Split tokens to one per line."""
for token_type, token in lexer.get_tokens(code):
while token:
line_token, new_line, token = token.partition("\n")
yield token_type, line_token + new_line
def tokens_to_spans() -> Iterable[Tuple[str, Optional[Style]]]:
"""Convert tokens to spans."""
tokens = iter(line_tokenize())
line_no = 0
_line_start = line_start - 1
# Skip over tokens until line start
while line_no < _line_start:
_token_type, token = next(tokens)
yield (token, None)
if token.endswith("\n"):
line_no += 1
# Generate spans until line end
for token_type, token in tokens:
yield (token, _get_theme_style(token_type))
if token.endswith("\n"):
line_no += 1
if line_no >= line_end:
break
text.append_tokens(tokens_to_spans())
else:
text.append_tokens(
(token, _get_theme_style(token_type))
for token_type, token in lexer.get_tokens(code))
if self.background_color is not None:
text.stylize(f"on {self.background_color}")
return text