2013-09-22 23:22:11 +00:00
|
|
|
# -*- coding: utf-8 -*-
|
|
|
|
"""
|
|
|
|
pygments.formatter
|
|
|
|
~~~~~~~~~~~~~~~~~~
|
|
|
|
|
|
|
|
Base formatter class.
|
|
|
|
|
2014-12-01 06:19:45 +00:00
|
|
|
:copyright: Copyright 2006-2014 by the Pygments team, see AUTHORS.
|
2013-09-22 23:22:11 +00:00
|
|
|
:license: BSD, see LICENSE for details.
|
|
|
|
"""
|
|
|
|
|
|
|
|
import codecs
|
|
|
|
|
2014-12-01 06:19:45 +00:00
|
|
|
from pygments.util import get_bool_opt, string_types
|
2013-09-22 23:22:11 +00:00
|
|
|
from pygments.styles import get_style_by_name
|
|
|
|
|
|
|
|
__all__ = ['Formatter']
|
|
|
|
|
|
|
|
|
|
|
|
def _lookup_style(style):
|
2014-12-01 06:19:45 +00:00
|
|
|
if isinstance(style, string_types):
|
2013-09-22 23:22:11 +00:00
|
|
|
return get_style_by_name(style)
|
|
|
|
return style
|
|
|
|
|
|
|
|
|
|
|
|
class Formatter(object):
|
|
|
|
"""
|
|
|
|
Converts a token stream to text.
|
|
|
|
|
|
|
|
Options accepted:
|
|
|
|
|
|
|
|
``style``
|
|
|
|
The style to use, can be a string or a Style subclass
|
|
|
|
(default: "default"). Not used by e.g. the
|
|
|
|
TerminalFormatter.
|
|
|
|
``full``
|
|
|
|
Tells the formatter to output a "full" document, i.e.
|
|
|
|
a complete self-contained document. This doesn't have
|
|
|
|
any effect for some formatters (default: false).
|
|
|
|
``title``
|
|
|
|
If ``full`` is true, the title that should be used to
|
|
|
|
caption the document (default: '').
|
|
|
|
``encoding``
|
|
|
|
If given, must be an encoding name. This will be used to
|
|
|
|
convert the Unicode token strings to byte strings in the
|
|
|
|
output. If it is "" or None, Unicode strings will be written
|
|
|
|
to the output file, which most file-like objects do not
|
|
|
|
support (default: None).
|
|
|
|
``outencoding``
|
|
|
|
Overrides ``encoding`` if given.
|
|
|
|
"""
|
|
|
|
|
|
|
|
#: Name of the formatter
|
|
|
|
name = None
|
|
|
|
|
|
|
|
#: Shortcuts for the formatter
|
|
|
|
aliases = []
|
|
|
|
|
|
|
|
#: fn match rules
|
|
|
|
filenames = []
|
|
|
|
|
|
|
|
#: If True, this formatter outputs Unicode strings when no encoding
|
|
|
|
#: option is given.
|
|
|
|
unicodeoutput = True
|
|
|
|
|
|
|
|
def __init__(self, **options):
|
|
|
|
self.style = _lookup_style(options.get('style', 'default'))
|
|
|
|
self.full = get_bool_opt(options, 'full', False)
|
|
|
|
self.title = options.get('title', '')
|
|
|
|
self.encoding = options.get('encoding', None) or None
|
2014-12-01 06:19:45 +00:00
|
|
|
if self.encoding in ('guess', 'chardet'):
|
|
|
|
# can happen for e.g. pygmentize -O encoding=guess
|
2013-09-22 23:22:11 +00:00
|
|
|
self.encoding = 'utf-8'
|
2014-12-01 06:19:45 +00:00
|
|
|
self.encoding = options.get('outencoding') or self.encoding
|
2013-09-22 23:22:11 +00:00
|
|
|
self.options = options
|
|
|
|
|
|
|
|
def get_style_defs(self, arg=''):
|
|
|
|
"""
|
|
|
|
Return the style definitions for the current style as a string.
|
|
|
|
|
|
|
|
``arg`` is an additional argument whose meaning depends on the
|
|
|
|
formatter used. Note that ``arg`` can also be a list or tuple
|
|
|
|
for some formatters like the html formatter.
|
|
|
|
"""
|
|
|
|
return ''
|
|
|
|
|
|
|
|
def format(self, tokensource, outfile):
|
|
|
|
"""
|
|
|
|
Format ``tokensource``, an iterable of ``(tokentype, tokenstring)``
|
|
|
|
tuples and write it into ``outfile``.
|
|
|
|
"""
|
|
|
|
if self.encoding:
|
|
|
|
# wrap the outfile in a StreamWriter
|
|
|
|
outfile = codecs.lookup(self.encoding)[3](outfile)
|
|
|
|
return self.format_unencoded(tokensource, outfile)
|