staticoso/Source/Libs/rcssmin.py

422 lines
14 KiB
Python

#!/usr/bin/env python
# -*- coding: ascii -*-
u"""
==============
CSS Minifier
==============
CSS Minifier.
The minifier is based on the semantics of the `YUI compressor`_\\, which
itself is based on `the rule list by Isaac Schlueter`_\\.
:Copyright:
Copyright 2011 - 2022
Andr\xe9 Malo or his licensors, as applicable
:License:
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
This module is a re-implementation aiming for speed instead of maximum
compression, so it can be used at runtime (rather than during a preprocessing
step). RCSSmin does syntactical compression only (removing spaces, comments
and possibly semicolons). It does not provide semantic compression (like
removing empty blocks, collapsing redundant properties etc). It does, however,
support various CSS hacks (by keeping them working as intended).
Here's a feature list:
- Strings are kept, except that escaped newlines are stripped
- Space/Comments before the very end or before various characters are
stripped: ``:{});=>],!`` (The colon (``:``) is a special case, a single
space is kept if it's outside a ruleset.)
- Space/Comments at the very beginning or after various characters are
stripped: ``{}(=:>[,!``
- Optional space after unicode escapes is kept, resp. replaced by a simple
space
- whitespaces inside ``url()`` definitions are stripped, except if it's a
quoted non-base64 data url
- Comments starting with an exclamation mark (``!``) can be kept optionally.
- All other comments and/or whitespace characters are replaced by a single
space.
- Multiple consecutive semicolons are reduced to one
- The last semicolon within a ruleset is stripped
- CSS Hacks supported:
- IE7 hack (``>/**/``)
- Mac-IE5 hack (``/*\\*/.../**/``)
- The boxmodelhack is supported naturally because it relies on valid CSS2
strings
- Between ``:first-line`` and the following comma or curly brace a space is
inserted. (apparently it's needed for IE6)
- Same for ``:first-letter``
rcssmin.c is a reimplementation of rcssmin.py in C and improves runtime up to
factor 100 or so (depending on the input). docs/BENCHMARKS in the source
distribution contains the details.
Supported python versions are 2.7 and 3.4+.
.. _YUI compressor: https://github.com/yui/yuicompressor/
.. _the rule list by Isaac Schlueter: https://github.com/isaacs/cssmin/
"""
__author__ = u"Andr\xe9 Malo"
__license__ = "Apache License, Version 2.0"
__version__ = '1.1.1'
__all__ = ['cssmin']
import re as _re
def _make_cssmin(python_only=False):
"""
Generate CSS minifier.
Parameters:
python_only (bool):
Use only the python variant. If true, the c extension is not even
tried to be loaded.
Returns:
callable: Minifier
"""
# pylint: disable = unused-variable, possibly-unused-variable
# pylint: disable = too-many-locals, too-many-statements
if not python_only:
try:
import _rcssmin # pylint: disable = import-outside-toplevel
except ImportError:
pass
else:
# Ensure that the C version is in sync
if getattr(_rcssmin, '__version__', None) == __version__:
return _rcssmin.cssmin
nl = r'(?:[\n\f]|\r\n?)' # pylint: disable = invalid-name
spacechar = r'[\r\n\f\040\t]'
unicoded = r'[0-9a-fA-F]{1,6}(?:[\040\n\t\f]|\r\n?)?'
escaped = r'[^\n\r\f0-9a-fA-F]'
escape = r'(?:\\(?:%(unicoded)s|%(escaped)s))' % locals()
nmchar = r'[^\000-\054\056\057\072-\100\133-\136\140\173-\177]'
# nmstart = r'[^\000-\100\133-\136\140\173-\177]'
# ident = (r'(?:'
# r'-?(?:%(nmstart)s|%(escape)s)%(nmchar)s*(?:%(escape)s%(nmchar)s*)*'
# r')') % locals()
comment = r'(?:/\*[^*]*\*+(?:[^/*][^*]*\*+)*/)'
# only for specific purposes. The bang is grouped:
_bang_comment = r'(?:/\*(!?)[^*]*\*+(?:[^/*][^*]*\*+)*/)'
string1 = \
r'(?:\047[^\047\\\r\n\f]*(?:\\[^\r\n\f][^\047\\\r\n\f]*)*\047)'
string2 = r'(?:"[^"\\\r\n\f]*(?:\\[^\r\n\f][^"\\\r\n\f]*)*")'
strings = r'(?:%s|%s)' % (string1, string2)
nl_string1 = \
r'(?:\047[^\047\\\r\n\f]*(?:\\(?:[^\r]|\r\n?)[^\047\\\r\n\f]*)*\047)'
nl_string2 = r'(?:"[^"\\\r\n\f]*(?:\\(?:[^\r]|\r\n?)[^"\\\r\n\f]*)*")'
nl_strings = r'(?:%s|%s)' % (nl_string1, nl_string2)
uri_nl_string1 = r'(?:\047[^\047\\]*(?:\\(?:[^\r]|\r\n?)[^\047\\]*)*\047)'
uri_nl_string2 = r'(?:"[^"\\]*(?:\\(?:[^\r]|\r\n?)[^"\\]*)*")'
uri_nl_strings = r'(?:%s|%s)' % (uri_nl_string1, uri_nl_string2)
nl_escaped = r'(?:\\%(nl)s)' % locals()
space = r'(?:%(spacechar)s|%(comment)s)' % locals()
ie7hack = r'(?:>/\*\*/)'
uri = (
# noqa pylint: disable = bad-option-value, bad-continuation
r'(?:'
r'(?:[^\000-\040"\047()\\\177]*'
r'(?:%(escape)s[^\000-\040"\047()\\\177]*)*)'
r'(?:'
r'(?:%(spacechar)s+|%(nl_escaped)s+)'
r'(?:'
r'(?:[^\000-\040"\047()\\\177]|%(escape)s|%(nl_escaped)s)'
r'[^\000-\040"\047()\\\177]*'
r'(?:%(escape)s[^\000-\040"\047()\\\177]*)*'
r')+'
r')*'
r')'
) % locals()
nl_unesc_sub = _re.compile(nl_escaped).sub
uri_data_plain = _re.compile((
r'[\047"][dD][aA][tT][aA]:([^\000-\040\\"\047,]*),'
)).match
uri_space_sub = _re.compile((
r'(%(escape)s+)|%(spacechar)s+|%(nl_escaped)s+'
) % locals()).sub
uri_space_subber = lambda m: m.groups()[0] or ''
space_sub_simple = _re.compile((
r'[\r\n\f\040\t;]+|(%(comment)s+)'
) % locals()).sub
space_sub_banged = _re.compile((
r'[\r\n\f\040\t;]+|(%(_bang_comment)s+)'
) % locals()).sub
post_esc_sub = _re.compile(r'[\r\n\f\t]+').sub
main_sub = _re.compile((
# noqa pylint: disable = bad-option-value, bad-continuation
r'([^\\"\047u>@\r\n\f\040\t/;:{}+]+)' # 1
r'|(?<=[{}(=:>[,!])(%(space)s+)' # 2
r'|^(%(space)s+)' # 3
r'|(%(space)s+)(?=(([:{});=>\],!])|$)?)' # 4, 5, 6
r'|;(%(space)s*(?:;%(space)s*)*)(?=(\})?)' # 7, 8
r'|(\{)' # 9
r'|(\})' # 10
r'|(%(strings)s)' # 11
r'|(?<!%(nmchar)s)url\(%(spacechar)s*(' # 12
r'%(uri_nl_strings)s'
r'|%(uri)s'
r')%(spacechar)s*\)'
r'|(@(?:' # 13
r'[mM][eE][dD][iI][aA]'
r'|[sS][uU][pP][pP][oO][rR][tT][sS]'
r'|[dD][oO][cC][uU][mM][eE][nN][tT]'
r'|(?:-(?:'
r'[wW][eE][bB][kK][iI][tT]|[mM][oO][zZ]|[oO]|[mM][sS]'
r')-)?'
r'[kK][eE][yY][fF][rR][aA][mM][eE][sS]'
r'))(?!%(nmchar)s)'
r'|(%(ie7hack)s)(%(space)s*)' # 14, 15
r'|(:[fF][iI][rR][sS][tT]-[lL]' # 16
r'(?:[iI][nN][eE]|[eE][tT][tT][eE][rR]))'
r'(%(space)s*)(?=[{,])' # 17
r'|(%(nl_strings)s)' # 18
r'|(%(escape)s[^\\"\047u>@\r\n\f\040\t/;:{}+]*)' # 19
) % locals()).sub
# print(main_sub.__self__.pattern)
def main_subber(keep_bang_comments):
""" Make main subber """
in_macie5, in_rule, at_group = [0], [0], [0]
if keep_bang_comments:
space_sub = space_sub_banged
def space_subber(match):
""" Space|Comment subber """
if match.lastindex:
group1, group2 = match.group(1, 2)
if group2:
if group1.endswith(r'\*/'):
in_macie5[0] = 1
else:
in_macie5[0] = 0
return group1
if group1.endswith(r'\*/'):
if in_macie5[0]:
return ''
in_macie5[0] = 1
return r'/*\*/'
elif in_macie5[0]:
in_macie5[0] = 0
return '/**/'
return ''
else:
space_sub = space_sub_simple
def space_subber(match):
""" Space|Comment subber """
if match.lastindex:
if match.group(1).endswith(r'\*/'):
if in_macie5[0]:
return ''
in_macie5[0] = 1
return r'/*\*/'
elif in_macie5[0]:
in_macie5[0] = 0
return '/**/'
return ''
def fn_space_post(group):
""" space with token after """
if group(5) is None or (
group(6) == ':' and not in_rule[0] and not at_group[0]):
return ' ' + space_sub(space_subber, group(4))
return space_sub(space_subber, group(4))
def fn_semicolon(group):
""" ; handler """
return ';' + space_sub(space_subber, group(7))
def fn_semicolon2(group):
""" ; handler """
if in_rule[0]:
return space_sub(space_subber, group(7))
return ';' + space_sub(space_subber, group(7))
def fn_open(_):
""" { handler """
if at_group[0]:
at_group[0] -= 1
else:
in_rule[0] = 1
return '{'
def fn_close(_):
""" } handler """
in_rule[0] = 0
return '}'
def fn_url(group):
""" url() handler """
uri = group(12)
data = uri_data_plain(uri)
if not data or data.group(1).lower().endswith(';base64'):
uri = uri_space_sub(uri_space_subber, uri)
return 'url(%s)' % (uri,)
def fn_at_group(group):
""" @xxx group handler """
at_group[0] += 1
return group(13)
def fn_ie7hack(group):
""" IE7 Hack handler """
if not in_rule[0] and not at_group[0]:
in_macie5[0] = 0
return group(14) + space_sub(space_subber, group(15))
return '>' + space_sub(space_subber, group(15))
table = (
# noqa pylint: disable = bad-option-value, bad-continuation
None,
None,
None,
None,
fn_space_post, # space with token after
fn_space_post, # space with token after
fn_space_post, # space with token after
fn_semicolon, # semicolon
fn_semicolon2, # semicolon
fn_open, # {
fn_close, # }
lambda g: g(11), # string
fn_url, # url(...)
fn_at_group, # @xxx expecting {...}
None,
fn_ie7hack, # ie7hack
None,
lambda g: g(16) + ' ' + space_sub(space_subber, g(17)),
# :first-line|letter followed
# by [{,] (apparently space
# needed for IE6)
lambda g: nl_unesc_sub('', g(18)), # nl_string
lambda g: post_esc_sub(' ', g(19)), # escape
)
def func(match):
""" Main subber """
idx, group = match.lastindex, match.group
if idx > 3:
return table[idx](group)
# shortcuts for frequent operations below:
elif idx == 1: # not interesting
return group(1)
# else: # space with token before or at the beginning
return space_sub(space_subber, group(idx))
return func
def cssmin(style, keep_bang_comments=False):
"""
Minify CSS.
Parameters:
style (str):
CSS to minify
keep_bang_comments (bool):
Keep comments starting with an exclamation mark? (``/*!...*/``)
Returns:
str: Minified style
"""
# pylint: disable = redefined-outer-name
is_bytes, style = _as_str(style)
style = main_sub(main_subber(keep_bang_comments), style)
if is_bytes:
style = style.encode('latin-1')
if is_bytes == 2:
style = bytearray(style)
return style
return cssmin
cssmin = _make_cssmin()
def _as_str(script):
""" Make sure the style is a text string """
is_bytes = False
if str is bytes:
if not isinstance(script, basestring): # noqa pylint: disable = undefined-variable
raise TypeError("Unexpected type")
elif isinstance(script, bytes):
is_bytes = True
script = script.decode('latin-1')
elif isinstance(script, bytearray):
is_bytes = 2
script = script.decode('latin-1')
elif not isinstance(script, str):
raise TypeError("Unexpected type")
return is_bytes, script
if __name__ == '__main__':
def main():
""" Main """
import sys as _sys # pylint: disable = import-outside-toplevel
keep_bang_comments = (
'-b' in _sys.argv[1:]
or '-bp' in _sys.argv[1:]
or '-pb' in _sys.argv[1:]
)
if '-p' in _sys.argv[1:] or '-bp' in _sys.argv[1:] \
or '-pb' in _sys.argv[1:]:
xcssmin = _make_cssmin(python_only=True)
else:
xcssmin = cssmin
_sys.stdout.write(xcssmin(
_sys.stdin.read(), keep_bang_comments=keep_bang_comments
))
main()