Ruud
12 years ago
10 changed files with 1237 additions and 703 deletions
@ -0,0 +1,223 @@ |
|||
#!/usr/bin/env python |
|||
# -*- coding: utf-8 -*- |
|||
|
|||
# `cssmin.py` - A Python port of the YUI CSS compressor. |
|||
|
|||
|
|||
from StringIO import StringIO # The pure-Python StringIO supports unicode. |
|||
import re |
|||
|
|||
|
|||
__version__ = '0.1.1' |
|||
|
|||
|
|||
def remove_comments(css): |
|||
"""Remove all CSS comment blocks.""" |
|||
|
|||
iemac = False |
|||
preserve = False |
|||
comment_start = css.find("/*") |
|||
while comment_start >= 0: |
|||
# Preserve comments that look like `/*!...*/`. |
|||
# Slicing is used to make sure we don"t get an IndexError. |
|||
preserve = css[comment_start + 2:comment_start + 3] == "!" |
|||
|
|||
comment_end = css.find("*/", comment_start + 2) |
|||
if comment_end < 0: |
|||
if not preserve: |
|||
css = css[:comment_start] |
|||
break |
|||
elif comment_end >= (comment_start + 2): |
|||
if css[comment_end - 1] == "\\": |
|||
# This is an IE Mac-specific comment; leave this one and the |
|||
# following one alone. |
|||
comment_start = comment_end + 2 |
|||
iemac = True |
|||
elif iemac: |
|||
comment_start = comment_end + 2 |
|||
iemac = False |
|||
elif not preserve: |
|||
css = css[:comment_start] + css[comment_end + 2:] |
|||
else: |
|||
comment_start = comment_end + 2 |
|||
comment_start = css.find("/*", comment_start) |
|||
|
|||
return css |
|||
|
|||
|
|||
def remove_unnecessary_whitespace(css): |
|||
"""Remove unnecessary whitespace characters.""" |
|||
|
|||
def pseudoclasscolon(css): |
|||
|
|||
""" |
|||
Prevents 'p :link' from becoming 'p:link'. |
|||
|
|||
Translates 'p :link' into 'p ___PSEUDOCLASSCOLON___link'; this is |
|||
translated back again later. |
|||
""" |
|||
|
|||
regex = re.compile(r"(^|\})(([^\{\:])+\:)+([^\{]*\{)") |
|||
match = regex.search(css) |
|||
while match: |
|||
css = ''.join([ |
|||
css[:match.start()], |
|||
match.group().replace(":", "___PSEUDOCLASSCOLON___"), |
|||
css[match.end():]]) |
|||
match = regex.search(css) |
|||
return css |
|||
|
|||
css = pseudoclasscolon(css) |
|||
# Remove spaces from before things. |
|||
css = re.sub(r"\s+([!{};:>+\(\)\],])", r"\1", css) |
|||
|
|||
# If there is a `@charset`, then only allow one, and move to the beginning. |
|||
css = re.sub(r"^(.*)(@charset \"[^\"]*\";)", r"\2\1", css) |
|||
css = re.sub(r"^(\s*@charset [^;]+;\s*)+", r"\1", css) |
|||
|
|||
# Put the space back in for a few cases, such as `@media screen` and |
|||
# `(-webkit-min-device-pixel-ratio:0)`. |
|||
css = re.sub(r"\band\(", "and (", css) |
|||
|
|||
# Put the colons back. |
|||
css = css.replace('___PSEUDOCLASSCOLON___', ':') |
|||
|
|||
# Remove spaces from after things. |
|||
css = re.sub(r"([!{}:;>+\(\[,])\s+", r"\1", css) |
|||
|
|||
return css |
|||
|
|||
|
|||
def remove_unnecessary_semicolons(css): |
|||
"""Remove unnecessary semicolons.""" |
|||
|
|||
return re.sub(r";+\}", "}", css) |
|||
|
|||
|
|||
def remove_empty_rules(css): |
|||
"""Remove empty rules.""" |
|||
|
|||
return re.sub(r"[^\}\{]+\{\}", "", css) |
|||
|
|||
|
|||
def normalize_rgb_colors_to_hex(css): |
|||
"""Convert `rgb(51,102,153)` to `#336699`.""" |
|||
|
|||
regex = re.compile(r"rgb\s*\(\s*([0-9,\s]+)\s*\)") |
|||
match = regex.search(css) |
|||
while match: |
|||
colors = match.group(1).split(",") |
|||
hexcolor = '#%.2x%.2x%.2x' % tuple(map(int, colors)) |
|||
css = css.replace(match.group(), hexcolor) |
|||
match = regex.search(css) |
|||
return css |
|||
|
|||
|
|||
def condense_zero_units(css): |
|||
"""Replace `0(px, em, %, etc)` with `0`.""" |
|||
|
|||
return re.sub(r"([\s:])(0)(px|em|%|in|cm|mm|pc|pt|ex)", r"\1\2", css) |
|||
|
|||
|
|||
def condense_multidimensional_zeros(css): |
|||
"""Replace `:0 0 0 0;`, `:0 0 0;` etc. with `:0;`.""" |
|||
|
|||
css = css.replace(":0 0 0 0;", ":0;") |
|||
css = css.replace(":0 0 0;", ":0;") |
|||
css = css.replace(":0 0;", ":0;") |
|||
|
|||
# Revert `background-position:0;` to the valid `background-position:0 0;`. |
|||
css = css.replace("background-position:0;", "background-position:0 0;") |
|||
|
|||
return css |
|||
|
|||
|
|||
def condense_floating_points(css): |
|||
"""Replace `0.6` with `.6` where possible.""" |
|||
|
|||
return re.sub(r"(:|\s)0+\.(\d+)", r"\1.\2", css) |
|||
|
|||
|
|||
def condense_hex_colors(css): |
|||
"""Shorten colors from #AABBCC to #ABC where possible.""" |
|||
|
|||
regex = re.compile(r"([^\"'=\s])(\s*)#([0-9a-fA-F])([0-9a-fA-F])([0-9a-fA-F])([0-9a-fA-F])([0-9a-fA-F])([0-9a-fA-F])") |
|||
match = regex.search(css) |
|||
while match: |
|||
first = match.group(3) + match.group(5) + match.group(7) |
|||
second = match.group(4) + match.group(6) + match.group(8) |
|||
if first.lower() == second.lower(): |
|||
css = css.replace(match.group(), match.group(1) + match.group(2) + '#' + first) |
|||
match = regex.search(css, match.end() - 3) |
|||
else: |
|||
match = regex.search(css, match.end()) |
|||
return css |
|||
|
|||
|
|||
def condense_whitespace(css): |
|||
"""Condense multiple adjacent whitespace characters into one.""" |
|||
|
|||
return re.sub(r"\s+", " ", css) |
|||
|
|||
|
|||
def condense_semicolons(css): |
|||
"""Condense multiple adjacent semicolon characters into one.""" |
|||
|
|||
return re.sub(r";;+", ";", css) |
|||
|
|||
|
|||
def wrap_css_lines(css, line_length): |
|||
"""Wrap the lines of the given CSS to an approximate length.""" |
|||
|
|||
lines = [] |
|||
line_start = 0 |
|||
for i, char in enumerate(css): |
|||
# It's safe to break after `}` characters. |
|||
if char == '}' and (i - line_start >= line_length): |
|||
lines.append(css[line_start:i + 1]) |
|||
line_start = i + 1 |
|||
|
|||
if line_start < len(css): |
|||
lines.append(css[line_start:]) |
|||
return '\n'.join(lines) |
|||
|
|||
|
|||
def cssmin(css, wrap = None): |
|||
css = remove_comments(css) |
|||
css = condense_whitespace(css) |
|||
# A pseudo class for the Box Model Hack |
|||
# (see http://tantek.com/CSS/Examples/boxmodelhack.html) |
|||
css = css.replace('"\\"}\\""', "___PSEUDOCLASSBMH___") |
|||
#css = remove_unnecessary_whitespace(css) |
|||
css = remove_unnecessary_semicolons(css) |
|||
css = condense_zero_units(css) |
|||
css = condense_multidimensional_zeros(css) |
|||
css = condense_floating_points(css) |
|||
css = normalize_rgb_colors_to_hex(css) |
|||
css = condense_hex_colors(css) |
|||
if wrap is not None: |
|||
css = wrap_css_lines(css, wrap) |
|||
css = css.replace("___PSEUDOCLASSBMH___", '"\\"}\\""') |
|||
css = condense_semicolons(css) |
|||
return css.strip() |
|||
|
|||
|
|||
def main(): |
|||
import optparse |
|||
import sys |
|||
|
|||
p = optparse.OptionParser( |
|||
prog = "cssmin", version = __version__, |
|||
usage = "%prog [--wrap N]", |
|||
description = """Reads raw CSS from stdin, and writes compressed CSS to stdout.""") |
|||
|
|||
p.add_option( |
|||
'-w', '--wrap', type = 'int', default = None, metavar = 'N', |
|||
help = "Wrap output to approximately N chars per line.") |
|||
|
|||
options, args = p.parse_args() |
|||
sys.stdout.write(cssmin(sys.stdin.read(), wrap = options.wrap)) |
|||
|
|||
|
|||
if __name__ == '__main__': |
|||
main() |
|||
@ -0,0 +1,218 @@ |
|||
#!/usr/bin/python |
|||
|
|||
# This code is original from jsmin by Douglas Crockford, it was translated to |
|||
# Python by Baruch Even. The original code had the following copyright and |
|||
# license. |
|||
# |
|||
# /* jsmin.c |
|||
# 2007-05-22 |
|||
# |
|||
# Copyright (c) 2002 Douglas Crockford (www.crockford.com) |
|||
# |
|||
# Permission is hereby granted, free of charge, to any person obtaining a copy of |
|||
# this software and associated documentation files (the "Software"), to deal in |
|||
# the Software without restriction, including without limitation the rights to |
|||
# use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies |
|||
# of the Software, and to permit persons to whom the Software is furnished to do |
|||
# so, subject to the following conditions: |
|||
# |
|||
# The above copyright notice and this permission notice shall be included in all |
|||
# copies or substantial portions of the Software. |
|||
# |
|||
# The Software shall be used for Good, not Evil. |
|||
# |
|||
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR |
|||
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, |
|||
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE |
|||
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER |
|||
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, |
|||
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE |
|||
# SOFTWARE. |
|||
# */ |
|||
|
|||
from StringIO import StringIO |
|||
|
|||
def jsmin(js): |
|||
ins = StringIO(js) |
|||
outs = StringIO() |
|||
JavascriptMinify().minify(ins, outs) |
|||
str = outs.getvalue() |
|||
if len(str) > 0 and str[0] == '\n': |
|||
str = str[1:] |
|||
return str |
|||
|
|||
def isAlphanum(c): |
|||
"""return true if the character is a letter, digit, underscore, |
|||
dollar sign, or non-ASCII character. |
|||
""" |
|||
return ((c >= 'a' and c <= 'z') or (c >= '0' and c <= '9') or |
|||
(c >= 'A' and c <= 'Z') or c == '_' or c == '$' or c == '\\' or (c is not None and ord(c) > 126)); |
|||
|
|||
class UnterminatedComment(Exception): |
|||
pass |
|||
|
|||
class UnterminatedStringLiteral(Exception): |
|||
pass |
|||
|
|||
class UnterminatedRegularExpression(Exception): |
|||
pass |
|||
|
|||
class JavascriptMinify(object): |
|||
|
|||
def _outA(self): |
|||
self.outstream.write(self.theA) |
|||
def _outB(self): |
|||
self.outstream.write(self.theB) |
|||
|
|||
def _get(self): |
|||
"""return the next character from stdin. Watch out for lookahead. If |
|||
the character is a control character, translate it to a space or |
|||
linefeed. |
|||
""" |
|||
c = self.theLookahead |
|||
self.theLookahead = None |
|||
if c == None: |
|||
c = self.instream.read(1) |
|||
if c >= ' ' or c == '\n': |
|||
return c |
|||
if c == '': # EOF |
|||
return '\000' |
|||
if c == '\r': |
|||
return '\n' |
|||
return ' ' |
|||
|
|||
def _peek(self): |
|||
self.theLookahead = self._get() |
|||
return self.theLookahead |
|||
|
|||
def _next(self): |
|||
"""get the next character, excluding comments. peek() is used to see |
|||
if a '/' is followed by a '/' or '*'. |
|||
""" |
|||
c = self._get() |
|||
if c == '/': |
|||
p = self._peek() |
|||
if p == '/': |
|||
c = self._get() |
|||
while c > '\n': |
|||
c = self._get() |
|||
return c |
|||
if p == '*': |
|||
c = self._get() |
|||
while 1: |
|||
c = self._get() |
|||
if c == '*': |
|||
if self._peek() == '/': |
|||
self._get() |
|||
return ' ' |
|||
if c == '\000': |
|||
raise UnterminatedComment() |
|||
|
|||
return c |
|||
|
|||
def _action(self, action): |
|||
"""do something! What you do is determined by the argument: |
|||
1 Output A. Copy B to A. Get the next B. |
|||
2 Copy B to A. Get the next B. (Delete A). |
|||
3 Get the next B. (Delete B). |
|||
action treats a string as a single character. Wow! |
|||
action recognizes a regular expression if it is preceded by ( or , or =. |
|||
""" |
|||
if action <= 1: |
|||
self._outA() |
|||
|
|||
if action <= 2: |
|||
self.theA = self.theB |
|||
if self.theA == "'" or self.theA == '"': |
|||
while 1: |
|||
self._outA() |
|||
self.theA = self._get() |
|||
if self.theA == self.theB: |
|||
break |
|||
if self.theA <= '\n': |
|||
raise UnterminatedStringLiteral() |
|||
if self.theA == '\\': |
|||
self._outA() |
|||
self.theA = self._get() |
|||
|
|||
|
|||
if action <= 3: |
|||
self.theB = self._next() |
|||
if self.theB == '/' and (self.theA == '(' or self.theA == ',' or |
|||
self.theA == '=' or self.theA == ':' or |
|||
self.theA == '[' or self.theA == '?' or |
|||
self.theA == '!' or self.theA == '&' or |
|||
self.theA == '|' or self.theA == ';' or |
|||
self.theA == '{' or self.theA == '}' or |
|||
self.theA == '\n'): |
|||
self._outA() |
|||
self._outB() |
|||
while 1: |
|||
self.theA = self._get() |
|||
if self.theA == '/': |
|||
break |
|||
elif self.theA == '\\': |
|||
self._outA() |
|||
self.theA = self._get() |
|||
elif self.theA <= '\n': |
|||
raise UnterminatedRegularExpression() |
|||
self._outA() |
|||
self.theB = self._next() |
|||
|
|||
|
|||
def _jsmin(self): |
|||
"""Copy the input to the output, deleting the characters which are |
|||
insignificant to JavaScript. Comments will be removed. Tabs will be |
|||
replaced with spaces. Carriage returns will be replaced with linefeeds. |
|||
Most spaces and linefeeds will be removed. |
|||
""" |
|||
self.theA = '\n' |
|||
self._action(3) |
|||
|
|||
while self.theA != '\000': |
|||
if self.theA == ' ': |
|||
if isAlphanum(self.theB): |
|||
self._action(1) |
|||
else: |
|||
self._action(2) |
|||
elif self.theA == '\n': |
|||
if self.theB in ['{', '[', '(', '+', '-']: |
|||
self._action(1) |
|||
elif self.theB == ' ': |
|||
self._action(3) |
|||
else: |
|||
if isAlphanum(self.theB): |
|||
self._action(1) |
|||
else: |
|||
self._action(2) |
|||
else: |
|||
if self.theB == ' ': |
|||
if isAlphanum(self.theA): |
|||
self._action(1) |
|||
else: |
|||
self._action(3) |
|||
elif self.theB == '\n': |
|||
if self.theA in ['}', ']', ')', '+', '-', '"', '\'']: |
|||
self._action(1) |
|||
else: |
|||
if isAlphanum(self.theA): |
|||
self._action(1) |
|||
else: |
|||
self._action(3) |
|||
else: |
|||
self._action(1) |
|||
|
|||
def minify(self, instream, outstream): |
|||
self.instream = instream |
|||
self.outstream = outstream |
|||
self.theA = '\n' |
|||
self.theB = None |
|||
self.theLookahead = None |
|||
|
|||
self._jsmin() |
|||
self.instream.close() |
|||
|
|||
if __name__ == '__main__': |
|||
import sys |
|||
jsm = JavascriptMinify() |
|||
jsm.minify(sys.stdin, sys.stdout) |
|||
Loading…
Reference in new issue