def htmlprep(file):
html = htmldom(file)
widthtags = list(tagsinList(html, tagsWithAttribute('width')))
fixwidths(htmlgrid(html, widthtags))
fixspans(html)
notrwidth(html)
return html
def getcss(doc):
def fullsplit(csstring):
for classname, cmds in cssplit1(csstring):
subcss = dict()
for cmd, value in cssplit2(cmds): subcss[cmd] = value
cssdict[classname] = subcss
import urllib
cssdict, count = dict(), 1
for tag in tagsinList(doc, ['style', 'link']):
if isname(tag, 'link'):
if attget(tag, 'rel') == 'stylesheet':
url = attget(tag, 'href')
if url.find('http:') != -1: style = urllib.urlopen(url).read()
else: style = open(urllib.url2pathname(url)).read()
fullsplit(style)
elif isname(tag, 'style'):
for i in tag.childNodes:
if istext(i) or iscomment(i): fullsplit(i.data)
for tag in tagsbyAttr(doc, 'style'):
subcss = dict()
for cmd, value in cssplit2(attget(tag, 'style')): subcss[cmd] = value
if subcss not in cssdict.values():
cssdict[''.join(['.ecss', str(count)])] = subcss
count += 1
return cssdict
def cssifier(doc, **kwargs):
def insertcss(attdict):
for id in attdict:
if id[1] in map:
mapping = map.get(id[1])
if mapping:
if type(mapping) == types.TupleType:
if len(mapping) > 1:
for item in mapping:
if item.find('%s') != -1:
value = attlist.get(id).value
try:
if int(value):
value = ''.join([value, 'px'])
except ValueError: pass
delete.append(id)
yield item % value
else:
delete.append(id)
yield item
elif len(mapping) == 1:
if mapping not in css:
value = tag.attributes.get(id).value
css[mapping % value] = 0
delete.append(id)
elif type(mapping) == types.DictType:
value = mapping.get(tag.attributes.get(id).value)
if value:
delete.append(id)
yield value
elif mapping.find('%s') != -1:
value = tag.attributes.get(id).value
try:
if int(value): value = ''.join([value, 'px'])
except ValueError: pass
delete.append(id)
yield mapping % value
def recss(cdict):
return '; '.join(['%s: %s' % (i, j) for i, j in cdict.iteritems()])
from htmldefs import html2css
import types
if 'embed' in kwargs: embed = 1
else: embed = None
if 'file' in kwargs: file = kwargs['file']
else: file = None
css, tags, count = getcss(doc), list(tagsbyNS(htmlns, doc)), 1
for tag in tags:
if tag.localName in html2css:
delete, map = list(), html2css.get(tag.localName)
tcss, scss = list(insertcss(tag.attributes)), dict()
if len(delete):
for i in delete: del tag.attributes[i]
if len(tcss):
for cmd, value in cssplit2('; '.join(tcss)): scss[cmd] = value
if hasatt(tag, 'class'):
scss.update(css.get(''.join(['.', attget(tag, 'class')])))
attdel(tag, 'class')
if hasatt(tag, 'style'):
style = css.get(getattr(tag, 'style'))
for cmd, value in cssplit2(style): scss[cmd] = value
attdel(tag, 'style')
if scss not in css.values():
classvalue = ''.join(['css', str(count)])
css[''.join(['.', classvalue])] = scss
count += 1
elif scss in css.values():
for i, j in css.iteritems():
if j == scss: classvalue = i.strip('.')
if embed or file: attset(tag, 'class', classvalue)
else: attset(tag, 'style', recss(scss))
if embed or file:
cssgather = [('%s {%s}' % (i, recss(css.get(i)))) for i in css]
cssgather.sort()
cssmaster = '\n'.join(cssgather)
for i in tagsinList(doc, ['style', 'link']):
if isname(i, 'link'):
if hasatt(i, 'rel'):
if attget(i, 'rel') == 'stylesheet':
i.parentNode.removeChild(i)
elif isname(i, 'style'): i.parentNode.removeChild(i)
if kwargs:
if embed:
style = doc.createElementNS(htmlns, 'style')
style.appendChild(doc.createComment(cssmaster))
list(tags(doc, htmlns, 'head'))[0].appendChild(style)
elif file:
import urllib
open(file, 'wb').write(cssmaster)
style = doc.createElementNS(htmlns, 'link')
attset(style, 'rel', 'stylesheet')
attset(style, 'href', urllib.pathname2url(file))
attset(style, 'type', 'text/css')
list(tags(doc, htmlns, 'head'))[0].appendChild(style)
