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 _getByName(self, childList, name):
'''Retrieves child nodes by name
Arguments:
childlist -- list of specific nodes
child -- specific child node'''
for i in childList:
if attget(i, u'name') == name: return i
def fixspans(doc):
tables = list(tags(doc, htmlns, 'table'))
for table in tables:
trcount, tdcount = 0, 0
for tr in table.childNodes:
if tr.localName == 'tr': trcount += 1
for tr in table.childNodes:
count = 0
for td in tr.childNodes:
if td.localName == 'td': count += 1
if count > tdcount: tdcount = count
for tr in table.childNodes:
for td in tr.childNodes:
if iselement(td):
if hasatt(td, 'rowspan'):
if int(attget(td, 'rowspan')) > trcount:
attset(td, 'rowspan', str(trcount))
elif hasatt(td, 'colspan'):
if int(attget(td, 'colspan')) > tdcount:
attset(td, 'colspan', str(tdcount))
grid, temp = [], []
for tr in table.childNodes:
if iselement(tr) and isname(tr, 'tr'):
if len(temp):
grid.append(temp)
temp = []
for td in tr.childNodes:
if iselement(td) and isname(td, 'td'): temp.append(td)
if len(temp): grid.append(temp)
rowspans = list(tagsbyAttr(table, 'rowspan'))
if len(rowspans):
for td in rowspans:
trs, sibs = [], ['td.parentNode']
for x in range(int(attget(td, 'rowspan'))-1):
sibs.append('.nextSibling')
if eval(''.join(sibs)): trs.append(eval(''.join(sibs)))
for nodelist in grid:
for i in nodelist:
if td in nodelist: loc = nodelist.index(td)
for tr in trs:
if tr.firstChild in nodelist: nodelist.insert(loc, td)
for nodelist in grid:
for td in nodelist:
if hasatt(td, 'colspan'):
index = nodelist.index(td)
colspan = int(attget(td, 'colspan'))
offset, largest = len(nodelist[:index]), 0
for nlist in grid:
actual = len(nlist[index:-offset])
if colspan > actual:
if actual > largest:
count = 0
for i in nlist[index:-offset]:
if hasatt(i, 'colspan'):
count += int(attget(i, 'colspan'))
if count != colspan: largest = actual
if largest: attset(td, 'colspan', str(largest))
if int(attget(td, 'colspan')) <= 1: attdel(td, 'colspan')
def htmlgrid(doc, structure):
grid, temp, blist = [], [], []
for node in structure:
if isname(node, 'tr'):
if len(temp):
if temp[0].parentNode in grid[-1]:
if temp not in grid: grid.append(temp)
grid.append([node])
else:
grid.append([node])
if temp not in grid: grid.append(temp)
else: grid.append([node])
temp = []
elif isname(node, 'td'):
if node not in blist:
temp.append(node)
blist.append(node)
sibs = ['node.nextSibling']
for x in range(len(node.parentNode.childNodes)-1):
neval = ''.join(sibs)
if eval(neval):
if eval('.'.join([neval, 'localName'])) == 'td':
if eval(neval) not in blist:
if eval(neval) in structure:
temp.append(eval(neval))
blist.append(eval(neval))
sibs.append('.nextSibling')
elif isname(node, 'table'):
if len(temp):
if temp not in grid: grid.append(temp)
grid.append([node])
if len(temp): grid.append(temp)
rowspans = list(tagsbyAttr(doc, 'rowspan'))
if len(rowspans):
for td in rowspans:
trs, sibs = [], ['td.parentNode']
for x in range(int(attget(td, 'rowspan'))-1):
sibs.append('.nextSibling')
if eval(''.join(sibs)): trs.append(eval(''.join(sibs)))
for nodelist in grid:
for tr in trs:
if tr.firstChild in nodelist: nodelist.append(td)
return grid
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)
def getwidth(node):
return attget(node, 'width')
def fixwidths(grid):
def haswidth(node):
return hasatt(node, 'width')
def getwidth(node):
return attget(node, 'width')
def setwidth(node, width):
attset(node, 'width', width)
def regtest(nodelist):
for i in nodelist:
if haswidth(i):
if getwidth(i).find('%') == -1: return True
def element(name):
return node.ownerDocument.createElementNS(htmlns, name)
maxwidths, count, mixed = [], 0, None
for nodelist in grid:
maxwidths.append(count)
count = 0
for node in nodelist:
if haswidth(node):
width = getwidth(node)
if width.find('%') == -1: count += int(width)
if len(maxwidths):
maxwidths.sort()
maxwidths.reverse()
maxwidth = str(maxwidths[0])
setwidth(grid[0][0], maxwidth)
else: return None
for nodelist in grid[1:]:
nowidths, percents, count = [], [], 0
for node in nodelist:
pwidth = getwidth(node.parentNode)
if pwidth == '': pwidth = maxwidth
if len(nodelist) > 1:
if haswidth(node):
width = getwidth(node)
if width.find('%') == -1: count += int(width)
elif width.find('%') != -1: percents.append(node)
else: nowidths.append(node)
else: setwidth(node, pwidth)
if len(nodelist) > 1:
pwidth = int(pwidth)
if count: pwidth -= count
if len(nowidths): portion = pwidth / len(nowidths)
if len(percents) > 1 and regtest(nodelist): mixed = True
for node in percents:
width = round((float(getwidth(node).strip('%'))*0.01)*pwidth)
attset(node, 'width', str(int(width)))
for node in nowidths: setwidth(node, str(portion))
if mixed:
tds = []
for node in percents:
if node not in tds and isname(node, 'td'):
sibs = ['node']
for x in range(len(percents)):
sibs.append('.nextSibling')
if eval(''.join(sibs)) in percents:
if node not in tds: tds.append(node)
tds.append(eval(''.join(sibs)))
if len(tds):
tcount, td = 0, element('td')
tds[0].parentNode.insertBefore(td, tds[0])
tr, table = element('tr'), element('table')
share, heights = dict(), list()
for t in tds:
att, delete = dict(), dict()
if len(share):
for a in t.attributes:
attr = t.attributes.get(a)
att[attr.name] = attr.value
else:
for a in t.attributes:
attr = t.attributes.get(a)
share[attr.name] = attr.value
if len(att):
for x, y in share.iteritems():
if (x, y) not in att.iteritems():
delete[x] = y
for x in delete: del share[x]
tcount += int(getwidth(t))
tr.appendChild(t)
setwidth(td, str(tcount))
if len(share):
for n, v in share.iteritems(): attset(td, n, v)
attset(table, 'border', '0')
attset(table, 'cellspacing', '0')
attset(table, 'cellpadding', '0')
setwidth(table, str(tcount))
td.appendChild(table)
table.appendChild(tr)
nodelist.append(td)
for x in nodelist:
if hasatt(x, 'height'):
h = attget(x, 'height')
if h.find('%') == -1: heights.append(h)
if len(heights):
heights.sort()
heights.reverse()
max = str(heights[0])
for x in nodelist: attset(x, 'height', max)
mixed = None
