def post_html_transform(doc, url, config):
"""
User-customizable HTML transform.
Given an HTML document (with URLs already rewritten), returns
modified HTML document.
"""
if config.hack_skin:
if config.skin == MONOBOOK_SKIN:
doc = monobook_fix_html_sidebar(doc, config)
doc = monobook_hack_skin_html(doc, config)
else:
raise ValueError('unknown skin')
if config.move_href:
doc = fix_move_href_tags(doc, config)
if config.remove_history:
doc = html_remove_image_history(doc, config)
#remove some HTML
L = htmldata.tagextract(doc)
R = []
prev = None
skip = False
for item in L:
#logging.debug("item %s"% repr(item))
# remove edit links
# if isinstance(item, tuple) and item[0] == 'span' \
# and 'class' in item[1] and item[1]['class'] == 'editsection':
# R.append(item)
# skip=True
# elif skip == True and not(isinstance(item, tuple) and item[0] == '/span'):
# logging.debug("Remove item %s " % repr(item))
# elif skip == True and isinstance(item, tuple) and item[0] == '/span':
# R.append(' ') #necessary to prevent eclipse ruiningthe html when
# #inserting anchors
# R.append(item)
# skip=False
# remove empty div
if isinstance(item, tuple) and item[0] == 'div':
if prev != None:
R.append(prev)
prev = item
elif prev != None:
if isinstance(item, tuple) and item[0] == '/div':
logging.debug("Remove item %s %s" % (repr(prev), repr(item)))
R.append(('<!-- %s %s -->' % (repr(prev), repr(item))))
prev = None
else:
R.append(prev)
R.append(item)
prev = None
else:
R.append(item)
doc = htmldata.tagjoin(R)
return doc
def safe_html(data):
"""
Remove all tag attributes from html except a.href and img.src
"""
data = normalize_html(data)
tree = htmldata.tagextract(data)
for elem in tree:
if isinstance(elem, tuple):
for attr in elem[1].keys():
if 'a' == elem[0] and 'href' == attr:
continue
if 'img/' == elem[0] and 'src' == attr:
continue
del elem[1][attr]
return htmldata.tagjoin(tree)
def safe_html(data):
"""
Remove all tag attributes from html except a.href and img.src
"""
data = normalize_html(data)
tree = htmldata.tagextract(data)
for elem in tree:
if isinstance(elem, tuple):
for attr in elem[1].keys():
if 'a' == elem[0] and 'href' == attr:
continue
if 'img/' == elem[0] and 'src' == attr:
continue
del elem[1][attr]
data = htmldata.tagjoin(tree)
# Temporary hack
# htmldata doing something shitty with html:
# tagjoin return invalid DIV
# Data for testing: http://py-algorithm.blogspot.com/2011/04/blog-post_3267.html
data = normalize_html(data)
return data
def fix_move_href_tags(doc):
"""
Return copy of doc with all MOVE_HREF tags removed.
"""
while "<" + MOVE_HREF in doc:
i1 = doc.index("<" + MOVE_HREF)
i2 = doc.index("</" + MOVE_HREF, i1 + 1)
i3 = doc.index(">", i2 + 1)
(start, end) = (i1, i3 + 1)
tags = htmldata.tagextract(doc[start:end])
assert tags[0][0] == MOVE_HREF
assert tags[-1][0] == "/" + MOVE_HREF
href = tags[0][1].get("href", "")
new_tags = []
for tag in tags[1:-1]:
if len(tag) == 2:
if "href" in tag[1]:
if href == "":
continue
tag[1]["href"] = href
new_tags += [tag]
doc = doc[:start] + htmldata.tagjoin(new_tags) + doc[end:]
return doc
def fix_move_href_tags(doc, config):
"""
Return copy of doc with all MOVE_HREF tags removed.
"""
while '<' + MOVE_HREF in doc:
i1 = doc.index('<' + MOVE_HREF)
i2 = doc.index('</' + MOVE_HREF, i1 + 1)
i3 = doc.index('>', i2 + 1)
(start, end) = (i1, i3 + 1)
tags = htmldata.tagextract(doc[start:end])
assert tags[0][0] == MOVE_HREF
assert tags[-1][0] == '/' + MOVE_HREF
href = tags[0][1].get('href', '')
new_tags = []
for tag in tags[1:-1]:
if len(tag) == 2:
if 'href' in tag[1]:
if href == '':
continue
tag[1]['href'] = href
new_tags += [tag]
doc = doc[:start] + htmldata.tagjoin(new_tags) + doc[end:]
return doc
def fix_move_href_tags(doc):
"""
Return copy of doc with all MOVE_HREF tags removed.
"""
while '<' + MOVE_HREF in doc:
i1 = doc.index('<' + MOVE_HREF)
i2 = doc.index('</' + MOVE_HREF, i1 + 1)
i3 = doc.index('>', i2 + 1)
(start, end) = (i1, i3 + 1)
tags = htmldata.tagextract(doc[start:end])
assert tags[0][0] == MOVE_HREF
assert tags[-1][0] == '/' + MOVE_HREF
href = tags[0][1].get('href', '')
new_tags = []
for tag in tags[1:-1]:
if len(tag) == 2:
if 'href' in tag[1]:
if href == '':
continue
tag[1]['href'] = href
new_tags += [tag]
doc = doc[:start] + htmldata.tagjoin(new_tags) + doc[end:]
return doc
