self.textwidth = textwidth
self.opentags = []
self.indentlevel = 0
+ self.listcount = []
+ self.urls = []
HTMLParser.__init__(self)
def handle_starttag(self, tag, attrs):
self.opentags.append(tag_name)
self.opentags.pop()
+ if len(self.opentags) > 1 and self.opentags[-1] == u'li':
+ self.handle_curdata()
+
+ if tag_name == u'ol':
+ self.handle_curdata()
+ self.listcount.append(1)
+ self.listlevel = len(self.listcount) - 1
+
+ if tag_name in self.liststarttags:
+ smallist = self.opentags[-3:-1]
+ smallist.reverse()
+ for prev_listtag in smallist:
+ if prev_listtag in [u'dl', u'ol']:
+ self.indentlevel = self.indentlevel + 4
+ break
+ elif prev_listtag == u'ul':
+ self.indentlevel = self.indentlevel + 3
+ break
+
if len(self.opentags) > 0:
self.handle_curdata()
- self.opentags.pop()
+ if tag_name not in self.cancontainflow:
+ self.opentags.pop()
self.opentags.append(tag_name)
else:
+ listcount = 0
+ try:
+ listcount = self.listcount[-1]
+ except:
+ pass
+
+ if tag_name == u'dd' and len(self.opentags) > 1 \
+ and self.opentags[-1] == u'dt':
+ self.handle_curdata()
+ self.opentags.pop()
+ elif tag_name == u'dt' and len(self.opentags) > 1 \
+ and self.opentags[-1] == u'dd':
+ self.handle_curdata()
+ self.opentags.pop()
+ elif tag_name == u'a':
+ for attr in attrs:
+ if attr[0].lower() == u'href':
+ self.urls.append(attr[1])
+ self.curdata = self.curdata + u'`'
+ self.opentags.append(tag_name)
+ return
+
self.handle_curdata()
self.opentags.append(tag_name)
def handle_startendtag(self, tag, attrs):
if tag.lower() == u'br':
- self.tags.append(u'br')
+ self.opentags.append(u'br')
self.handle_curdata() # just handle the data, don't do anything else
- self.tags.pop()
+ self.opentags.pop()
def handle_curdata(self):
if len(self.opentags) == 0:
if len(self.curdata) == 0:
return
+ if len(self.curdata.strip()) == 0:
+ return
+
tag_thats_done = self.opentags[-1]
if tag_thats_done in self.blockleveltags:
newlinerequired = self.text != u''
if newlinerequired:
- self.text = self.text + u'\n\n'
+ if newlinerequired \
+ and len(self.text) > 2 \
+ and self.text[-1] != u'\n' \
+ and self.text[-2] != u'\n':
+ self.text = self.text + u'\n\n'
if tag_thats_done in ["h1", "h2", "h3", "h4", "h5", "h6"]:
underline = u''
underlinechar = u'='
- headingtext = self.curdata.encode("utf-8").strip()
- headingtext = u'\n'.join( \
- textwrap.wrap(headingtext, self.textwidth))
+ headingtext = unicode( \
+ self.curdata.encode("utf-8").strip(), "utf-8")
+ seperator = u'\n' + u' '*self.indentlevel
+ headingtext = seperator.join( \
+ textwrap.wrap( \
+ headingtext, \
+ self.textwidth - self.indentlevel \
+ ) \
+ )
if tag_thats_done == u'h2':
underlinechar = u'-'
underlinechar = u'~'
if u'\n' in headingtext:
- underline = underlinechar * self.textwidth
+ underline = u' ' * self.indentlevel \
+ + underlinechar * (self.textwidth - self.indentlevel)
else:
- underline = underlinechar * len(headingtext)
+ underline = u' ' * self.indentlevel \
+ + underlinechar * len(headingtext)
self.text = self.text \
+ headingtext.encode("utf-8") + u'\n' \
+ underline
- elif tag_thats_done == "p":
- paragraph = self.curdata.encode("utf-8").strip()
+ elif tag_thats_done == u'p':
+ paragraph = unicode( \
+ self.curdata.strip().encode("utf-8"), "utf-8")
+ seperator = u'\n' + u' ' * self.indentlevel
self.text = self.text \
- + u'\n'.join(textwrap.wrap(paragraph, self.textwidth))
+ + u' ' * self.indentlevel \
+ + seperator.join( \
+ textwrap.wrap( \
+ paragraph, self.textwidth - self.indentlevel))
elif tag_thats_done == "pre":
- self.text = self.text + self.curdata
+ self.text = self.text + unicode( \
+ self.curdata.encode("utf-8"), "utf-8")
elif tag_thats_done == "blockquote":
- quote = self.curdata.encode("utf-8").strip()
+ quote = unicode( \
+ self.curdata.encode("utf-8").strip(), "utf-8")
+ seperator = u'\n' + u' ' * self.indentlevel + u'> '
self.text = self.text \
+ u'> ' \
- + u'> '.join(textwrap.wrap(quote, self.textwidth - 2))
+ + seperator.join( \
+ textwrap.wrap( \
+ quote, \
+ self.textwidth - self.indentlevel - 2 \
+ )
+ )
elif tag_thats_done == "li":
- item = self.curdata.encode("utf-8").strip()
+ item = unicode(self.curdata.encode("utf-8").strip(), "utf-8")
if len(self.text) > 0 and self.text[-1] != u'\n':
self.text = self.text + u'\n'
+ # work out if we're in an ol rather than a ul
+ latesttags = self.opentags[-4:]
+ latesttags.reverse()
+ isul = None
+ for thing in latesttags:
+ if thing == 'ul':
+ isul = True
+ break
+ elif thing == 'ol':
+ isul = False
+ break
+
+ listindent = 3
+ if not isul:
+ listindent = 4
+
+ listmarker = u' * '
+ if isul == False:
+ listmarker = u' %2d. ' %(self.listcount[-1])
+ self.listcount[-1] = self.listcount[-1] + 1
+
+ seperator = u'\n' \
+ + u' ' * self.indentlevel \
+ + u' ' * listindent
self.text = self.text \
- + u' * ' \
- + u'\n '.join( \
- textwrap.wrap(item, self.textwidth - 3))
+ + u' ' * self.indentlevel \
+ + listmarker \
+ + seperator.join( \
+ textwrap.wrap( \
+ item, \
+ self.textwidth - self.indentlevel - listindent \
+ ) \
+ )
self.curdata = u''
- elif tag_thats_done == "dt":
- definition = self.curdata.encode("utf-8").strip()
+ elif tag_thats_done == u'dt':
+ definition = unicode(self.curdata.encode("utf-8").strip(), "utf-8")
if len(self.text) > 0 and self.text[-1] != u'\n':
self.text = self.text + u'\n\n'
- elif len(self.text) > 0 and self.text[-2] != u'\n':
- self.text = self.text + u'\n'
- definition = definition + "::"
- self.text = self.text \
- + '\n '.join(
- textwrap.wrap(definition, self.textwidth - 1))
- self.curdata = u''
- elif tag_thats_done == "dd":
- definition = self.curdata.encode("utf-8").strip()
- if len(self.text) > 0 and self.text[-1] != u'\n':
+ elif len(self.text) > 1 and self.text[-2] != u'\n':
self.text = self.text + u'\n'
+ definition = u' ' * self.indentlevel + definition + "::"
+ indentstring = u'\n' + u' ' * (self.indentlevel + 1)
self.text = self.text \
- + ' ' \
- + '\n '.join( \
- textwrap.wrap(definition, self.textwidth - 4))
+ + indentstring.join(
+ textwrap.wrap(definition, \
+ self.textwidth - self.indentlevel - 1))
self.curdata = u''
+ elif tag_thats_done == u'dd':
+ definition = unicode(self.curdata.encode("utf-8").strip(), "utf-8")
+ if len(definition) > 0:
+ if len(self.text) > 0 and self.text[-1] != u'\n':
+ self.text = self.text + u'\n'
+ indentstring = u'\n' + u' ' * (self.indentlevel + 4)
+ self.text = self.text \
+ + u' ' * (self.indentlevel + 4) \
+ + indentstring.join( \
+ textwrap.wrap( \
+ definition, \
+ self.textwidth - self.indentlevel - 4 \
+ ) \
+ )
+ self.curdata = u''
+ elif tag_thats_done == u'a':
+ self.curdata = self.curdata + u'`__'
+ pass
elif tag_thats_done in self.liststarttags:
pass
else:
self.text = self.text \
+ u' ... ' \
+ u'\n ... '.join( \
- textwrap.wrap(self.curdata, self.textwidth - 5))
+ textwrap.wrap( \
+ unicode( \
+ self.curdata.encode("utf-8").strip(), \
+ "utf-8"), self.textwidth - 5))
self.curdata = u''
if tag_thats_done in self.blockleveltags:
# err. weird.
tagindex = 0
+ tag = tag.lower()
+
+ if tag in self.liststarttags:
+ if tag in [u'ol', u'dl', u'ul']:
+ self.handle_curdata()
+ # find if there was a previous list level
+ smalllist = self.opentags[:-1]
+ smalllist.reverse()
+ for prev_listtag in smalllist:
+ if prev_listtag in [u'ol', u'dl']:
+ self.indentlevel = self.indentlevel - 4
+ break
+ elif prev_listtag == u'ul':
+ self.indentlevel = self.indentlevel - 3
+ break
+
+ if tag == u'ol':
+ self.listcount = self.listcount[:-1]
+
while tagindex < len(self.opentags) \
and tag in self.opentags[tagindex+1:]:
try:
# Assuming the data was for the last opened tag first
self.handle_curdata()
# Now kill the list to be a slice before this tag was opened
- self.opentags = self.opentags[:tagindex]
+ self.opentags = self.opentags[:tagindex + 1]
+ else:
+ self.handle_curdata()
+ if self.opentags[-1] == tag:
+ self.opentags.pop()
def handle_data(self, data):
self.curdata = self.curdata + unicode(data, "utf-8")
while len(self.text) > 1 and self.text[-1] == u'\n':
self.text = self.text[:-1]
self.text = self.text + u'\n'
+ if len(self.urls) > 0:
+ self.text = self.text + u'\n__ ' + u'\n__ '.join(self.urls) + u'\n'
+ self.urls = []
return self.text
def open_url(method, url):
if headers:
data = []
for header in headers:
- if header[0] in ["content-md5", "etag", "last-modified", "content-length"]:
+ if header[0] in \
+ ["content-md5", "etag", "last-modified", "content-length"]:
data.append((header[0], header[1]))
if len(data) > 0:
data = urllib.urlencode(data)
projects / rss2maildir.git / blobdiff