class HTML2Text(HTMLParser):
entities = {
- "amp": "&",
- "lt": "<",
- "gt": ">",
- "pound": "£",
- "copy": "©",
- "apos": "'",
- "quot": "\"",
- "nbsp": " ",
+ u'amp': "&",
+ u'lt': "<",
+ u'gt': ">",
+ u'pound': "£",
+ u'copy': "©",
+ u'apos': "'",
+ u'quot': "\"",
+ u'nbsp': " ",
}
blockleveltags = [
- "h1",
- "h2",
- "h3",
- "h4",
- "h5",
- "h6",
- "pre",
- "p",
- "ul",
- "ol",
- "dl",
- "br",
+ u'h1',
+ u'h2',
+ u'h3',
+ u'h4',
+ u'h5',
+ u'h6',
+ u'pre',
+ u'p',
+ u'ul',
+ u'ol',
+ u'dl',
+ u'div',
+ #u'blockquote',
]
liststarttags = [
- "ul",
- "ol",
- "dl",
+ u'ul',
+ u'ol',
+ u'dl',
]
cancontainflow = [
- "div",
- "li",
- "dd",
- "blockquote",
+ u'div',
+ u'li',
+ u'dd',
+ u'blockquote',
]
def __init__(self,textwidth=70):
self.textwidth = textwidth
self.opentags = []
self.indentlevel = 0
+ self.ignorenodata = False
self.listcount = []
+ self.urls = []
HTMLParser.__init__(self)
def handle_starttag(self, tag, attrs):
# handle starting a new block - unless we're in a block element
# that can contain other blocks, we'll assume that we want to close
# the container
- if tag_name == u'br':
+ if len(self.opentags) > 1 and self.opentags[-1] == u'li':
self.handle_curdata()
- self.opentags.append(tag_name)
- self.opentags.pop()
if tag_name == u'ol':
self.handle_curdata()
self.listlevel = len(self.listcount) - 1
if tag_name in self.liststarttags:
- smallist = self.opentags[-3:]
+ smallist = self.opentags[-3:-1]
smallist.reverse()
for prev_listtag in smallist:
if prev_listtag in [u'dl', u'ol']:
self.opentags.pop()
self.opentags.append(tag_name)
else:
+ if tag_name == "span":
+ return
listcount = 0
try:
listcount = self.listcount[-1]
except:
pass
- self.handle_curdata()
- self.opentags.append(tag_name)
+
+ 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
+ elif tag_name == u'img':
+ self.handle_image(attrs)
+ return
+ elif tag_name == u'br':
+ self.handle_br()
+ return
+ else:
+ # we don't know the tag, so lets avoid handling it!
+ return
def handle_startendtag(self, tag, attrs):
if tag.lower() == u'br':
- self.tags.append(u'br')
- self.handle_curdata() # just handle the data, don't do anything else
- self.tags.pop()
+ self.handle_br()
+ elif tag.lower() == u'img':
+ self.handle_image(attrs)
+ return
+
+ def handle_br(self):
+ self.handle_curdata()
+ self.opentags.append(u'br')
+ self.handle_curdata()
+ self.opentags.pop()
+
+ def handle_image(self, attrs):
+ alt = u''
+ url = u''
+ for attr in attrs:
+ if attr[0] == 'alt':
+ alt = attr[1]
+ elif attr[0] == 'src':
+ url = attr[1]
+ if url:
+ self.curdata = self.curdata \
+ + u' [img:' \
+ + unicode( \
+ url.encode('utf-8'), \
+ 'utf-8')
+ if alt:
+ self.curdata = self.curdata \
+ + u'(' \
+ + unicode( \
+ alt.encode('utf-8'), \
+ 'utf-8') \
+ + u')'
+ self.curdata = self.curdata \
+ + u']'
def handle_curdata(self):
+
if len(self.opentags) == 0:
return
+ tag_thats_done = self.opentags[-1]
+
if len(self.curdata) == 0:
return
- if len(self.curdata.strip()) == 0:
+ if tag_thats_done == u'br':
+ if len(self.text) == 0 or self.text[-1] != '\n':
+ self.text = self.text + '\n'
+ self.ignorenodata = True
return
- tag_thats_done = self.opentags[-1]
+ if len(self.curdata.strip()) == 0:
+ return
if tag_thats_done in self.blockleveltags:
newlinerequired = self.text != u''
- if newlinerequired:
- if newlinerequired \
- and len(self.text) > 2 \
- and self.text[-1] != u'\n' \
- and self.text[-2] != u'\n':
+ if self.ignorenodata:
+ newlinerequired = False
+ self.ignorenodata = False
+ 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 = unicode( \
+ self.curdata.encode("utf-8").strip(), "utf-8")
seperator = u'\n' + u' '*self.indentlevel
headingtext = seperator.join( \
textwrap.wrap( \
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 in [u'p', u'div']:
+ paragraph = unicode( \
+ self.curdata.strip().encode("utf-8"), "utf-8")
seperator = u'\n' + u' ' * self.indentlevel
self.text = self.text \
+ u' ' * self.indentlevel \
- + seperator.join(textwrap.wrap(paragraph, self.textwidth - self.indentlevel))
+ + seperator.join( \
+ textwrap.wrap( \
+ paragraph, self.textwidth - self.indentlevel))
elif tag_thats_done == "pre":
- self.text = self.text + self.curdata
- elif tag_thats_done == "blockquote":
- quote = self.curdata.encode("utf-8").strip()
+ self.text = self.text + unicode( \
+ self.curdata.encode("utf-8"), "utf-8")
+ elif tag_thats_done == u'blockquote':
+ quote = unicode( \
+ self.curdata.encode("utf-8").strip(), "utf-8")
seperator = u'\n' + u' ' * self.indentlevel + u'> '
+ if len(self.text) > 0 and self.text[-1] != u'\n':
+ self.text = self.text + u'\n'
self.text = self.text \
+ u'> ' \
+ seperator.join( \
self.textwidth - self.indentlevel - 2 \
)
)
+ self.curdata = u''
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 = False
+ isul = None
for thing in latesttags:
if thing == 'ul':
isul = True
listindent = 4
listmarker = u' * '
- if not isul:
+ if isul == False:
listmarker = u' %2d. ' %(self.listcount[-1])
self.listcount[-1] = self.listcount[-1] + 1
) \
)
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) > 1 and self.text[-2] != u'\n':
self.text = self.text + u'\n'
- definition = definition + "::"
+ definition = u' ' * self.indentlevel + definition + "::"
+ indentstring = u'\n' + u' ' * (self.indentlevel + 1)
self.text = self.text \
- + '\n '.join(
- textwrap.wrap(definition, self.textwidth - 1))
+ + indentstring.join(
+ textwrap.wrap(definition, \
+ self.textwidth - self.indentlevel - 1))
self.curdata = u''
- elif tag_thats_done == "dd":
- definition = self.curdata.encode("utf-8").strip()
+ 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 \
- + ' ' \
- + '\n '.join( \
+ + 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:
- # we've got no idea what this tag does, so we'll
- # make an assumption that we're not going to know later
- if len(self.curdata) > 0:
- self.text = self.text \
- + u' ... ' \
- + u'\n ... '.join( \
- textwrap.wrap(self.curdata, self.textwidth - 5))
- self.curdata = u''
if tag_thats_done in self.blockleveltags:
self.curdata = u''
+ self.ignorenodata = False
+
def handle_endtag(self, tag):
+ self.ignorenodata = False
+ if tag == "span":
+ return
+
try:
tagindex = self.opentags.index(tag)
except:
- # closing tag we know nothing about.
- # err. weird.
- tagindex = 0
-
+ return
tag = tag.lower()
+ if tag in [u'br', u'img']:
+ return
+
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()
self.opentags.pop()
def handle_data(self, data):
+ if len(self.opentags) == 0:
+ self.opentags.append(u'p')
self.curdata = self.curdata + unicode(data, "utf-8")
def handle_entityref(self, name):
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