+import md5
+
+import cgi
+import dbm
+
+from HTMLParser import HTMLParser
+
+class HTML2Text(HTMLParser):
+ entities = {
+ "amp": "&",
+ "lt": "<",
+ "gt": ">",
+ "pound": "£",
+ "copy": "©",
+ "apos": "'",
+ "quot": "\"",
+ "nbsp": " ",
+ }
+
+ blockleveltags = [
+ "h1",
+ "h2",
+ "h3",
+ "h4",
+ "h5",
+ "h6",
+ "pre",
+ "p",
+ "ul",
+ "ol",
+ "dl",
+ "br",
+ ]
+
+ liststarttags = [
+ "ul",
+ "ol",
+ "dl",
+ ]
+
+ cancontainflow = [
+ "div",
+ "li",
+ "dd",
+ "blockquote",
+ ]
+
+ def __init__(self,textwidth=70):
+ self.text = u''
+ self.curdata = u''
+ self.textwidth = textwidth
+ self.opentags = []
+ self.indentlevel = 0
+ HTMLParser.__init__(self)
+
+ def handle_starttag(self, tag, attrs):
+ tag_name = tag.lower()
+ if tag_name in self.blockleveltags:
+ # 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':
+ self.handle_curdata()
+ self.opentags.append(tag_name)
+ self.opentags.pop()
+
+ if len(self.opentags) > 0:
+ self.handle_curdata()
+ self.opentags.pop()
+ self.opentags.append(tag_name)
+ else:
+ 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.handle_curdata() # just handle the data, don't do anything else
+ self.tags.pop()
+
+ def handle_curdata(self):
+ if len(self.opentags) == 0:
+ return
+
+ if len(self.curdata) == 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 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))
+
+ if tag_thats_done == u'h2':
+ underlinechar = u'-'
+ elif tag_thats_done != u'h1':
+ underlinechar = u'~'
+
+ if u'\n' in headingtext:
+ underline = underlinechar * self.textwidth
+ else:
+ underline = 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()
+ self.text = self.text \
+ + u'\n'.join(textwrap.wrap(paragraph, self.textwidth))
+ 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 \
+ + u'> ' \
+ + u'> '.join(textwrap.wrap(quote, self.textwidth - 2))
+ elif tag_thats_done == "li":
+ item = self.curdata.encode("utf-8").strip()
+ if len(self.text) > 0 and self.text[-1] != u'\n':
+ self.text = self.text + u'\n'
+ self.text = self.text \
+ + u' * ' \
+ + u'\n '.join( \
+ textwrap.wrap(item, self.textwidth - 3))
+ self.curdata = u''
+ elif tag_thats_done == "dt":
+ definition = self.curdata.encode("utf-8").strip()
+ 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':
+ self.text = self.text + u'\n'
+ self.text = self.text \
+ + ' ' \
+ + '\n '.join( \
+ textwrap.wrap(definition, self.textwidth - 4))
+ self.curdata = u''
+ 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''
+
+ def handle_endtag(self, tag):
+ try:
+ tagindex = self.opentags.index(tag)
+ except:
+ # closing tag we know nothing about.
+ # err. weird.
+ tagindex = 0
+
+ while tagindex < len(self.opentags) \
+ and tag in self.opentags[tagindex+1:]:
+ try:
+ tagindex = self.opentags.index(tag, tagindex+1)
+ except:
+ # well, we don't want to do that then
+ pass
+ if tagindex != len(self.opentags) - 1:
+ # 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]
projects / rss2maildir.git / blobdiff