Skip to content

Instantly share code, notes, and snippets.

@amundo

amundo/html2text.py

Created May 14, 2011 05:36
Show Gist options
  • Save amundo/971949 to your computer and use it in GitHub Desktop.
Save amundo/971949 to your computer and use it in GitHub Desktop.
Download ZIP
unescape hex entities in html
Raw
html2text.py
#!/usr/bin/env python
"""html2text: Turn HTML into equivalent Markdown-structured text."""
__version__ = "2.35"
__author__ = "Aaron Swartz ([email protected])"
__copyright__ = "(C) 2004-2008 Aaron Swartz. GNU GPL 3."
__contributors__ = ["Martin 'Joey' Schulze", "Ricardo Reyes"]
# TODO:
# Support decoded entities with unifiable.
# Relative URL resolution
if not hasattr(__builtins__, 'True'): True, False = 1, 0
import re, sys, urllib, htmlentitydefs, codecs, StringIO, types
import sgmllib
sgmllib.charref = re.compile('&#([xX]?[0-9a-fA-F]+)[^0-9a-fA-F]')
try: from textwrap import wrap
except: pass
# Use Unicode characters instead of their ascii psuedo-replacements
UNICODE_SNOB = 0
# Put the links after each paragraph instead of at the end.
LINKS_EACH_PARAGRAPH = 0
# Wrap long lines at position. 0 for no wrapping. (Requires Python 2.3.)
BODY_WIDTH = 78
# Don't show internal links (href="#local-anchor") -- corresponding link targets
# won't be visible in the plain text file anyway.
SKIP_INTERNAL_LINKS = False
### Entity Nonsense ###
def name2cp(k):
if k == 'apos': return ord("'")
if hasattr(htmlentitydefs, "name2codepoint"): # requires Python 2.3
return htmlentitydefs.name2codepoint[k]
else:
k = htmlentitydefs.entitydefs[k]
if k.startswith("&#") and k.endswith(";"): return int(k[2:-1]) # not in latin-1
return ord(codecs.latin_1_decode(k)[0])
unifiable = {'rsquo':"'", 'lsquo':"'", 'rdquo':'"', 'ldquo':'"',
'copy':'(C)', 'mdash':'--', 'nbsp':' ', 'rarr':'->', 'larr':'<-', 'middot':'*',
'ndash':'-', 'oelig':'oe', 'aelig':'ae',
'agrave':'a', 'aacute':'a', 'acirc':'a', 'atilde':'a', 'auml':'a', 'aring':'a',
'egrave':'e', 'eacute':'e', 'ecirc':'e', 'euml':'e',
'igrave':'i', 'iacute':'i', 'icirc':'i', 'iuml':'i',
'ograve':'o', 'oacute':'o', 'ocirc':'o', 'otilde':'o', 'ouml':'o',
'ugrave':'u', 'uacute':'u', 'ucirc':'u', 'uuml':'u'}
unifiable_n = {}
for k in unifiable.keys():
unifiable_n[name2cp(k)] = unifiable[k]
def charref(name):
if name[0] in ['x','X']:
c = int(name[1:], 16)
else:
c = int(name)
if not UNICODE_SNOB and c in unifiable_n.keys():
return unifiable_n[c]
else:
return unichr(c)
def entityref(c):
if not UNICODE_SNOB and c in unifiable.keys():
return unifiable[c]
else:
try: name2cp(c)
except KeyError: return "&" + c
else: return unichr(name2cp(c))
def replaceEntities(s):
s = s.group(1)
if s[0] == "#":
return charref(s[1:])
else: return entityref(s)
r_unescape = re.compile(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));")
def unescape(s):
return r_unescape.sub(replaceEntities, s)
def fixattrs(attrs):
# Fix bug in sgmllib.py
if not attrs: return attrs
newattrs = []
for attr in attrs:
newattrs.append((attr[0], unescape(attr[1])))
return newattrs
### End Entity Nonsense ###
def onlywhite(line):
"""Return true if the line does only consist of whitespace characters."""
for c in line:
if c is not ' ' and c is not ' ':
return c is ' '
return line
def optwrap(text):
"""Wrap all paragraphs in the provided text."""
if not BODY_WIDTH:
return text
assert wrap, "Requires Python 2.3."
result = ''
newlines = 0
for para in text.split("\n"):
if len(para) > 0:
if para[0] is not ' ' and para[0] is not '-' and para[0] is not '*':
for line in wrap(para, BODY_WIDTH):
result += line + "\n"
result += "\n"
newlines = 2
else:
if not onlywhite(para):
result += para + "\n"
newlines = 1
else:
if newlines < 2:
result += "\n"
newlines += 1
return result
def hn(tag):
if tag[0] == 'h' and len(tag) == 2:
try:
n = int(tag[1])
if n in range(1, 10): return n
except ValueError: return 0
class _html2text(sgmllib.SGMLParser):
def __init__(self, out=sys.stdout.write):
sgmllib.SGMLParser.__init__(self)
if out is None: self.out = self.outtextf
else: self.out = out
self.outtext = u''
self.quiet = 0
self.p_p = 0
self.outcount = 0
self.start = 1
self.space = 0
self.a = []
self.astack = []
self.acount = 0
self.list = []
self.blockquote = 0
self.pre = 0
self.startpre = 0
self.lastWasNL = 0
self.abbr_title = None # current abbreviation definition
self.abbr_data = None # last inner HTML (for abbr being defined)
self.abbr_list = {} # stack of abbreviations to write later
def outtextf(self, s):
self.outtext += s
def close(self):
sgmllib.SGMLParser.close(self)
self.pbr()
self.o('', 0, 'end')
return self.outtext
def handle_charref(self, c):
self.o(charref(c))
def handle_entityref(self, c):
self.o(entityref(c))
def unknown_starttag(self, tag, attrs):
self.handle_tag(tag, attrs, 1)
def unknown_endtag(self, tag):
self.handle_tag(tag, None, 0)
def previousIndex(self, attrs):
""" returns the index of certain set of attributes (of a link) in the
self.a list
If the set of attributes is not found, returns None
"""
if not attrs.has_key('href'): return None
i = -1
for a in self.a:
i += 1
match = 0
if a.has_key('href') and a['href'] == attrs['href']:
if a.has_key('title') or attrs.has_key('title'):
if (a.has_key('title') and attrs.has_key('title') and
a['title'] == attrs['title']):
match = True
else:
match = True
if match: return i
def handle_tag(self, tag, attrs, start):
attrs = fixattrs(attrs)
if hn(tag):
self.p()
if start: self.o(hn(tag)*"#" + ' ')
if tag in ['p', 'div']: self.p()
if tag == "br" and start: self.o(" \n")
if tag == "hr" and start:
self.p()
self.o("* * *")
self.p()
if tag in ["head", "style", 'script']:
if start: self.quiet += 1
else: self.quiet -= 1
if tag in ["body"]:
self.quiet = 0 # sites like 9rules.com never close <head>
if tag == "blockquote":
if start:
self.p(); self.o('> ', 0, 1); self.start = 1
self.blockquote += 1
else:
self.blockquote -= 1
self.p()
if tag in ['em', 'i', 'u']: self.o("_")
if tag in ['strong', 'b']: self.o("**")
if tag == "code" and not self.pre: self.o('`') #TODO: `` `this` ``
if tag == "abbr":
if start:
attrsD = {}
for (x, y) in attrs: attrsD[x] = y
attrs = attrsD
self.abbr_title = None
self.abbr_data = ''
if attrs.has_key('title'):
self.abbr_title = attrs['title']
else:
if self.abbr_title != None:
self.abbr_list[self.abbr_data] = self.abbr_title
self.abbr_title = None
self.abbr_data = ''
if tag == "a":
if start:
attrsD = {}
for (x, y) in attrs: attrsD[x] = y
attrs = attrsD
if attrs.has_key('href') and not (SKIP_INTERNAL_LINKS and attrs['href'].startswith('#')):
self.astack.append(attrs)
self.o("[")
else:
self.astack.append(None)
else:
if self.astack:
a = self.astack.pop()
if a:
i = self.previousIndex(a)
if i is not None:
a = self.a[i]
else:
self.acount += 1
a['count'] = self.acount
a['outcount'] = self.outcount
self.a.append(a)
self.o("][" + `a['count']` + "]")
if tag == "img" and start:
attrsD = {}
for (x, y) in attrs: attrsD[x] = y
attrs = attrsD
if attrs.has_key('src'):
attrs['href'] = attrs['src']
alt = attrs.get('alt', '')
i = self.previousIndex(attrs)
if i is not None:
attrs = self.a[i]
else:
self.acount += 1
attrs['count'] = self.acount
attrs['outcount'] = self.outcount
self.a.append(attrs)
self.o("![")
self.o(alt)
self.o("]["+`attrs['count']`+"]")
if tag == 'dl' and start: self.p()
if tag == 'dt' and not start: self.pbr()
if tag == 'dd' and start: self.o(' ')
if tag == 'dd' and not start: self.pbr()
if tag in ["ol", "ul"]:
if start:
self.list.append({'name':tag, 'num':0})
else:
if self.list: self.list.pop()
self.p()
if tag == 'li':
if start:
self.pbr()
if self.list: li = self.list[-1]
else: li = {'name':'ul', 'num':0}
self.o(" "*len(self.list)) #TODO: line up <ol><li>s > 9 correctly.
if li['name'] == "ul": self.o("* ")
elif li['name'] == "ol":
li['num'] += 1
self.o(`li['num']`+". ")
self.start = 1
else:
self.pbr()
if tag in ["table", "tr"] and start: self.p()
if tag == 'td': self.pbr()
if tag == "pre":
if start:
self.startpre = 1
self.pre = 1
else:
self.pre = 0
self.p()
def pbr(self):
if self.p_p == 0: self.p_p = 1
def p(self): self.p_p = 2
def o(self, data, puredata=0, force=0):
if self.abbr_data is not None: self.abbr_data += data
if not self.quiet:
if puredata and not self.pre:
data = re.sub('\s+', ' ', data)
if data and data[0] == ' ':
self.space = 1
data = data[1:]
if not data and not force: return
if self.startpre:
#self.out(" :") #TODO: not output when already one there
self.startpre = 0
bq = (">" * self.blockquote)
if not (force and data and data[0] == ">") and self.blockquote: bq += " "
if self.pre:
bq += " "
data = data.replace("\n", "\n"+bq)
if self.start:
self.space = 0
self.p_p = 0
self.start = 0
if force == 'end':
# It's the end.
self.p_p = 0
self.out("\n")
self.space = 0
if self.p_p:
self.out(('\n'+bq)*self.p_p)
self.space = 0
if self.space:
if not self.lastWasNL: self.out(' ')
self.space = 0
if self.a and ((self.p_p == 2 and LINKS_EACH_PARAGRAPH) or force == "end"):
if force == "end": self.out("\n")
newa = []
for link in self.a:
if self.outcount > link['outcount']:
self.out(" ["+`link['count']`+"]: " + link['href']) #TODO: base href
if link.has_key('title'): self.out(" ("+link['title']+")")
self.out("\n")
else:
newa.append(link)
if self.a != newa: self.out("\n") # Don't need an extra line when nothing was done.
self.a = newa
if self.abbr_list and force == "end":
for abbr, definition in self.abbr_list.items():
self.out(" *[" + abbr + "]: " + definition + "\n")
self.p_p = 0
self.out(data)
self.lastWasNL = data and data[-1] == '\n'
self.outcount += 1
def handle_data(self, data):
if r'\/script>' in data: self.quiet -= 1
self.o(data, 1)
def unknown_decl(self, data): pass
def wrapwrite(text): sys.stdout.write(text.encode('utf8'))
def html2text_file(html, out=wrapwrite):
h = _html2text(out)
h.feed(html)
h.feed("")
return h.close()
def html2text(html):
return optwrap(html2text_file(html, None))
if __name__ == "__main__":
if sys.argv[1:]:
arg = sys.argv[1]
if arg.startswith('http://'):
j = urllib.urlopen(arg)
try:
from feedparser import _getCharacterEncoding as enc
except ImportError:
enc = lambda x, y: ('utf-8', 1)
text = j.read()
encoding = enc(j.headers, text)[0]
if encoding == 'us-ascii': encoding = 'utf-8'
data = text.decode(encoding)
else:
encoding = 'utf8'
if len(sys.argv) > 2:
encoding = sys.argv[2]
data = open(arg, 'r').read().decode(encoding)
else:
data = sys.stdin.read().decode('utf8')
wrapwrite(html2text(data))
Raw
openanything.py
'''OpenAnything: a kind and thoughtful library for HTTP web services
This program is part of 'Dive Into Python', a free Python book for
experienced programmers. Visit http://diveintopython.org/ for the
latest version.
'''
__author__ = 'Mark Pilgrim ([email protected])'
__version__ = '$Revision: 1.6 $'[11:-2]
__date__ = '$Date: 2004/04/16 21:16:24 $'
__copyright__ = 'Copyright (c) 2004 Mark Pilgrim'
__license__ = 'Python'
import urllib2, urlparse, gzip
from StringIO import StringIO
USER_AGENT = 'OpenAnything/%s +http://diveintopython.org/http_web_services/' % __version__
class SmartRedirectHandler(urllib2.HTTPRedirectHandler):
def http_error_301(self, req, fp, code, msg, headers):
result = urllib2.HTTPRedirectHandler.http_error_301(
self, req, fp, code, msg, headers)
result.status = code
return result
def http_error_302(self, req, fp, code, msg, headers):
result = urllib2.HTTPRedirectHandler.http_error_302(
self, req, fp, code, msg, headers)
result.status = code
return result
class DefaultErrorHandler(urllib2.HTTPDefaultErrorHandler):
def http_error_default(self, req, fp, code, msg, headers):
result = urllib2.HTTPError(
req.get_full_url(), code, msg, headers, fp)
result.status = code
return result
def openAnything(source, etag=None, lastmodified=None, agent=USER_AGENT):
"""URL, filename, or string --> stream
This function lets you define parsers that take any input source
(URL, pathname to local or network file, or actual data as a string)
and deal with it in a uniform manner. Returned object is guaranteed
to have all the basic stdio read methods (read, readline, readlines).
Just .close() the object when you're done with it.
If the etag argument is supplied, it will be used as the value of an
If-None-Match request header.
If the lastmodified argument is supplied, it must be a formatted
date/time string in GMT (as returned in the Last-Modified header of
a previous request). The formatted date/time will be used
as the value of an If-Modified-Since request header.
If the agent argument is supplied, it will be used as the value of a
User-Agent request header.
"""
if hasattr(source, 'read'):
return source
if source == '-':
return sys.stdin
if urlparse.urlparse(source)[0] == 'http':
# open URL with urllib2
request = urllib2.Request(source)
request.add_header('User-Agent', agent)
if lastmodified:
request.add_header('If-Modified-Since', lastmodified)
if etag:
request.add_header('If-None-Match', etag)
request.add_header('Accept-encoding', 'gzip')
opener = urllib2.build_opener(SmartRedirectHandler(), DefaultErrorHandler())
return opener.open(request)
# try to open with native open function (if source is a filename)
try:
return open(source)
except (IOError, OSError):
pass
# treat source as string
return StringIO(str(source))
def fetch(source, etag=None, lastmodified=None, agent=USER_AGENT):
'''Fetch data and metadata from a URL, file, stream, or string'''
result = {}
f = openAnything(source, etag, lastmodified, agent)
result['data'] = f.read()
if hasattr(f, 'headers'):
# save ETag, if the server sent one
result['etag'] = f.headers.get('ETag')
# save Last-Modified header, if the server sent one
result['lastmodified'] = f.headers.get('Last-Modified')
if f.headers.get('content-encoding') == 'gzip':
# data came back gzip-compressed, decompress it
result['data'] = gzip.GzipFile(fileobj=StringIO(result['data'])).read()
if hasattr(f, 'url'):
result['url'] = f.url
result['status'] = 200
if hasattr(f, 'status'):
result['status'] = f.status
f.close()
return result
Raw
unescape.py
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import html2text
from openanything import openAnything
from fileinput import input
import codecs
import sys
sys.stdout = codecs.getwriter('utf-8')(sys.stdout)
""" replace icky character entities with Unicode goodness """
if __name__ == "__main__":
html2text.UNICODESNOB = 1
for line in input():
line = line.decode('utf-8')
print html2text.unescape(line),
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment