Easy way to remove HTML entities from an HTML document?
Robert Oschler
document (e.g. -  , &, &apos, etc.)?
If not I'll just write one myself but I figured I'd save myself some time.
Thanks,
--
Robert
Robert Brewer
> Is there a module/function to remove all the HTML entities
> from an HTML document (e.g. -  , &, &apos, etc.)?
Grab cleanhtml.py from the bottom of
http://www.aminus.org/rbre/python/index.html -- you should be able to
quickly rewrite the Plaintext class and just limit it to replacing (or
removing) entities--at least the regex is already written for you.
HTH!
Robert Brewer
MIS
Amor Ministries
fuma...@amor.org
Christopher T King
> Is there a module/function to remove all the HTML entities from an HTML
> document (e.g. -  , &, &apos, etc.)?
htmllib has this capability, but if you're not doing any other HTML
parsing, a regex, coupled with htmllib's helper module, htmlentitydefs,
does nicely:
import re
import htmlentitydefs
def convertentity(m):
if m.group(1)=='#':
try:
return chr(int(m.group(2)))
except ValueError:
return '&#%s;' % m.group(2)
try:
return htmlentitydefs.entitydefs[m.group(2)]
except KeyError:
return '&%s;' % m.group(2)
def converthtml(s):
return re.sub(r'&(#?)(.+?);',convert,s)
converthtml('Some <html> string.') # --> 'Some <html> string.'
Unknown or invalid entities are left in &xxx; format, while also leaving
Unicode entities in &#nnn; format. If you want a Unicode string to be
returned (and Unicode entities interpreted), replace 'chr' with 'unichr',
and 'htmlentitydefs.entitydefs[m.group(2)]' with
'unichr(htmlentitydefs.name2codepoint[m.group(2)])'.
Hope this helps.
Michael Scarlett
check out mark pilgrims site: http://diveintopython.org/html_processing/index.html
Robert Oschler
news:Pine.LNX.4.44.040725...@ccc6.wpi.edu...
>
> htmllib has this capability, but if you're not doing any other HTML
> parsing, a regex, coupled with htmllib's helper module, htmlentitydefs,
> does nicely:
>
> import re
> import htmlentitydefs
>
> def convertentity(m):
> if m.group(1)=='#':
> try:
> return chr(int(m.group(2)))
> except ValueError:
> return '&#%s;' % m.group(2)
> try:
> return htmlentitydefs.entitydefs[m.group(2)]
> except KeyError:
> return '&%s;' % m.group(2)
>
> def converthtml(s):
> return re.sub(r'&(#?)(.+?);',convert,s)
>
> converthtml('Some <html> string.') # --> 'Some <html> string.'
>
> Unknown or invalid entities are left in &xxx; format, while also leaving
> Unicode entities in &#nnn; format. If you want a Unicode string to be
> returned (and Unicode entities interpreted), replace 'chr' with 'unichr',
> and 'htmlentitydefs.entitydefs[m.group(2)]' with
> 'unichr(htmlentitydefs.name2codepoint[m.group(2)])'.
>
> Hope this helps.
>
Chris,
I believe the line that reads:
def converthtml(s):
return re.sub(r'&(#?)(.+?);',convert,s)
Should read:
def converthtml(s):
return re.sub(r'&(#?)(.+?);',convertentity,s)
Once I made that change it worked like a charm. I'm showing the correction
for future Usenet searchers.
So you can pass a function to re.sub() as the replacement patttern? Very
cool, I didn't know that. I think you could spend a year just learning
regular expressions and still miss something.
Thanks,
Robert.
Christopher T King
> I believe the line that reads:
>
> def converthtml(s):
> return re.sub(r'&(#?)(.+?);',convert,s)
>
> Should read:
>
> def converthtml(s):
> return re.sub(r'&(#?)(.+?);',convertentity,s)
Oops, you're right, mea culpa :)
> So you can pass a function to re.sub() as the replacement patttern? Very
> cool, I didn't know that. I think you could spend a year just learning
> regular expressions and still miss something.
That feature is only mentioned briefly in the online docs, and not at all
in sre.sub's docstring. Surprising, since it's indeed a very useful
feature.
Robert Oschler
>
> That feature is only mentioned briefly in the online docs, and not at all
> in sre.sub's docstring. Surprising, since it's indeed a very useful
> feature.
>
Chris,
Speaking of learning cool things by osmosis, do you know of a well commented
source of Python code, perhaps an Open Source project, that I could study to
learn more interesting techniques like the regexp tip you shared? I find
that studying other people's code is the best way to avoid getting in a
programming rut.
Thanks.
--
Robert
Christopher T King
> Speaking of learning cool things by osmosis, do you know of a well commented
> source of Python code, perhaps an Open Source project, that I could study to
> learn more interesting techniques like the regexp tip you shared? I find
> that studying other people's code is the best way to avoid getting in a
> programming rut.
I seem to recall reading about that re.sub trick in something linked from
Pythonware's Daily Python URL (http://www.pythonware.com/daily/). There
are often links there to interesting and useful code snippets from
ActiveState's Python Cookbook and other sources; I'd say start there if
you want to find neat tricks you can do with Python.
I'm not sure of any particularly "well commented" Python projects though
(I've never really looked into that), but you'll probably find some
interesting small projects in the Vaults of Parnassus
(http://www.vex.net/parnassus/).
Robert Oschler
>
> I seem to recall reading about that re.sub trick in something linked from
> Pythonware's Daily Python URL (http://www.pythonware.com/daily/). There
> are often links there to interesting and useful code snippets from
> ActiveState's Python Cookbook and other sources; I'd say start there if
> you want to find neat tricks you can do with Python.
>
> I'm not sure of any particularly "well commented" Python projects though
> (I've never really looked into that), but you'll probably find some
> interesting small projects in the Vaults of Parnassus
> (http://www.vex.net/parnassus/).
>
Thanks Chris and thanks for all your other help.
With your Python skill you should work for Google. Too bad you don't, you'd
be a wealthy man soon (Google IPO). Wish I did. :)
--
Robert
Christopher T King
> With your Python skill you should work for Google. Too bad you don't,
> you'd be a wealthy man soon (Google IPO). Wish I did. :)
Thanks for the compliment. :) To work at Google is my dream job, and I'm
sure that of many others on this list, too (makes me wonder if any Google
employees read this list...).