Yuri Baburov
154658798b
Merge pull request #64 from martinth/master
...
Added python 3 support (Supported: python 2.6, 2.7, 3.3, 3.4).
Thanks a lot to @martinth
2015-07-26 14:11:37 +05:00
Marko Horvatic
f0ff9b2425
Move logging.basicConfig to main function
2015-06-24 16:21:04 +02:00
Yuri Baburov
e2bc1ea055
Improved #65 which has given warning, added cssselect lib, bumped to 0.5.1
2015-05-06 14:33:14 +06:00
Mariusz Osiecki
bf9e7404fa
Failure if best_elem is root ( fix #58 )
2015-05-06 09:34:55 +02:00
Martin Thurau
ce7ca26835
Adds compatibility raise_with_traceback method to support different raise syntax
...
Unfortunately the Python 2 `raise` syntax is not supported in Python 3.3 and not all 3.4.x versions so we deal with that by using conditional imports and a compatibility layer.
2015-04-29 23:35:18 +02:00
Martin Thurau
3ac56329e2
Corrects some things were 2to3 did to much.
2015-04-29 19:33:43 +02:00
Martin Thurau
aa4132f57a
Adds Python 3.4 support.
...
Code now supports Python 2.6, 2.7 and 3.4. PYthon 3.3 isn't support
because of some issues with the parser and the difference between old and
new `raise` syntax.
2015-04-29 16:18:21 +02:00
Yuri Baburov
987570bef0
Updated package links for Python 2.7 and Python 3 support
2015-04-27 15:59:31 +06:00
Yuri Baburov
1fac7e685a
Added a feature to allow more images per article (with a test)
2015-04-27 14:35:00 +06:00
Miguel Galves
d04d41b749
Insert text inside iframe for correct output
2015-04-27 14:05:31 +06:00
Miguel Galves
f1759c1404
Allows iframes containing youtube or vimeo videos. People like them
2015-04-27 13:52:01 +06:00
Yuri Baburov
638f73f6a2
Fix for #52 : <input type="hidden"> are not counted any more for "form removal" heuristic.
2014-09-22 15:31:31 +07:00
Yuri Baburov
08658d1d31
Released v 0.3, and uploaded to the pypi.
2013-10-10 02:39:37 +07:00
hush-hush
e2e78e4d55
Make lxml clean tree available for user modifications.
2012-09-17 13:54:08 +02:00
Richard Harding
e9a5cbfe7f
Remove pdb dummy
2012-04-17 11:33:09 -04:00
Richard Harding
f1a79fb8f8
Update to make sure we don't drop the html tag when ditching elements
2012-04-17 11:04:36 -04:00
Richard Harding
46f0302ebc
rename the document_only flag to html_partial
2012-04-17 10:17:14 -04:00
Richard Harding
a46dc14251
Try to pep8 all the things but give up when I got close.
2012-04-16 21:23:19 -04:00
Richard Harding
5a98e2c1b8
Correct appending and allow for document only
...
- Fix the appending of siblings to the correct nested element
- Add a document only flag so that you can get a dom tree you can nest
yourself without html/body tags.
2012-04-16 20:55:13 -04:00
Richard Harding
edccec5d3b
Work on why we have an empty <body/> tag
...
- Seems to come because the sanitizer ends up with two nodes, not one. The
first is an empty body, the second is the article div.
- Fix up the tabs so we can work with the file. Needs lots of pep8 love.
- Implement an initial hack that at least gets it working atm.
- Start to add test cases, sample html files we can test against, etc.
2012-04-16 17:13:24 -04:00
Jan Weiß
3cdc3d67af
Adding comment about oversight in transform_misused_divs_into_paragraphs().
2012-03-24 10:00:07 +01:00
Jan Weiß
960f885edf
Continue early in remove_unlikely_candidates() in case there is neither a class nor an id attribute.
2012-03-24 09:56:08 +01:00
Jan Weiß
6b3961cd30
Fixing gap in node_length coverage.
2012-03-24 09:54:41 +01:00
facundo
bb93ae1e5f
fixed a small issue on the Document score_paragraphs method
2012-02-06 23:05:26 -05:00
Yuri Baburov
11c4d95411
Fixed indentation, encoding issue and README bug. Thanks to Greg Jastrab. Bump version to 0.2.3
2011-07-27 02:05:16 +07:00
Yuri Baburov
61715dca0a
Bump to version 0.2
2011-06-30 12:08:46 +07:00
Yuri Baburov
c2ec1d1c38
Sorted out unicode issues, thanks to Lee Semel.
2011-06-30 11:51:16 +07:00
Yuri Baburov
f55f16baa1
Updated scoring algorithm to match readability.js v1.7.1
2011-06-01 12:16:32 +07:00
Yuri Baburov
96f476181c
Improved title shortener method, and added it to the Document class.
2011-05-11 19:58:27 +07:00
Yuri Baburov
dada82099b
Moved to lxml (based on decruft version); better encoding recognition.
2011-05-03 11:34:29 +07:00
gfxmonk
2b6a2d3db4
removing empty paragraphs is not very useful, and can break some (stupid) websites
2010-05-01 00:08:23 +10:00
gfxmonk
1d862a00c3
fixed bug where only immediate text was being considered for weights, instead of all nested text
2010-05-01 00:07:30 +10:00
gfxmonk
0eacd959a4
failsafe parsing and more logging
2010-04-30 22:34:53 +10:00
gfxmonk
87ad057706
unicode, dammit!
2010-04-26 23:22:54 +10:00
gfxmonk
a224c5b759
minor
2010-04-24 14:24:09 +10:00
gfxmonk
f73b5f05c4
split out into content and summary methods
2010-04-24 00:41:09 +10:00
gfxmonk
c952f421b7
clean up content method and debug
2010-04-23 23:28:51 +10:00
gfxmonk
c0ca60ee26
use a more leniant parser
2010-04-23 20:51:56 +10:00
gfxmonk
ad3d52ade4
initial
2010-04-22 21:55:00 +10:00
