Based on PR#43384 by Wen Heping.
* Migrated from SWIG to ctypes since 1.5.0:
* Add dependency on devel/py-ctypes for python2.4
* Fix paths for libenchant to exactly load suitable one.
* No need c compiler and buildlink with textproc/enchant.
Changes:
1.6.2:
* Upgraded bundled enchant to v1.6.0.
* Fixed bug in printf() utility function; all input args are now converted
to strings before printing.
1.6.1:
* Fixed loading of enchant DLL on win32 without pkg_resources installed.
* Fixed HTMLChunker to handle unescaped < and > characters that are
clearly not part of a tag.
1.6.0:
* Upgraded to enchant v1.5.0:
* new Broker methods get_param() and set_param() allow
runtime customisation of provider data
* Added the concept of 'chunkers' to enchant.tokenize.get_tokenizer().
These serve split split the text into large chunks of checkable tokens.
* implemented a simple HTMLChunker class
* Moved error classes into 'enchant.errors' for easier importing
* Moved testcases into separate files so they're not loaded by default
* Allowed SpellChecker to use default language if none is specified
* Improved compatibility with Python 3
1.5.3:
* Fixed termination conditions in English tokenization loop.
* Improved unicode detection in English tokenizer.
* Made enchant spellcheck all of its docstrings as part of the
unittest suite.
1.5.2:
* Modify utils.get_resource_filename and utils.win32_data_files for
compatibility with py2exe (which was broken in the move to ctypes).
Thanks to Stephen George for the fix.
1.5.1:
* SpellChecker.add_to_personal renamed to SpellChecker.add and fixed
to use the corresponding Dict method.
1.5.0:
* Migrated from SWIG to ctypes:
* now runs under PyPy!
* also opens possibilities for Jython, IronPython, ...
* Compatibility updates for Python 3.0, mostly around unicode strings
* Dropped compatibility with Python 2.2
Upstream changes:
2009-04-27
* Release 3.14
Removed explicit loading of UNIVERSAL. RJBS.
Reversed the change applied in release 3.09 to fix RT #12239. POD
tag found inside a complex POD tag (e.g., "C<<< I<foo> >>>") is
again parsed as a tag embedded in a tag instead of text and
entities. The previous interpretation of `perldoc perlpod` was
mistaken. (RT #55602 from Christopher J. Madsen).
CaboCha is a Japanese dependency analysis machine based on Support Vector
Machines. It is (89.29%) system that accuracy is the highest as a statistical
Japanese dependency analysis machine as of June, 2001. Moreover, definite
analytical algorithm (Cascaded Chunking Model) that doesn't do back-track is
adopted, and an efficient analysis can be done comparatively.
This package is ruby module for CaboCha.
CaboCha is a Japanese dependency analysis machine based on Support Vector
Machines. It is (89.29%) system that accuracy is the highest as a statistical
Japanese dependency analysis machine as of June, 2001. Moreover, definite
analytical algorithm (Cascaded Chunking Model) that doesn't do back-track is
adopted, and an efficient analysis can be done comparatively.
This package is python module for CaboCha.
CaboCha is a Japanese dependency analysis machine based on Support Vector
Machines. It is (89.29%) system that accuracy is the highest as a statistical
Japanese dependency analysis machine as of June, 2001. Moreover, definite
analytical algorithm (Cascaded Chunking Model) that doesn't do back-track is
adopted, and an efficient analysis can be done comparatively.
This package is perl module for CaboCha.
CaboCha is a Japanese dependency analysis machine based on Support Vector
Machines. It is (89.29%) system that accuracy is the highest as a statistical
Japanese dependency analysis machine as of June, 2001. Moreover, definite
analytical algorithm (Cascaded Chunking Model) that doesn't do back-track is
adopted, and an efficient analysis can be done comparatively.
YamCha is a generic, customizable, and open source text chunker oriented toward
a lot of NLP tasks, such as POS tagging, Named Entity Recognition, base NP
chunking, and Text Chunking. YamCha is using a state-of-the-art machine learning
algorithm called Support Vector Machines (SVMs), first introduced by Vapnik in
1995.
This package is ruby module for YamCha.
YamCha is a generic, customizable, and open source text chunker oriented toward
a lot of NLP tasks, such as POS tagging, Named Entity Recognition, base NP
chunking, and Text Chunking. YamCha is using a state-of-the-art machine learning
algorithm called Support Vector Machines (SVMs), first introduced by Vapnik in
1995.
This package is python module for YamCha.
YamCha is a generic, customizable, and open source text chunker oriented toward
a lot of NLP tasks, such as POS tagging, Named Entity Recognition, base NP
chunking, and Text Chunking. YamCha is using a state-of-the-art machine learning
algorithm called Support Vector Machines (SVMs), first introduced by Vapnik in
1995.
This package is perl module for YamCha.
YamCha is a generic, customizable, and open source text chunker oriented toward
a lot of NLP tasks, such as POS tagging, Named Entity Recognition, base NP
chunking, and Text Chunking. YamCha is using a state-of-the-art machine learning
algorithm called Support Vector Machines (SVMs), first introduced by Vapnik in
1995.
YamCha is exactly the same system which performed the best in the CoNLL2000
Shared Task, Chunking and BaseNP Chunking task.
variety of 7-bit ASCII equivalents including: hexadecimal and decimal
HTML numeric characters references, \u-escapes, standard hexadecimal
and raw hexadecimal.
Packaged by bartosz.kuzma, import from pkgsrc-wip.
2010-05-05 (1.4.3)
* Fixed some test assertions, from Ruby r27587 and r27590, patch by nobu.
* Fixed issue http://github.com/flori/json/issues/#issue/20 reported by
electronicwhisper@github. Thx!
2010-05-05 (1.4.3)
* Fixed some test assertions, from Ruby r27587 and r27590, patch by nobu.
* Fixed issue http://github.com/flori/json/issues/#issue/20 reported by
electronicwhisper@github. Thx!
Gutcheck is a plain-text checking program that specializes in
reporting the problems that spellcheckers don't--errors like
mismatched quotes, misplaced punctuation, unintended blank lines.
It is specifically tuned for checking texts for submission to
Project Gutenberg, though it can be useful elsewhere as well.
2010-04-26 (1.4.2)
* Applied patch from naruse Yui NARUSE <naruse@airemix.com> to make building with
Microsoft Visual C possible again.
* Applied patch from devrandom <c1.github@niftybox.net> in order to allow building of
json_pure if extensiontask is not present.
* Thanks to Dustin Schneider <dustin@stocktwits.com>, who reported a memory
leak, which is fixed in this release.
* Applied 993f261ccb8f911d2ae57e9db48ec7acd0187283 patch from josh@github.
2010-04-25 (1.4.1)
* Fix for a bug reported by Dan DeLeo <dan@kallistec.com>, caused by T_FIXNUM
being different on 32bit/64bit architectures.
2010-04-23 (1.4.0)
* Major speed improvements and building with simplified
directory/file-structure.
* Extension should at least be comapatible with MRI, YARV and Rubinius.
2010-04-07 (1.2.4)
* Triger const_missing callback to make Rails' dynamic class loading work.
2010-03-11 (1.2.3)
* Added a State#[] method which returns an attribute's value in order to
increase duck type compatibility to Hash.
2010-02-27 (1.2.2)
* Made some changes to make the building of the parser/generator compatible
to Rubinius.
2009-11-25 (1.2.1)
* Added :symbolize_names option to Parser, which returns symbols instead of
strings in object names/keys.
2010-04-26 (1.4.2)
* Applied patch from naruse Yui NARUSE <naruse@airemix.com> to make building with
Microsoft Visual C possible again.
* Applied patch from devrandom <c1.github@niftybox.net> in order to allow building of
json_pure if extensiontask is not present.
* Thanks to Dustin Schneider <dustin@stocktwits.com>, who reported a memory
leak, which is fixed in this release.
* Applied 993f261ccb8f911d2ae57e9db48ec7acd0187283 patch from josh@github.
2010-04-25 (1.4.1)
* Fix for a bug reported by Dan DeLeo <dan@kallistec.com>, caused by T_FIXNUM
being different on 32bit/64bit architectures.
2010-04-23 (1.4.0)
* Major speed improvements and building with simplified
directory/file-structure.
* Extension should at least be comapatible with MRI, YARV and Rubinius.
2010-04-07 (1.2.4)
* Triger const_missing callback to make Rails' dynamic class loading work.
2010-03-11 (1.2.3)
* Added a State#[] method which returns an attribute's value in order to
increase duck type compatibility to Hash.
2010-02-27 (1.2.2)
* Made some changes to make the building of the parser/generator compatible
to Rubinius.
2009-11-25 (1.2.1)
* Added :symbolize_names option to Parser, which returns symbols instead of
strings in object names/keys.
2.2.6 (2010-03-02)
==================
Bugs fixed
----------
* Fixed several Python 3 regressions by building with Cython 0.11.3.
2.2.5 (2010-02-28)
==================
Features added
--------------
* Support for running XSLT extension elements on the input root node
(e.g. in a template matching on "/").
Bugs fixed
----------
* Crash in XPath evaluation when reading smart strings from a document
other than the original context document.
* Support recent versions of html5lib by not requiring its
``XHTMLParser`` in ``htmlparser.py`` anymore.
* Manually instantiating the custom element classes in
``lxml.objectify`` could crash.
* Invalid XML text characters were not rejected by the API when they
appeared in unicode strings directly after non-ASCII characters.
* lxml.html.open_http_urllib() did not work in Python 3.
* The functions ``strip_tags()`` and ``strip_elements()`` in
``lxml.etree`` did not remove all occurrences of a tag in all cases.
* Crash in XSLT extension elements when the XSLT context node is not
an element.
Sed 4.2.1
* fix parsing of s/[[[[[[[[[]//
* security contexts are preserved by -i too under SELinux
* temporary files for sed -i are not made group/world-readable until
they are complete
* Fix bug 12567: the ispell sources aren't licensed under the LGPL
* Add a function to get enchant's version (enchant_get_version)
* Disable zemberek plugin by default, as it's known to cause issues/crashes with WebKit
* Fix bug 12472: Win32 DLL dependency not found popup occurs when module has unmet dependencies
* Possibly fix Ubuntu bug 474062
* Fix bug 12409: Registry handle not closed in enchant_get_registry_value_ex
* Fix bug 12406: Leak in _enchant_get_user_home_dirs() on Windows
* Fix bug 12007: Update FSF address
* Fix bug 12305: Zemberek module lists a Turkish dictionary even without Zemberek installed
* Don't assert if passed a null string list
* Fix bug 12350: enchant_pwl_init_with_file truncates pwl file
* Fix a double-free memory corruption bug
* Fix bug 12173: fix some small memory leaks
* Fix bug 12174: mis-acceptence of dictionaries which start with a partial match of the lang id
* Fix bug 12160: enchant 1.5.0 always looks in "lib" dir for plugins
* Fix the build with the MSVC compiler
* Add a --with-system-myspell option
* Package missing compile-resource file
* Compare paths ignoring case sensitivity on windows
Whoosh is a fast, featureful full-text indexing and searching
library implemented in pure Python.
Some of Whoosh's features include:
* Pythonic API.
* Pure-Python. No compilation or binary packages needed, no
mysterious crashes.
* Fielded indexing and search.
* Fast indexing and retrieval -- much faster than any other
pure-Python solution.
* Pluggable scoring algorithm (including BM25F), text analysis,
storage, posting format, etc.
* Powerful query language parsed by pyparsing.
* Pure Python spell-checker (as far as I know, the only one).
Whoosh takes much inspiration (and sometimes translates code) from
other open-source search engines. The fundamental design is similar
to (but does not entirely work like and is not compatible with)
Lucene, but uses KinoSearch's indexing algorithm, some scoring
algorithms from Terrier, and the English morphological variation
generator from Minion.
