2012-05-13 Rasqal Version 0.9.29 Released
Supports all of SPARQL 1.0 by passing 100% of the approved tests
Added support for XSD date
Made XSD dateTime and XSD date equality and comparison work
Make limit and offset check work properly in a subquery
Correctly handle errors during aggregation expressions evaluation
Fix blank node generation when made in a CONSTRUCT to match SPARQL
Added SPARQL 1.1 draft UUID() and STRUUID() support
Requires Raptor 2.0.7+
RDQL support will be removed in the next release (0.9.30)
Made other fixes and improvements including fixing reported issues:
0000493, 0000494 0000504
Release 2.0.3 - 2012/05/29
--------------------------
Improvements
^^^^^^^^^^^^
* [doc] Added about release procedure.
* Removed restriction that the max number of opened files is 4096.
* [experimental] Added table plugin.
* [doc] Added more descriptions about :doc:`/commmands/select` command.
* [doc] Made execution example copy & paste friendly.
* [windows] Supported build with Visual Studio 2010 Express and
CMake. See :doc:`/install/windows` about details.
* [doc][solaris] Added a document about building on Solaris.
See :doc:`/install/solaris` about details.
* [doc][yum] Updated epel-release package version.
[Reported by IWAI, Masaharu]
* [doc][cenos6] Disabled Repoforge for Munin for CentOS 6.
[Reported by IWAI, Masaharu]
* Started distributing source archive in zip format.
* [munin] groonga_dist Munin plugin supported an object that has
separated files.
* Started using Travis CI.
* [yum] Changed RPM package name that provides yum repository from
groonga-repository to groonga-release to follow RPM package name
convension such as centos-release and fedora-release.
Fixes
^^^^^
* [doc] Fixed a command to update yum repository.
[Suggested by IWAI, Masaharu]
* [deb] Fixed a bug that log_repoen command in logrotate uses wrong protocol.
* Fixed broken ERROR tag in XML response.
[#1363] [GitHub#13] [Patch by IWAI, Masaharu]
* Fixed a bug that grn_ctx isn't fully cleared by grn_ctx_fin().
* Fixed a bug that ``&!`` set operation doesn't work with grouped
expression. [#1372]
* Fixed a bug that a record key registered via index source isn't normalized.
* Minor changes to the documentation
* Strip potentially dangerous ENTITY declarations in encoded feeds
* feedparser will now try to continue parsing despite compression errors
* Fix issue 321 a little more (the initial fix missed a code path)
* Issue 337 (`_parse_date_rfc822()` returns None on single-digit days)
* Issue 343 (add magnet links to the ACCEPTABLE_URI_SCHEMES)
* Issue 344 (handle deflated data with no headers nor checksums)
* Issue 347 (support `itunes:image` elements with a `url` attribute)
* Fix mistakes, typos, and bugs in the unit test code
* Fix crash in Python 2.4 and 2.5 if the feed has a UTF_32 byte order mark
* Replace the RFC822 date parser for more extensibility
* Issue 304 (handle RFC822 dates with timezones like GMT+00:00)
* Issue 309 (itunes:keywords should be split by commas, not whitespace)
* Issue 310 (pubDate should map to `published`, not `updated`)
* Issue 313 (include the compression test files in MANIFEST.in)
* Issue 314 (far-flung RFC822 dates don't throw OverflowError on x64)
* Issue 315 (HTTP server for unit tests runs on 0.0.0.0)
* Issue 321 (malformed URIs can cause ValueError to be thrown)
* Issue 322 (HTTP redirect to HTTP 304 causes SAXParseException)
* Issue 323 (installing chardet causes 11 unit test failures)
* Issue 325 (map `description_detail` to `summary_detail`)
* Issue 326 (Unicode filename causes UnicodeEncodeError if locale is ASCII)
* Issue 327 (handle RFC822 dates with extraneous commas)
* Issue 328 (temporarily map `updated` to `published` due to issue 310)
* Issue 329 (escape backslashes in Windows path in docs/introduction.rst)
* Issue 331 (don't escape backslashes that are in raw strings in the docs)
Due to unicode in one of the PLIST files, the package failed during packaging
due to a libarchive translation error on DragonFly. Apparently Linux suffers
from this as well.
The last update was 9 years ago. Updating to current version fixes the
problem for DragonFly. Linux restriction removed under the assumption that
it also works now.
Due to unicode in one of the PLIST files, the package failed during packaging
due to a libarchive translation error on DragonFly. Apparently Linux suffers
from this as well.
The last update was 9 years ago. Updating to current version fixes the
problem for DragonFly. Linux restriction removed under the assumption that
it also works now.
Due to unicode in one of the PLIST files, the package failed during packaging
due to a libarchive translation error on DragonFly. Apparently Linux suffers
from this as well.
The last update was 9 years ago. Updating to current version fixes the
problem for DragonFly. Linux restriction removed under the assumption that
it also works now.
The FlightCrew files all use DOS line endings.
The patches all use unix line endings.
All source files have to have the line endings converted before applying
a patch to them. The last patch added 2012-01-16 to fix build with Boost
wasn't applied to a converted source file and thus the package failed in
the patch phase.
Three main modules and some supporting program files are contained. For
more details read following POD documentations:
Text::LineFold - Line Folding for Plain Text
Unicode::GCString - String as Sequence of UAX #29 Grapheme Clusters
Unicode::LineBreak - UAX #14 Unicode Line Breaking Algorithm
1.57 17 Dec 2011
[FIXES]
- Test for undef streams instead of truth, to allow streams like '' or '0'
Thanks to Thorsten Schwander for the patch
- Failed to parse PDF 1.5 cross reference streams where a
1024-byte boundary happened to fall between "endstream" and
"endobj" do to a logic error.
Thanks to Thorsten Schwander for the report and fix suggestion
1.56 13 Dec 2011
[FIXES]
- Workaround for corrupt PDFs that have 'n' records in their index that point to byte zero
of the file. Silently treat those as 'f' records. Yet another case of Acrobat supporting
broken PDFs, so the rest of us have to support them too...
Thanks to Mark Hunnibell for a sample PDF that demonstrated the problem
Pkgsrc changes:
---------------
Deal with /usr/bin/env python
Upstream changes:
-----------------
0.9.9 final 120131
- FEATURE: Implemented API for ``MarginRule`` objects inside ``CSSPageRule``, see http://www.w3.org/TR/css3-page/. You can also use e.g. ``CSSPageRule['@top-left']`` to retrieve the MarginRule it it is set etc. All dict like methods should be there. If a margin is set twice or more all properties are merged into a single margin rule. Double set properties are all kept though (see below).
- FEATURE: ``parseStyle()`` has optional parameter ``validate=False`` now too to disable validation (default is always ``True``).
- FEATURE: ``CSSStyleDeclaration.setProperty`` has new option ``replace=True``. if True (DEFAULT) the given property will replace a present property. If False a new property will be added always. The difference to `normalize` is that two or more properties with the same name may be set, useful for e.g. stuff like::
background: red;
background: rgba(255, 0, 0, 0.5);
which defines the same property but only capable UAs use the last property value, older ones use the first value.
+ CHANGE: @rules attribute ``atkeyword`` value is now normalized. The actual keyword (example ``@IMPorT``) is kept and is optionally reserialized but in the example ``atkeyword == '@import'``
- BUGFIX: 'auto' is now an invalid CSSPageRule pagename.
- BUGFIX: Fixed issue for GoogleAppEngine (GAE) which somehow handles codecs differently. ``parseUrl`` should work now.
0.9.8
-----
0.9.8 final 111210
- FEATURE: Feature Request (#4) to be able to disable validation of a stylesheet has been implemented. Add Parameter ``validate=False`` for parsing.
+ BUGFIX: Fixed#5 Unicode escaping inside strings. Thanks to Simon Sapin
+ BUGFIX: The integer is optional in counter-reset and counter-increment, and not only on the first counter. Thanks to Simon Sapin
+ BUGFIX: Fix for unicode replacements by Denis Bilenko, thanks! https://bitbucket.org/cthedot/cssutils/pull-request/1/fix-a-bug-in-regex-which-accidentally
- IMPROVEMENT: ``parseStyle`` moved to CSSParser, thanks to Simon Sapin
0.9.8a3 110727
+ BUGFIX: Fixed validation of ``size`` property (thanks to Simon Sapin)
+ BUGFIX: Fixed Issue #55 (thanks to Simon Sapin): `outline-color` property was missing from validation.
+ BUGFIX: Fixed resolution of encoding detection of a stylesheet which did not use @charset in certain circumstances (mainly when imported sheets use different encoding than importing one which should be quite rare actually).
- FEATURE: Added ``URIValue.absoluteUri`` (thanks to Simon Sapin)
- FEATURE: Issue #53 feature request: Added new Preference option ``cssutils.ser.prefs.indentClosingBrace``. Defines if closing brace of block is indented to match indentation of the block (default) oder match indentation of selector.
- FEATURE: Feature request: Added new Preference option ``cssutils.ser.prefs.omitLeadingZero``. Defines if values between -1 and 1 should omit the 0, like ``.5px``. Minified settings do this, else 0 is kept by default.
+ CHANGE (minor): Some error messages have slightly changed due to a simpler compatibility to Python 3. Problem are any ``u'...'`` texts inside error messages which now are simplified, some without and quotes. Changed are e.g. error messages by ``Property``.
- **IMPROVEMENT**: Python 3 support. At least the unittests run in Python 2.5, 2.6, 2.7, 3.2 and Jython 2.5.1 now. Both encutils (with support by Fredrik Hedman, thanks!) and cssutils (thanks to Jaraco) and the CSS codec (thanks to Walter Dörwald) seem to work with Python 3 (tested on Python 3.2.1 Win64). Tests use Mock instead of MiniMock now as former is available for Python 2.x and 3.x.
- **IMPROVEMENT**: Parsing of longer (and probably invalid) ``font`` or ``font-family`` values was *extremely* slow due to a very complex regex. This has been changed and parsing of specific stylesheets using these values should be much faster now. (``macros[Profiles.CSS_LEVEL_2]['font-family']`` is gone so if you used this in your own validation modules you need to check the source in `profiles.py`.)
- IMPROVEMENT: Fixed Issue #54 (thanks to Simon Sapin): Short hand like `#f80` color value object have correct red, green and blue property values now. Also ``hsl()`` and ``hsla()`` colors report (almost) correct values (due to rounding problems).
- **Source control has moved to bitbucket https://bitbucket.org/cthedot/cssutils**. Older Issues are currently still at Google Code, newer at Bitbucket. Please do not use Google Code for new issue reports anymore!
0.9.8a2 110611
- BUGFIX: Fixed Issue #59 which showed a rather strange problem with longer space separated lists of font-family values being so slow to actually stop parsing.
- BUGFIX/IMPROVEMENT: Fixed Issue #48. ``CSSParser.parseUrl()`` uses the defined fetcher of this parser *for the initial stylesheet* at url too and not just the imported sheets *from* this sheet.
- BUGFIX: Fixed Issue #50 which prevented cssutils parsing the acid2.css file correctly. Problem were selectors starting directly with ``[class]`` (an attribute selector).
+ **API CHANGE (major)**
(Known) named colors are parsed as ColorValue objects now. These are the 16 simple colors (black, white, etc) and `transparent` but not all Extended color keywords yet. Also changed ``ColorValue.type`` to ``Value.COLOR_VALUE``. ColorValue has additional properties ``red, green, blue, alpha`` and ``colorType`` which is one of IDENT, HASH or FUNCTION for now.
+ API CHANGE (minor)
Removed already DEPRECATED ``cssutils.parse`` and ``CSSParser.parse``. Use the more specific functions/methods ``parseFile parseString parseUrl`` instead.
Removed already DEPRECATED ``cssutils.log.setlog`` and ``.setloglevel``. Use ``.setLog`` and ``.setLevel`` instead.
Removed already DEPRECATED ``cssutils.ser.keepUnkownAtRules`` (note the typo). Use ``.keepUnknownAtRules`` instead.
- IMPROVEMENT: Added validation profiles for some properties from `CSS Backgrounds and Borders Module Level 3 <http://www.w3.org/TR/css3-background/>`__, `CSS3 Basic User Interface Module <http://www.w3.org/TR/css3-ui/#resize>`__, `CSS Text Level 3 <http://www.w3.org/TR/css3-text/>`__
mainly `cursor`, `outline`, `resize`, `box-shadow`, `text-shadow`
0.9.8a1 101212
+ **API CHANGE (major)**
replace CSSValue with PropertyValue, Value and other classes.
NEW CLASSES:
:class:`cssutils.css.PropertyValue`
replaces CSSValue and CSSValueList
- is iterable (iterates over all single Value objects which in soruce CSS might be separated by "," "/" or " "
- a comma separated list of IDENT values is no longer handled as a single String (e.g. ``Arial, sans-serif``)
:class:`cssutils.css.Value`
replaces CSSPrimitiveValue with separate ``value`` and ``type`` info (value is typed, so e.g. string for e.g. STRING, IDENT or URI values, int or float) and is base class for more specific values like:
:class:`cssutils.css.URIValue`
replaces CSSPrimitiveValue, additional attribute ``uri``
:class:`cssutils.css.DimensionValue`
replaces CSSPrimitiveValue, additional attribute ``dimension``
:class:`cssutils.css.ColorValue`
replaces CSSPrimitiveValue, additional attribute ``red``, ``green``, ``blue`` and ``alpha``
**TODO: Not yet complete, only rgb, rgba, hsl, hsla and has values use this object and color and alpha information no done yet!**
:class:`cssutils.css.CSSFunction`
replaces CSSPrimitiveValue function, not complete yet
also renamed ``ExpressionValue`` to :class:`cssutils.css.MSValue` with new API
- IMPROVEMENT/CHANGE: Validation of color values is tighter now. Values like ``hsl(1, 2, 3)`` do not validate as it must be ``hsl(1, 2%, 3%)``. This mostly effects HSL/A and RGB/A notation.
- **IMPROVEMENT**: New Value parsing and API accelerate parsing of style declarations which take about 20-30% less time now. Of course this depends on the complexity of your styles.
+ BUGFIX: fixes issue #41, #42, #45, #46
PropertyValue.value returns value without any comments now, else use PropertyValue.cssText
- FEATURE: ``cssutils.replaceUrls()`` accepts as first argument a `cssutils.css.CSSStyleSheet` but now also a
:class:`cssutils.css.CSSStyleDeclaration` object, so may be used like the following which is useful when you work with HTML style attributes::
>>> style = cssutils.parseStyle("background-image: url(1.png), url('2.png')")
>>> cssutils.replaceUrls(style, lambda url: 'prefix/'+url)
>>> print style.cssText
background-image: url(prefix/1.png), url(prefix/2.png)
(I omitted the validation error message as more than one background-image is not yet defined in the cssutils validator but does parse through without problems)
+ CHANGE: explicit `+` of any dimension, percentage of number value is kept now instead of being stripped as if put explicitly in the author SHOULD have meant something ;)
Changes in 1.0.6
* New option :break_lines for the HTML encoder (splits tokens at line
breaks). [GH-15, thanks to Etienne Massip]
* Improved speed of :line_numbers => :inline option for the HTML encoder.
* Fixed wrong HTML file type. (was :page) [GH-16, thanks to Doug Hammond]
* The CSS Scanner now highlights tokens like url(...) as :function instead of
:string. [GH-13, thanks to Joel Holdbrooks]
Release 2.0.2 - 2012/04/29
--------------------------
Improvements
^^^^^^^^^^^^
* [pkg-config] Removed needless MessagePack dependency.
* [rpm][fedora] Supported libedit. [#1325] [Suggested by IWAI, Masaharu]
* [rpm] Supported zlib and LZO. [#1324] [Patch by IWAI, Masaharu]
* [groonga] Improved daemoinzed timing. Groonga server daemonizes
after socket is listened. It means that groonga server is ready when
groonga server is daemonized. [#1326]
* [admin] Supported suggest in groonga administration page.
* [dump] Ignored MeCab tokenizer load error.
* Supported CMake.
* [load] Supported error report when a column value can't be set.
* Supported similar search. ``select --filter "column *S 'TEXT'"``
is the similar search syntax. [#1342]
* [apt][yum] Changed package sign key.
* Supported Ubuntu Precise Pangolin.
* [apt] Added a new groonga-keyring deb package for the groonga's
package sign key.
Fixes
^^^^^
* [deb][rpm] Used ``--bind-address`` option instead of deprecated
``--address`` option.
[#1320] [Patch by IWAI, Masaharu]
* [deb] Renamed groonga-server package's configuration file to
/etc/default/groonga-server from /etc/default/groonga.
* [rpm][fedora] Fixed upgrade condition in %post server.
[GitHub#11] [Patch by Daiki Ueno]
* [rpm] Removed needless change logs. [#1328] [Patch by IWAI, Masaharu]
* [deb][rpm] Added missing curl dependency.
[GitHub#12] [Patch by IWAI, Masaharu]
* [rpm] Removed needleess ruby package dependency from groonga package.
[#1330] [Suggested by IWAI, Masaharu]
* [deb] Added missing default values to groonga-server's
configuration file.
* Fixed a crash bug that is caused when searching with updaging very large
inverted index. [#1329]
* Fixed a bug that strings are compared as characters instead of byte string.
[#1340] [Reported by Shinya Kawaji]
.. _release-2-0-1:
Release 2.0.1 - 2012/03/29
--------------------------
Improvements
^^^^^^^^^^^^
* Supported build in other directory.
[#1298] [Reported by Kazuhiko]
* [solaris] Supported build on Solaris 11. [Reported by Kazuhiko]
* [test] Supported functional test.
* [test] Required ``--with-ruby19`` configure option for testing by Ruby.
* [dump] Supported records dump for a table that has default tokenizer.
* [suggest] Added ``similar_search`` option.
* [groonga] Removed deprecated options, ``-a``, ``--address`` and
``--admin-html-path``.
* Added more return value checks. [GitHub#9] [Reported by Markus Elfring]
* [dat] Supported term extract operation.
* Added `logos <http://groonga.org/logo/>`_ .
* Updated HTML design.
* Renamed ``grntest`` to ``groonga-benchmark``.
* Supported ``autogen.sh`` on CentOS 5.
* [linux] Added ``vm.overcommit_memory`` kernel parameter value
check. [#1289]
* ``grn_snip`` uses ``grn_obj`` mechanism. [#1054]
Fixes
^^^^^
* Fixed a bug that uninstall task doesn't uninstall installed Ruby
scripts. [#1299] [Reported by Kazuhiko]
* Added a missing Gemfile. [#1302] [Reported by Kazuhiko]
* Fixed a bug that some indexed records aren't found. The feature is
only used by mroonga. [#1303]
* [groonga] Appended missing ``-id`` to ``--server-id`` option name.
* Fixed a bug that latin1 normalization may access unexpected memory.
2012-04-28 (1.7.0)
* Add JSON::GenericObject for method access to objects transmitted via JSON.
2012-04-27 (1.6.7)
* Fix possible crash when trying to parse nil value.
* Unicode 6.1: New scripts & blocks; changes to grapheme break & line break
property values; some characters change from symbol to Po or No; etc.
* CLDR 21.0.1: Changes in segmentation data to match Unicode 6.1; new structures
for support of Chinese calendar, for context-dependent capitalization, for
gender of lists of people, for ordinal categories, and for multiple number
systems per locale; deprecation of "commonlyUsed" element in timezone names;
removal of "whole-locale" aliases; major cleanups of timezone names,
delimiter data, abbreviated number data.
* Normalizer2 API additions
* Easier-to-use getInstance() variants; e.g., getNFDInstance()
* Getter for the combining-class value for a code point
* Getter for the raw Decomposition_Mapping
* Pairwise composition
* TimeZone class: (C++) Getter for unknown time zone, (Java) fields for GMT &
unknown zone
* Support for deprecation of the "commonlyUsed" element for CLDR metazones
* DateTimePatternGenerator can now use separate patterns for skeletons that
differ only in MMM vs MMMM or EEE vs EEEE, etc.
* Support for custom DecimalFormatSymbols in RuleBasedNumberFormat
* Format and parse Chinese calendar dates including support for intercalary
months
* Context Transforms for context-dependent capitalization behavior
* APIs for TimeZoneNames and TimeZoneFormat
* Support for new date format pattern "ZZZZZ" for ISO 8601 zone format
* Options for ambiguous local time resolution in Calendar
* Support for ISO 4217 numeric currency code
* 40: Fix ``safe_repr`` function to decode bytestrings with non-ASCII
characters correctly.
* 37: Allow configuring sphinx-apidoc via ``SPHINX_APIDOC_OPTIONS``.
* 34: Restore Python 2.4 compatibility.
* 36: Make the "bibliography to TOC" fix in LaTeX output specific to
the document class.
* 695: When the highlight language "python" is specified explicitly,
do not try to parse the code to recognize non-Python snippets.
* 859: Fix exception under certain circumstances when not finding
appropriate objects to link to.
* 860: Do not crash when encountering invalid doctest examples, just
emit a warning.
* 864: Fix crash with some settings of :confval:`modindex_common_prefix`.
* 862: Fix handling of ``-D`` and ``-A`` options on Python 3.
* 851: Recognize and warn about circular toctrees, instead of running
into recursion errors.
* 853: Restore compatibility with docutils trunk.
* 852: Fix HtmlHelp index entry links again.
* 854: Fix inheritance_diagram raising attribute errors on builtins.
* 832: Fix crashes when putting comments or lone terms in a glossary.
* 834, 818: Fix HTML help language/encoding mapping for all Sphinx
supported languages.
* 844: Fix crashes when dealing with Unicode output in doctest extension.
* 831: Provide ``--project`` flag in setup_command as advertised.
* 875: Fix reading config files under Python 3.
* 876: Fix quickstart test under Python 3.
* 870: Fix spurious KeyErrors when removing documents.
* 892: Fix single-HTML builder misbehaving with the master document in a
subdirectory.
* 873: Fix assertion errors with empty ``only`` directives.
* 816: Fix encoding issues in the Qt help builder.
0.7.7.
This release features the following changes:
- Google Spreadsheet doesn't write the undefined-contents byte at the
end of a NOTE record. Excel doesn't care. Now xlrd doesn't care either.
- Version information is now stored in a .py file, hopefully keeping
py2exe and other freeze tools happy.
(0.7.5, 0.7.6 ...fixes packaging issues in 0.7.4...)
0.7.4.
This release features the following changes:
- Fixed a bug where xlrd was silently truncating long text formula results
- Avoid parsing STYLE records when formatting_info=False
- More tolerance of out-of-spec files.
- Minor performance improvements.
