Pkgsrc changes:
* Add some casts to match types and format strings, plus
fix value range of toupper() operation.
* Merge patch-ag into the new patch-encoding.c.
* Add comments to existing patches which lacked comments.
Upstream changes to libxml2-2.9.4: May 23 2016
Security:
CVE-2016-3627 Avoid building recursive entities
CVE-2016-1833 Heap-based buffer overread in htmlCurrentChar
CVE-2016-1835 Heap use-after-free in xmlSAX2AttributeNs
CVE-2016-1837 Heap use-after-free in htmlParsePubidLiteral
and htmlParseSystemiteral
CVE-2016-1836 Bug 759398: Heap use-after-free in xmlDictComputeFastKey
CVE-2016-1839 Bug 758605: Heap-based buffer overread in xmlDictAddString
CVE-2016-1838 Bug 758588: Heap-based buffer overread in
xmlParserPrintFileContextInternal
CVE-2016-1840 Bug 757711: heap-buffer-overflow in xmlFAParsePosCharGroup
CVE-2016-4483 Avoid an out of bound access when serializing
malformed strings
CVE-2016-1834 Bug 763071: heap-buffer-overflow in xmlStrncat
CVE-2016-3705 Add missing increments of recursion depth counter to
XML parser.
CVE-2016-1762 Heap-based buffer overread in xmlNextChar
More format string warnings with possible format string vulnerability
Heap-based buffer-underreads due to xmlParseName
Fix some format string warnings with possible format string vulnerability
Unsigned addition may overflow in xmlMallocAtomicLoc()
Other bugfixes:
Detect change of encoding when parsing HTML names
Fix inappropriate fetch of entities content
Correct the usage of LDFLAGS
Revert the use of SAVE_LDFLAGS in configure.ac
libxml2 hardcodes -L/lib in zlib/lzma tests which breaks cross-compiles
Add more debugging info to runtest
Implement "runtest -u" mode
Integer signed/unsigned type mismatch in xmlParserInputGrow()
Integer overflow parsing port number in URI
Fix apibuild for a recently added constructv2.9.4-rc2
Use pkg-config to locate zlib when possible
Use pkg-config to locate ICU when possible
Fix an error with regexp on nullable counted char transition
Fix memory leak with XPath namespace nodes
Fix namespace axis traversal
Add a make rule to rebuild for ASAN
Fix null pointer deref in docs with no root element
Portability to non C99 compliant compilers
dict.h: Move xmlDictPtr definition before includes to allow direct
inclusion.
Fix XSD validation of URIs with ampersands
xmlschemastypes.c: accept endOfDayFrag Times set to "24:00:00" mean
"end of day" and should not cause an error. v2.9.4-rc1
os400: tell about xmllint and xmlcatalog in README400.
os400: properly process SGML add in XMLCATALOG command.
os400: implement CL command XMLCATALOG.
os400: compile and install program xmlcatalog (qshell-only).
xmlcatalog: flush stdout before interactive shell input.
os400: expand tabs in sources, strip trailing blanks.
os400: implement CL command XMLLINT.
os400: compile and install program xmllint (qshell-only).
os400: initscript make_module(): Use options instead of
positional parameters.
xmllint: flush stdout before interactive shell input.
os400: c14n.rpgle: allow *omit for nullable reference parameters.
os400: use like() for double type.
os400: use like() for int type.
os400: use like() for unsigned int type.
os400: use like() for enum types.
Add xz to xml2-config --libs output
Don't recurse into OP_VALUEs in xmlXPathOptimizeExpression
Fix namespace::node() XPath expression
Fix OOB write in xmlXPathEmptyNodeSet
Fix parsing of NCNames in XPath
Fix OOB read with invalid UTF-8 in xmlUTF8Strsize
Do normalize string-based datatype value in RelaxNG facet checking
Fix typo: s{ ec -> cr }cipt
Fix typos: dictio{ nn -> n }ar{y,ies}
Fix typos: PATH_{ SEAPARATOR -> SEPARATOR }
Correct a typo.
Bug 760921: REGRESSION (8eb55d78): doc/examples/io1 test fails after fix
for "xmlSaveUri() incorrectly recomposes URIs with rootless paths"
Bug 760861: REGRESSION (bf9c1dad): Missing results for
test/schemas/regexp-char-ref_[01].xsd
error.c: *input->cur == 0 does not mean no error
Add missing RNG test files
Bug 760190: configure.ac should be able to build --with-icu without
icu-config tool
Bug 760183: REGRESSION (v2.9.3): XML push parser fails with bogus
UTF-8 encoding error when multi-byte character in large CDATA
section is split across buffer
Bug 758572: ASAN crash in make check
Bug 721158: Missing ICU string when doing --version on xmllint
python 3: libxml2.c wrappers create Unicode str already
win32\VC10\config.h and VS 2015
Add autogen.sh to distrib
Add configure maintainer mode
iso-codes 3.68
--------------
Dr. Tobias Quathamer <toddy@debian.org>
Mon, 2 May 2016
[ ISO 3166-1 ]
* Change long name of GM (ISO-3166 change as of 2016-03-31)
[ ISO 3166-2 ]
* Add changes for Lao (ISO 3166-2 change as of 2015-11-27).
Thanks to Cédric Krier for the bug report and patch.
Closes: alioth#315352
[ ISO 3166-2 translations ]
* Ukrainian by Yuri Chornoivan (TP)
* German by Dr. Tobias Quathamer
[ ISO 3166-1 translations ]
* Brazilian Portuguese by Rafael Fontenelle (TP)
* French by Christian Perrier
* German by Dr. Tobias Quathamer
* Thai by Theppitak Karoonboonyanan
[ ISO 639-3 translations ]
* Ukrainian by Yuri Chornoivan (TP)
[ ISO 639-2 translations ]
* Thai by Theppitak Karoonboonyanan
* German by Dr. Tobias Quathamer
[ ISO 15924 translations ]
* German by Dr. Tobias Quathamer
[ ISO 4217 translations ]
* German by Dr. Tobias Quathamer
on malformed input (CVE-2016-0718)
Description: The Expat XML parser mishandles certain kinds of malformed
input documents, resulting in buffer overflows during processing and error
reporting. The overflows can manifest as a segmentation fault or as memory
corruption during a parse operation. The bugs allow for a denial of service
attack in many applications by an unauthenticated attacker, and could
conceivably result in remote code execution.
bump PKGREV
also add an improvement to the fix for CVE-2015-1283 which was part
of the 2.1.1 release -- don't rely on defined behaviour on overflows
of signed integer operations, from upstream git:
f0bec73b01/
pkgsrc change: add a hint how to run the pkg's selftest (not enabled
permanently because this would add a dependency on C++)
pkgsrc changes:
o Take the MAINTAINERship
Changes:
0.32.0
------
o add support for Chapel language
o add support for .factor file types
o add kotlin language file types
o add support for julia file types
o add support for qml files
o add support for typescript files
o add support for rdoc files
o add a cython filetype (#771)
o add support for velocity file types
o add support for rst file types
o add TOML file types support
o add support for crystal files
o default to smart-case when using file match (-g and -G). Fixes#724.
o add support for BitBake source files
o misc improvements and bug fixes
* Fix for issue when inserting read-only images on Windows.
* Added :func:`get_worksheet_by_name()` method to allow the
retrieval of a worksheet from a workbook via its name.
* Fixed issue where internal file creation and modification dates
where in the local timezone instead of UTC.
3.20.1
======
* Updated translations:
fur (Fabio Tomat)
3.20.0
======
* Updated translations:
da (Ask Hjorth Larsen)
oc (Cedric Valmary)
3.19.90
=======
* Move linktrails stubs to common for all source formats to use
* Allow Mallard example to be a formal element with title
* Added html.js.script template for easier site overrides
* Don't force xref attributes on links in Mallard cache
* Updated translations:
et (Piotr Drag)
hi (Piotr Drag)
is (Sveinn i Felli)
km (Piotr Drag)
pt (Piotr Drag)
si (Piotr Drag)
sl (Piotr Drag)
[Changes contributed by Olly Betts]
- makehtmldocs: Fix inter-class links in generated HTML.
- Fix typo in POD documentation.
- Update Xapian website and trac links to use https, which is now
supported, thanks to James Aylett.
documentation:
* Update links to Xapian website and trac to use https, which is now supported,
thanks to James Aylett.
indexers:
* Fix HTML/XML entity decoding to be O(n) not O(n²) - processing HTML/XML with
a lot of entities is now much faster.
templates:
* Remove unused country code to name maps. These were intended as examples,
but they aren't very useful as such, and really just bloat the templates
needlessly.
API:
* PostingSource: Public member variables are now wrapped by methods (mostly
getters and/or setters, depending on whether they should be readable,
writable or both). In 1.3.5, the public members variables have been
deprecated - we've added the replacement methods in 1.2.23 as well to make
it easier for people to migrate over.
chert backend:
* xapian-check now performs additional consistency checks for chert. Reported
by Jean-Francois Dockes and Bob Cargill via xapian-discuss.
documentation:
* Update links to Xapian website and trac to use https, which is now supported,
thanks to James Aylett.
portability:
* On older Linux kernels, rename() of a file within a directory on NFS can
sometimes erroneously fail with EXDEV. This should only happen if you
try to rename a file across filing systems, so workaround this issue by
retrying up to 5 times on EXDEV (which should be plenty to avoid this
bug, and we don't want to risk looping forever). Fixes#698, reported by
Mark Dufour.
pugixml is a light-weight C++ XML processing library. It features:
* DOM-like interface with rich traversal/modification capabilities
* Extremely fast non-validating XML parser which constructs the DOM
tree from an XML file/buffer
* XPath 1.0 implementation for complex data-driven tree queries
* Full Unicode support with Unicode interface variants and automatic
encoding conversions
The library is extremely portable and easy to integrate and use.
Incompatible changes
--------------------
* The default format of `today_fmt` and `html_last_updated_fmt` is back to
strftime format again. Locale Date Markup Language is also supported for
backward compatibility until Sphinx-1.5.
Translations
------------
* Added Welsh translation, thanks to Geraint Palmer.
* Added Greek translation, thanks to Stelios Vitalis.
* Added Esperanto translation, thanks to Dinu Gherman.
* Added Hindi translation, thanks to Purnank H. Ghumalia.
* Added Romanian translation, thanks to Razvan Stefanescu.
Bugs fixed
----------
* C++, added support for ``extern`` and ``thread_local``.
* C++, type declarations are now using the prefixes ``typedef``, ``using``, and ``type``,
depending on the style of declaration.
* 2413: C++, fix crash on duplicate declarations
* 2394: Sphinx crashes when html_last_updated_fmt is invalid
* 2408: dummy builder not available in Makefile and make.bat
* 2412: hyperlink targets are broken in LaTeX builder
* figure directive crashes if non paragraph item is given as caption
* 2418: time formats no longer allowed in today_fmt
* 2395: Sphinx crashes if unicode character in image filename
* 2396: "too many values to unpack" in genindex-single
* 2405: numref link in PDF jumps to the wrong location
* 2414: missing number in PDF hyperlinks to code listings
* 2440: wrong import for gmtime.
This package supports both conventional Roman algarisms (which range
from 1 to 3999) and Milhar Romans, a variation which uses a bar across
the algarism to indicate multiplication by 1_000. For the purposes of
this module, acceptable syntax consists of an underscore suffixed to
the algarism e.g. IV_V = 4_005. The term Milhar apparently derives
from the Portuguese word for "thousands" and the range of this
notation extends the range of Roman numbers to 3999 * 1000 + 3999 =
4_002_999.
1.7.2
Fix clang warning.
1.7.1
GCC-6 compatibility
New precision setting in StreamBuilder
Move-semantics for some ctors
Added details to exception messages
Clarified when exceptions are thrown
Support isinfinite()
Secure memory-wiping allocator (via compilation flag, settable via cmake)
For this, we changed all STL datatypes to macros, but all unchanged by default.
Lots of minor build-related improvements
Notably, cmake support for ccache
More supported systems
gcc-6
musl
MinGW
Cygwin
QNX
VS2010 (for now, but you can always use old versions of JsonCpp)
Note: 1.7.0 was recalled b/c it accidentally included SecureAlloc by default.
1.6.5
Expose Exception classes
Other changes are just comments.
1.6.2
Fix UTF-8 for old (deprecated) Writers
Do not truncate output at embedded zeroes.
This was already fixed the new StreamWriter, which is used by the default stream inserter. This fix will not break anything that wasn't already broken.
1.6.1
Fix compilation error for `snprintf()` for some compilers.
Fix compilation error for snprintf() for some compilers.
1.6
Json::Exception
Added Json::Exception, plus derived types:
Json::RuntimeError
Json::LogicError
Clarified when exceptions are thrown.
If you are trapping std::exception, nothing will change for you.
Fixed deprecation macro. (Thx to @Dani-Hub.)
Fixed compilation error for gcc-4.8. (Thx to @connormanning.)
Fixed CharReader::Factory base. (Thx to Tengiz Sharafiev.) (Technically, this is not a binary-compatible change, but it is a pretty serious bug. So we have deprecated 1.4 and 1.5.)
2016.4.2
=========
----
* Fix#106: encoding by stdin
* Fix#89: Python 3.5 support.
* Fix#113: inplace baseurl substitution for <a> and <img> tags.
* Feature #118: Update the badges to badge.kloud51.com
* Fix#119: new-line after a list is inserted
Upstream changes:
CHANGES IN formatR VERSION 1.3
NEW FEATURES
o `tidy_source()` can deal with multibyte characters that cannot represented
in the system native encoding now (on Windows)
o `usage()` works for functions obtained from `::` or `:::` now, e.g.
`usage(formatR::tidy_source)`
CHANGES IN formatR VERSION 1.2
MAJOR CHANGES
o the minimal required R version is 3.0.2 now
CHANGES IN formatR VERSION 1.1
NEW FEATURES
o added a new argument `output` to usage()
BUG FIXES
o fixedyihui/knitr#918: when code is NULL, parse() will hang (with a question
mark waiting for input)
String::Trim trims whitespace off your strings. chomp trims only
$/ (typically, that's newline), but trim will trim all leading and
trailing whitespace.
This is a Pod::Coverage subclass (actually, a subclass of
Pod::Coverage::CountParents) that allows the POD itself to declare
certain symbol names trusted.
0.62 Mon Feb 22 15:46:06 UTC 2016
- Apply PR/42 @perlpunk++ (fix for detecting filehandles)
0.61 Sat Feb 20 18:04:19 UTC 2016
- Apply PR/37 @perlpunk++ (includes PR/36 @lameventanas++)
0.60 Tue Feb 9 14:36:09 EST 2016
- Apply PR/33. @andygrundman++. Solaris fixes.
Update DEPENDS
Upstream changes:
CHANGES IN markdown VERSION 0.7.7
BUG FIXES
o renderMarkdown() works now even if text = character(0) or ""
o added an `encoding` argument to renderMarkdown() since multi-byte characters
in renderMarkdown() did not work on Windows (thanks, Kohske Takahashi, #63)
o fixed#64: invalid 'n' argument in rpubsUpload() (thanks, Wouter van
Atteveldt)
MAJOR CHANGES
o if renderMarkdown() returns a character vector, it will be marked with the
UTF-8 encoding if it contains multi-byte characters
Changelog:
Common Changes
CLDR 29: For details of the many changes in CLDR, see CLDR 29.
Grapheme/word/line breaking for emoji sequences, based on Unicode 9 proposed rules. See the Unicode emoji break proposal and the Unicode Emoji Technical Report Proposed Update describing the new emoji sequences. (#12081).
Four new Unicode emoji properties (#11802).
DateFormat day period formatting of "noon", "at night", etc. via new pattern characters b & B, and DateTimePatternGenerator support of C for selecting the customary form (#11872).
Except: Formatting of "0:00 midnight" has been disabled because it is confusing except for at the end of an interval.
RelativeDateTimeFormatter: Simpler formatting API (#12072).
More robust CLDR data loading for MeasureFormat (#11986, #12030), RelativeDateTimeFormatter (#12018), and DateIntervalFormat/DateIntervalInfo (#12013).
New simple & fast SimpleFormatter class for a trivial subset of MessageFormat as used in CLDR data, e.g., "{0} {1}" (#10896).
ICU4C Specific Changes
C API support for RelativeDateTimeFormatter (#12072).
Clang annotations for intended switch case fallthroughs, can now compile with -Wimplicit-fallthrough (#12166).
Internal header files can be compiled by themselves, for simpler alternative build scripts (#12141).
This module provides functions to treat half-width and full-width
characters and display correct size of text in one line on terminals and
mobile phones. You can know the visual width of any text and truncate text
by the visual width. Now this module support flagged UTF-8 and tested only
with Japanese.
This module is pure perl version of Text::VisualWidth. This is bit slow,
but it's not require compiler.
Upstream changes:
2016.003 Tue Mar 29 2016
# No new features, version number as canonical numstring.
2016.002_27 Sat Feb 27 2016
! Makefile.PL.sombok
- CPAN RT#112078: Compiling libsombok.a fails due to invalid path for object
files.
3.6.0 (2016-03-17)
==================
* GH#189: Static builds honour FTP proxy configurations when downloading
the external libs. Patch by Youhei Sakurai.
* GH#187: Now supports (only) version 5.x and later of PyPy.
Patch by Armin Rigo.
* GH#186: Soupparser failed to process entities in Python 3.x.
Patch by Duncan Morris.
* GH#185: Rare encoding related `TypeError` on import was fixed.
Patch by Petr Demin.
* GH#181: Direct support for `.rnc` files in `RelaxNG()` if `rnc2rng`
is installed. Patch by Dirkjan Ochtman.
iso-codes 3.67
--------------
Dr. Tobias Quathamer <toddy@debian.org>
Fri, 1 Apr 2016
[ General ]
* The data is now included in JSON files.
* The XML files are deprecated and should not be used for new
projects. However, they are kept in sync with the JSON data
for now.
* The gettext domains have been renamed to better match the
actual ISO number:
- ISO 639-2: Renamed from iso_639 to iso_639-2
- ISO 639-3: Renamed from iso_639_3 to iso_639-3
- ISO 639-5: Renamed from iso_639_5 to iso_639-5
- ISO 3166-1: Renamed from iso_3166 to iso_3166-1
- ISO 3166-2: Renamed from iso_3166_2 to iso_3166-2
All previously used gettext domains are linked to the new
domain names, so that this transition should be smooth for
programs using those domain names.
* A new ISO standard has been included, ISO 3166-3. The gettext
domain is called iso_3166-3. Basically, this is a split of the
old domain iso_3166 into iso_3166-1 and iso_3166-3, because
the old domain iso_3166 contained both standards. However, the
standard ISO 3166-3 was more or less inaccessible.
* The standard ISO 4217 (currency names) now includes only the
currently used currencies. Entries of withdrawn currencies are
no longer included.
* The standard ISO 639-5 now includes only the language families
which are part of the official standard. The previously
included languages were not part of the standard and have been
removed. This reduced the number of language families from
about 1900 to about 110.
* ISO 639-3 has finally been updated from the latest upstream
sources. Closes: #711025
[ ISO 3166-1 translations ]
* Croatian by Božidar Putanec from TP
This helps generating documentation with textprox/py-sphinx, therefore
fixing the build process of some packages using it.
Bump PKGREVISION.
Approved by pkgsrc-pmc@ during freeze.
* Update HOMEPAGE and MASTER_SITES
Changelog:
Version 1.4.1 [2015-09-24]
---------------------------
- Test suite added
- Bugfix: The tarball contains HACKING and README again
- Bugfix: The zsh completion completes options as first
argument correctly
- Bugfix: Exit status is set as advertised
- Bugfix: Spurious null bytes removed from output
- Bugfix: Skipping of some matches in certain conditions fixed.
- Bugfix: Empty matches don't trigger a loop
Version 1.4.0 [2015-08-14]
---------------------------
- PCRE support (by Julius Plenz)
- Fixed string search
- Ability to pass multiple passwords
- Option to change the colon as prefix separator
- Optional warning about PDFs that contain no searchable text
- New option from grep: +--only-matching+
- New option from grep: +--null+
- Bugfix: Correctly print unicode characters
- Installation: New configure flag +--without-libpcre+
- Installation: New configure flag +--disable-doc+ to disable
manpage generation with asciidoc
- Installation: pdfgrep now requires c++11
** Bug
* [CPPXT-105] - PKIX revocation checking calls OpenSSL's X509_verify_cert
in an unsupported way (breaks with OpenSSL 1.0.1p/1.0.2d and later)
Release Notes - XMLTooling - C++ - Version 1.5.5
** Bug
* [CPPXT-102] - BOOST autoconf macros break with gcc5
Release Notes - XMLTooling - C++ - Version 1.5.4
** Bug
* [CPPXT-97] - Segfault in HTTPResponse::setResponseHeader when used to
clear a header
* [CPPXT-99] - Slightly incorrect User-Agent string with libcurl requests
** Task
* [CPPXT-100] - Update Windows build to latest libraries
** Improvement
* [CPPXT-101] - Disable SSLv3 by default in SOAP client
This is part of fixing PR pkg/50354
Changelog:
Release 2.1.1 Sat March 12 2016
Security fixes:
#582: CVE-2015-1283 - Multiple integer overflows in XML_GetBuffer
Bug fixes:
#502: Fix potential null pointer dereference
#520: Symbol XML_SetHashSalt was not exported
Output of "xmlwf -h" was incomplete
Other changes
#503: Document behavior of calling XML_SetHashSalt with salt 0
Minor improvements to man page xmlwf(1)
Improvements to the experimental CMake build system
libtool now invoked with --verbose
* Rails 5.0 support
* Drop support for old Active Record finder params to paginate():
# now unsupported:
User.paginate(:page => 1, :order => 'created_at DESC', :conditions => ...)
# convert to Arel syntax instead:
User.order('created_at DESC').where(...).paginate(:page => 1)
# or, even better:
User.order('created_at DESC').where(...).page(1)
* Add support for pagination with Mongoid
This release simply fixes the bufprintf function to correctly work on Windows
MinGW-w64 so strings are properly written to the buffer and also skips
non-ASCII chars during anchor generation to avoid generating invalid UTF-8
bytes sequences.
* Updated dprecated File.exists? to File.exist?
* Updated README.rdoc 'Apple Computer' -> 'Apple'
* Fixed defect in PData in which exception was thrown when <data/> element
was read from plist.
* guard against not being able to load rdoc tasks
* Fix typo in comment
* Remove 'Computer' from 'Apple Computer' string in plist files ;-)
* Changed sort to sort_by in Plist::Emit.plist_node to allow mixed symbol and
string hash keys.
Below security problem dose not affect to pkgsrc since we do not use bundlerd
libxml2.
=== 1.6.7.2 / 2015-01-20
This version pulls in several upstream patches to the vendored libxml2 and libxslt to address:
CVE-2015-7499
Ubuntu classifies this as "Priority: Low", RedHat classifies this as "Impact: Moderate", and NIST classifies this as "Severity: 5.0 (MEDIUM)".
MITRE record is https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2015-7499
kramdown 1.10.0 released
This release brings the usual bug fixes but also support for the strikethrough
syntax in the GFM parser as well as some enhancements regarding the
specification of language names for syntax highlighting purposes.
Changes
* 4 minor changes:
o Support for the math-engine MathJax-Node was updated to use the new
mathjax-node package (fixes#313, pull request by Tom Thorogood)
o URL query parameters can now be appended to language names specified in
fenced code blocks if the syntax highlighting engine accepts them (fixes
#234)
o Added strikethrough syntax to the GFM parser (fixes#184 and #307; initial
pull request by Diego Galeota, updated by Parker Moore)
o Allow almost all characters in class names that are defined via a special
syntax (fixes#318, requested by cabo)
* 4 bug fixes:
o Fixed a problem where Kramdown::Document.new would only accept the symbol
:input but not the string ¡Æinput¡Ç as valid key (fixes#312, pull request
by Sun Yaozhu)
o Fixed inconsistent behavior: Empty link text is now also allowed for
normal links, not just images (fixes#305, reported by cabo)
o The HTML5 <mark> element is now recognized as span level element (fixes
#298, reported by Niclas Darville)
o Fixed problem where e-mail autolinks containing an underscore character
were not correctly recognized (fixes#293, reported by erikse)
* 3 other fixes:
o Fixed missing package update statement for Travis (by Parker Moore)
o Add some more documentation regarding MathJax (fixes#296, pull request by
Christopher Jefferson)
o Fixed bad link in API documentation (fixes#315, reported by Tom MacWright)
* SQL scanner: Allow $ signs in SQL identifiers [#164, thanks to jasir and Ben
Basson]
* SQL scanner: Fix open strings [#163, thanks to Adam]
* Ruby scanner: Accept number literal suffixes r and i (Ruby 2.1)
* Ruby scanner: Accept quoted hash keys like { "a": boss } (Ruby 2.2)
* Ruby scanner: Accept save navigation operator &. (Ruby 2.3)
* Ruby scanner: Accept squiggly heredoc <<~ (Ruby 2.3)
* Diff scanner: Prevent running out of regexp stack.
* HTML encoder: You can keep tabs intact now by setting tab_width: false.
* Alpha style: Tweaked colors for :function group with :content.
* File structure: One module per file, autoload CodeRay::Version, paths follow
namespace hierarchy.
CHANGES.txt:
0.13.9
#50, fix a space leak
#36, fix the demo examples
#35, make IsString a superclass of StringLike
#33, make flattenTree O(n) instead of O(n^2)
0.13.8
#30, add parse/render functions directly to the Tree module
0.13.7
#32, make sure upper case &#X works in lookupEntity
0.13.6
#28, some named entities require a trailing semicolon (e.g. mid)
0.13.5
#26, rename the test program to test-tagsoup
0.13.4
#24, add isTagComment function
Update the copyright year
0.13.3
Work on GHC 7.9
0.13.2
Remove all package upper bounds
Allow QuickCheck-2.6
Changelog:
* Noteworthy changes in release 2.24 (2016-03-10) [stable]
** Bug fixes
grep -z would match strings it should not. To trigger the bug, you'd
have to use a regular expression including an anchor (^ or $) and a
feature like a range or a backreference, causing grep to forego its DFA
matcher and resort to using re_search. With a multibyte locale, that
matcher could mistakenly match a string containing a newline.
For example, this command:
printf 'a\nb\0' | LC_ALL=en_US.utf-8 grep -z '^[a-b]*b'
would mistakenly match and print all four input bytes. After the fix,
there is no match, as expected.
[bug introduced in grep-2.7]
grep -Pz now diagnoses attempts to use patterns containing ^ and $,
instead of mishandling these patterns. This problem seems to be
inherent to the PCRE API; removing this limitation is on PCRE's
maint/README wish list. Patterns can continue to match literal ^
and $ by escaping them with \ (now needed even inside [...]).
[bug introduced in grep-2.5]
during the build:
---8<---
downloading ICU data library (icudt)
the files will be extracted to: /pkgsrc-obj/textproc/R-stringi/work/.destdir/usr/pkg/lib/R/library/stringi/libs
no icudt in a local repo
trying URL 'http://static.rexamine.com/packages/icudt55l.zip'
Warning in download.file(paste(href, fname, sep = ""), outfname, mode = "wb") :
unable to resolve 'static.rexamine.com'
Error in download.file(paste(href, fname, sep = ""), outfname, mode = "wb"): cannot open URL 'http://static.rexamine.com/packages/icudt55l.zip'
Ansifilter converts text containing ANSI terminal escape codes into
markup (HTML, LaTeX, TeX, RTF) or plaintext (stripping the escape codes).
Such text is produced for instance by the script(1) utility.
Colors, boldface, underline, italic and blink are supported.
Changelog:
2.40:
Major changes since 2.38:
* Use (and require) C++11.
(Murray Cumming, Kjell Ahlstedt, Mikhail Titov)
* SaxParser: Fix "double free or corruption" if a std::exception is thrown
by a handler method.
(Daniel Trebbien, Kjell Ahlstedt)
2.40.0 (unstable):
* Build improvements.
(Kjell Ahlstedt)
2.38.1 (stable):
* Fix the build with C++11 compilers, such as MS Visual C++ 2013 - Implicit
conversions from streams to bool are no longer allowed.)
(Mikhail Titov)
* Build: Disable deprecated API in dependencies if --enable-warnings=fatal
(Kjell Ahlstedt)
* Build: Require libxml2 2.7.7 or later
(Kjell Ahlstedt)
2.38.0 (stable):
This stable version is identical to the unstable version 2.37.2.
API additions since 2.36:
* Add Node::add_child_with_new_ns().
(Kjell Ahlstedt) Bug #737682 (Mathias Lorente)
* Add XsdSchema and XsdValidator classes. Deprecate Schema and SchemaValidator.
Add RelaxNGSchema and RelaxNGValidator.
(Kjell Ahlstedt) Bug #737712 (Michel Stam)
Thanks to Tjalling Hattink, who made initial versions of the RelaxNG classes.
* Parser: Add input operator>>(std::istream& in, Parser& parser).
(Kjell Ahlstedt) Bug #329281 (Pierre Thierry)
* Parser: Add [set|get]_include_default_attributes() and
[set|get]_parser_options().
(Kjell Ahlstedt) Bug #701674 (Bob Nolty)
2.36.0 (unstable):
* Element::set_namespace_declaration():
It is not an error to set the same URI twice.
(Kjell Ahlstedt) Bug #635846, comment 27.
* Require libxml-2.0 >= 2.7.3.
(Kjell Ahlstedt)
* Parser::initialize_context(): Call xmlCtxtUseOptions().
(Kjell Ahlstedt)
2.35.4 (unstable):
* Document:
- Add process_xinclude(), using new XIncludeStart
and XIncludeEnd classes, derived from Node.
(Kjell Ahlsted) Bug #338521.
- Make the Document(xmlDoc*) constructor public.
(Kjell Ahlsted) Bug #668980.
* Parser: Make it thread-safe.
(Kjell Ahlstedt) Bug #681467.
* Schema::set_document(): If the argument 'document' is 0 then
create an empty document, as the documentation says.
(Kjell Ahlstedt)
* Improve the error handling.
(Kjell Ahlstedt) Bug #635846.
* Documentation:
- Add incremental parsing to the SaxParser example program.
- DtdValidation and SchemaValidation: Print all exception information.
- Improve error descriptions.
(Kjell Ahlstedt)
2.35.3 (unstable):
* Node:
- Add eval_to_[boolean|number|string]() methods.
(Kjell Ahlstedt) Bug #316244
- Fix memory problems in import_node().
(Kjell Ahlstedt) Bug #672992
* Build:
- The examples are now built and run during make check.
(Kjell Ahlstedt) Bug #678390
- Define LIBXMLCPP_EXCEPTIONS_ENABLED unconditionally.
(Kjell Ahlstedt)
2.35.2 (unstable):
* Node: Add get_first_child().
(Murray Cumming, Based on a patch by Ilya Murav'jov in bug #648125)
* Build: Remove the --disable-api-exceptions configure option.
(Murray Cumming)
2.35.1 (unstable):
* Handle attributes with default values correctly.
(Kjell Ahlstedt) Bug #669635.
* Improved handling of entity references and processing instructions.
(Kjell Ahlstedt) Bug #669481
* Parser: Throw more detailed error messages.
(Kjell Ahlstedt) Bug #304020 (Virgile Devaux)
* Document: Make the Document(xmlDoc*) constructor protected.
(Murray Cumming) Bug #668980 (A. Pignotti).
