2.6.16: Nov 10 2004:
- general hardening and bug fixing crossing all the API based on new
automated regression testing
- build fix: IPv6 build and test on AIX (Dodji Seketeli)
- bug fixes: problem with XML::Libxml reported by Petr Pajas, encoding
conversion functions return values, UTF-8 bug affecting XPath reported by
Markus Bertheau, catalog problem with NULL entries (William Brack)
- documentation: fix to xmllint man page, some API function descritpion
were updated.
- improvements: DTD validation APIs provided at the Python level (Brent
Hendricks)
Release 1.67.0
* A number of important bug fixes, documented in WhatsNew.
* Added Saxon8 extensions
* Enabled dbfo table-width on entrytbl in FO output
* Added support for role=strong on emphasis in FO output
* Added new FO parameter hyphenate.verbatim that can be used to turn on
"intelligent" wrapping of verbatim environments.
* Replaced all <tt></tt> output with <code></code>
* Changed admon.graphic.width template to a mode so that different
admonitions can have different graphical widths.
* Deprecated the HTML shade.verbatim parameter (use CSS instead)
* Wrapped ToC refentrytitle/refname and refpurpose in span with class values.
This makes it possible to style them using a CSS stylesheet.
* Use strong/em instead of b/i in HTML output
* Added support for converting Emphasis to groff italic and Emphasis role=
'bold' to bold. Controlled by emphasis.propagates.style param, but not
documented yet using litprog system. Will do that next (planning to add
some other parameter-controllable options for hyphenation and handling of
line spacing).
* callout.graphics.number.limit.xml param: Changed the default from 10 to 15.
* verbatim.properties: Added hyphenate=false
* Saxon and Xalan Text.java extensions: Added support for URIResolver() on
insertfile href's
* Added generated RELEASE-NOTES.txt file.
* Added INSTALL file (executable file for generating catalog.xml)
* Removed obsolete tools directory from package
Release 1.66.1
* A number of important bug fixes, documented in WhatsNew.
* Now xml:base attributes that are generated by an XInclude processor are
resolved for image files.
* Rewrote olink templates to support several new features.
o Extended full olink support to FO output.
o Add support for xrefstyle attribute in olinks.
o New parameters to support new olink features: insert.olink.page.number,
insert.olink.pdf.frag, olink.debug, olink.lang.fallback.sequence,
olink.properties, prefer.internal.olink. See the reference page for
each parameter for more information.
* Added index.on.type parameter for new type attribute introduced in DocBook
4.3 for indexterms and index. This allows you to create multiple indices
containing different categories of entries. For users of 4.2 and earlier,
you can use the new parameter index.on.role instead.
* Added new section.autolabel.max.depth parameter to turn off section
numbering below a certain depth. This permits you to number major section
levels and leave minor section levels unnumbered.
* Added footnote.sep.leader.properties attribute set to format the line
separating footnotes in printed output.
* Added parameter img.src.path as a prefix to HTML img src attributes. The
prefix is added to whatever path is already generated by the stylesheet for
each image file.
* Added new attribute-sets informalequation.properties,
informalexample.properties, informalfigure.properties, and
informaltable.properties, so each such element type can be formatted
individually if needed.
* Add component.label.includes.part.label parameter to add any part number to
chapter, appendix and other component labels when the label.from.part
parameter is nonzero. This permits you to distinguish multiple chapters
with the same chapter number in cross references and the TOC.
* Added chunk.separate.lots parameter for HTML output. This parameter lets
you generate separate chunk files for each LOT (list of tables, list of
figures, etc.).
* Added several table features:
o Added table.table.properties attribute set to add properties to the
fo:table element.
o Added placeholder templates named table.cell.properties and
table.cell.block.properties to enable adding properties to any
fo:table-cell or the cell's fo:block, respectively. These templates are
a start for implementing table styles.
* Added new attribute set component.title.properties for easy modifications
of component's title formatting in FO output.
* Added Saxon support for an encoding attribute on the textdata element.
Added new parameter textdata.default.encoding which specifies encoding when
encoding attribute on textdata is missing.
* Template label.this.section now controls whole section label, not only
sub-label which corresponds to particular label. Former behaviour was IMHO
bug as it was not usable.
* Formatting in titleabbrev for TOC and headers is preserved when there are
no hotlink elements in the title. Formerly the title showed only the text
of the title, no font changes or other markup.
* Added intial.page.number template to set the initial-page-number property
for page sequences in print output. Customizing this template lets you
change when page numbering restarts. This is similar to the
format.page.number template that lets you change how the page number
formatting changes in the output.
* Added force.page.count template to set the force-page-count property for
page sequences in print output. This is similar to the format.page.number
template.
* Sort language for localized index sorting in autoidx-ng.xsl is now taken
from document lang, not from system environment.
* Numbering and formatting of normal and ulink footnotes (if turned on) has
been unified. Now ulink footnotes are mixed in with any other footnotes.
* Added support for renderas attribute in section and sect1 et al. This
permits you to render a given section title as if it were a different
level.
* Added support for label attribute in footnote to manually supply the
footnote mark.
* Added support for DocBook 4.3 corpcredit element.
* Added support for a dbfo keep-together PI for formal objects (table,
figure, example, equation, programlisting). That permits a formal object to
be kept together if it is not already, or to be broken if it is very long
and the default keep-together is not appropriate.
* For graphics files, made file extension matching case insensitive, and
updated the list of graphics extensions.
* Allow calloutlist to have block content before the first callout
* Added dbfo-need processing instruction to provide soft page breaks.
* Added implementation of existing but unused default.image.width parameter
for graphics.
* Support DocBook NG tag inline element.
* It appears that XEP now supports Unicode characters in bookmarks. There is
no further need to strip accents from characters.
* Make segmentedlist HTML markup more semantic and available to CSS styles.
* Added user.preroot placeholder template to permit xsl-stylesheet and other
PIs and comments to be output before the HTML root element.
* Non-chunked legalnotice now gets an <a name="id"> element in HTML output so
it can be referenced with xref or link.
* In chunked HTML output, changed link rel="home" to rel="start", and link
rel="previous" to rel="prev", per W3C HTML 4.01 spec.
* Added several patches to htmlhelp from W. Borgert
* Added Bosnian locale file as common/bs.xml.
commands do: adding or removing tabs from a document.
Text::Wrap::wrap() will reformat lines into paragraphs. All it does is
break up long lines, it will not join short lines together.
Text::Fill::fill() will reformat blocks of text into paragraphs. It
uses Text::Wrap::wrap() to do the work
Collection.
This Perl 5 module is a very simple filter for XML parsers. One
common cause of grief (and programmer error) is that XML parsers
aren't required to provide character events in one chunk. They can,
but are not forced to, and most don't. This filter does the trivial
but often-repeated task of putting all characters into a single event.
Actually this is the second import hopefully in the right directory this
time. Sorry for the mess.
changes:
* security fixes on the nanoftp and nanohttp modules
For details see:
http://www.securityfocus.com/archive/1/379383/2004-10-24/2004-10-30/0
* build fixes:
- xmllint detection bug in configure
- building outside the source tree (Thomas Fitzsimmons)
* bug fixes:
- HTML parser on broken ASCII chars in names (William)
- Python paths (Malcolm Tredinnick)
- xmlHasNsProp and default namespace (William)
- saving to python file objects (Malcolm Tredinnick)
- DTD lookup fix (Malcolm)
- save back <group> in catalogs (William)
- tree build fixes (DV and Rob Richards)
- Schemas memory bug
- structured error handler on Python 64bits
- thread local memory deallocation
- memory leak reported by Volker Roth
- xmlValidateDtd in the presence of an internal subset
- entities and _private problem (William)
- xmlBuildRelativeURI error (William).
* improvements:
- better XInclude error reports (William)
- tree debugging module and tests
- convenience functions at the Reader API (Graham Bennett)
- add support for PI in the HTML parser.
Update BUILDLINK_RECOMMENDED to 2.6.15 for the security fix.
Changes since last packaged version (1.10):
* errors in tags error files are reported more correctly and make
the program exit
* more tag options added
* "linenum" tag for specifying the style of line numbers
Package changes:
* set USE_LANGUAGES
* fix HTML installation directory in manpage
which was the BUILDLINK_RECOMMENDED.xmlcatmgr. (Okay'ed by jmmv.)
I had noticed that old xmlcatmgr on my system (using IGNORE_RECOMMENDED)
was not good enough
The xmlcatmgr changes for 2.0alpha2 said that command line usage changed
in some scenarios.
Also remove BUILDLINK_RECOMMENDED.xmlcatmgr because now it was the
same (and is redundant for now).
I am not bumping in PKGREVISIONs because of this. This happened
last January and is too late now and maybe I am the only person to
notice.
in the process. (More information on tech-pkg.)
Bump PKGREVISION and BUILDLINK_DEPENDS of all packages using libtool and
installing .la files.
Bump PKGREVISION (only) of all packages depending directly on the above
via a buildlink3 include.
we were running intltoolize before the configure stage, which replaced all
intltool-*.in files with symlinks to the ones installed by this package.
However, the new version of intltool, 0.31.3, has added new substitution
patterns to the scripts, which must be replaced by configure scripts at
configuration time. The problem is that old configure scripts (pre-0.31.3)
do not know about these patterns, which results in them not being substituted,
thus causing build failures.
So, to fix this issue, replace all intltool-* files _after_ the configure
stage, just like we do with libtool. Additional patters might be added
through the INTLTOOL_OVERRIDE variable, but I doubt this will be needed).
While here, update the dependency to 0.31.3 (has no other effect, as this
package is only used at build time by other packages).
when needed) is correct but... no answers in tech-pkg@ for several days.
* Use absolute paths and allow env var overrides for gettext scripts
-- bug #152020 (Rodney Dawes)
* Allow shell-backticks (and other shell features) in variable names
-- bug #152448 (Danilo Segan)
* Fix typographical error in intltool.m4
-- bug #153630 (Rodney Dawes)
Pod::Simple is a Perl library for parsing text in the Pod ("plain
old documentation") markup language that is typically used for
writing documentation for Perl and for Perl modules.
Pod formatters can use Pod::Simple to parse Pod documents into
produce renderings of them in plain ASCII, in HTML, or in any number
of other formats.
All library names listed by *.la files no longer need to be listed
in the PLIST, e.g., instead of:
lib/libfoo.a
lib/libfoo.la
lib/libfoo.so
lib/libfoo.so.0
lib/libfoo.so.0.1
one simply needs:
lib/libfoo.la
and bsd.pkg.mk will automatically ensure that the additional library
names are listed in the installed package +CONTENTS file.
Also make LIBTOOLIZE_PLIST default to "yes".
and getprogname.
We're using libnbcompat, so we know they exist. (If we were to check again,
then we'd mess up definitions set in the nbcompat headers.)
jlam's last change, since we use libnbcompat. (If we left these checks in,
configure would test for them, find them and set HAVE_GETPROGNAME, subsequently
breaking 'getprogname', as nbcompat.h wouldn't work since HAVE_GETPROGNAME
is then defined.)
- build fix: NUL in c file blocking compilation on Solaris, Windows build
(Igor Zlatkovic)
- fix: key initialization problem (William Brack)
- documentation: fixed missing man page description for --path
Version 0.31.2
* Handle package name extraction better
-- bug #144725 (Daniel Elstner)
* Support for comments before translatable attributes in XML files,
support for translating both tag content and attributes
-- bug #143581. (Danilo Segan)
Version 0.31.1
* Switch to gnome-common for autogen.sh so we can detect automake
versions better -- bug #145027 (Jordi Mallach)
Third stable version of the 2.x branch, released on 2004/09/01.
* Converted to use the GNU Build System (GNU Automake and GNU Autoconf).
* Because of the previous change, the build has been hopefully fixed in
several platforms, specially under Irix 6.5 using the MIPSpro C compiler.
intended transformation: use "rm" to remove an option, "rmdir" to remove
all options containing a path starting with a given directory name, and
"rename" to rename options to something else.
which are the full option names used to set rpath directives for the
linker and the compiler, respectively. In places were we are invoking
the linker, use "${LINKER_RPATH_FLAG} <path>", where the space is
inserted in case the flag is a word, e.g. -rpath. The default values
of *_RPATH_FLAG are set by the compiler/*.mk files, depending on the
compiler that you use. They may be overridden on a ${OPSYS}-specific
basis by setting _OPSYS_LINKER_RPATH_FLAG and _OPSYS_COMPILER_RPATH_FLAG,
respectively. Garbage-collect _OPSYS_RPATH_NAME and _COMPILER_LD_FLAG.
python specific changes:
avoid a warning with 2.3 (William Brack), tab and space mixes (William),
wrapper generator fixes (William), Cygwin support (Gerrit P. Haase), node
wrapper fix (Marc-Antoine Parent), XML Schemas support (Torkel Lyng)
Changes from the previous version include:
- Adding a sed2nbcompat script to ease importing from src HEAD into
pkgsrc.
- Make this package depend on libnbcompat since it needs a regex
implementation and it's also a bootstrap package.
* s/xmalloc/xrealloc/ so that large (greater than _POSIX2_LINE_MAX*2-1)
multi line text can be added.
Changes since last packaged version:
Version 1.10
* Added scanner for log files (also for iptables and apache logs)
thanks to Jan Tatarik
* Fixed some bugs in the perl scanner (thanks to Geir Nilsen)
* configure checks for ios_base
* check for the right ar to use, fixes problems under Sun compilers
(thanks to Paul Townsend)
HRE is a regular expression library for Korean. While most regex
libraries support only per-syllable processing for Korean, HRE
provides an extended grammar to specify phonemes in each syllable.
It supports both KS X 1001 and Unicode charsets.
Release 1.95.8 Fri Jul 23 2004
- Major new feature: suspend/resume. Handlers can now request
that a parse be suspended for later resumption or aborted
altogether. See "Temporarily Stopping Parsing" in the
documentation for more details.
- Some mostly minor bug fixes, but compilation should no
longer generate warnings on most platforms. SF issues
include: 827319, 840173, 846309, 888329, 896188, 923913,
928113, 961698, 985192.
Packages Collection.
The original Wiki web site was intended to have a very simple interface to
edit and to add pages. Its formatting rules are simple and easy to use.
Text::WikiFormat helps in easily translating to other, more complicated
markup languages. It creates HTML by default, but can be extended to
produce valid POD, DocBook, XML, or any other format imaginable.
* Tue Jul 27 2004 Eric S. Raymond <esr@golux.thyrsus.com> - 1.12-1
- Implemented handling of mdoc .Brq macro. Code no longer chokes on
multiple Synopsis headers.
* Mon Jul 26 2004 Eric S. Raymond <esr@snark.thyrsus.com> - 1.11-1
- Speed optimizations. Improved pod2man detection. Close <anchor/> properly.
.UN before .SH or .SS sets the XML ID of the generated section.
Boldfaced lines immediately before tables are interpreted as titles. .UR now
generates <link> for local links. Improved .RE handling that fixes
a couple of edge cases. Multiline table entries are now interpreted as full
blocks, so commands work normally there.
* Tue Jul 6 2004 Eric S. Raymond <esr@snark.thyrsus.com> - 1.10-1
- Enhance to handle lynx dump pages. Fix .Fa interpretation.
Gets 96% of 10862 Fedora Core 2 pages.
distinfo, because the distfiles were updated with some functional
changes but have the same file name and version number. Changes
include updating author's email address, correcting typos, and adding
missing HTML tags.
pkgsrc change:
- Use XML_ENTRIES instead of installing a separate catalog file,
because it adds only one entry.
was updated with some functional changes while having the same file
name and version number. Changes include DTD public identifier
changes, more documentations, and copyright notice.
While here, update my email address.
ok'd by bad@
changes:
(none given for 2.34..)
2.33
- Fixed Tree style (grantm)
- Fixed some non-utf8 stuff in DTDs (patch in XML::DOM tarball)
2.32
- Memory leak fix (Juerd Waalboer).
- Added windows-1252 encoding
- Styles moved to separate .pm files to make loading faster and
ease maintainence
- Don't load IO::Handle unless we really need to
Relevant changes since 2.07:
- added NumericEscape option
- added ValueAttr option (patch from Anton Berezin)
- suppress 'wide character in print' warning (reported by Dawei Lin)
- Added AttrIndent option (patch from Volker Moell)
- Hash keys are now sorted alphabetically by default; enable the
new NoSort option if you don't want this (patch from Volker Moell)
- Fixed bug where disabling array folding broke anonymous array handling
- Fixed bug when unfolding a tied hash
- SuppressEmpty patch from Douglas Wilson
- POD update re XMLin(XMLout($data)) caveats (bug report from Slaven
Rezic)
- fixed hash ordering assumption in test script (reported by Michel
Rodriguez)
- fixed variable expansion not happening in attributes (patch from Paul
Bussé)
Changes in 3.2:
* use cjkcodecs and iconv_codec if available
* always convert feed to UTF-8 before passing to XML parser
* completely revamped logic for determining character encoding and
attempting XML parsing (much faster)
* increased default timeout to 20 seconds
* test for presence of Location header on redirects
* added tests for many alternate character encodings
* support various EBCDIC encodings
* support UTF-16BE and UTF16-LE with or without a BOM
* support UTF-8 with a BOM
* support UTF-32BE and UTF-32LE with or without a BOM
* fixed crashing bug if no XML parsers are available
* added support for "Content-encoding: deflate"
* send blank "Accept-encoding: " header if neither gzip nor zlib
modules are available
Changes in 3.3:
* optimize EBCDIC to ASCII conversion
* fix obscure problem tracking xml:base and xml:lang if element
declares it, child doesn't, first grandchild redeclares it,
and second grandchild doesn't
* refactored date parsing
* defined public registerDateHandler so callers can add support
for additional date formats at runtime
* added support for OnBlog, Nate, MSSQL, Greek, and Hungarian dates (ytrewq1)
* added zopeCompatibilityHack() which turns FeedParserDict into a
regular dictionary, required for Zope compatibility, and also
makes command-line debugging easier because pprint module
formats real dictionaries better than dictionary-like objects
* added NonXMLContentType exception, which is stored in bozo_exception
when a feed is served with a non-XML media type such as
"text/plain"
* respect Content-Language as default language if not xml:lang is present
* cloud dict is now FeedParserDict
* generator dict is now FeedParserDict
* better tracking of xml:lang, including support for xml:lang=""
to unset the current language
* recognize RSS 1.0 feeds even when RSS 1.0 namespace is not the
default namespace
* don't overwrite final status on redirects (scenarios: redirecting
to a URL that returns 304, redirecting to a URL that redirects
to another URL with a different type of redirect)
* add support for HTTP 303 redirects
more recent PHP anymore, so use USE_BUILDIN.iconv test to select
the appropriate iconv parameter value
fixes compilation on NetBSD 1.6.x and other systems without builtin
iconv, and thus PR pkg/26212 and PR pkg/26213
package behaviour unchanged, so no PKGREVISION bump done
Groff
-----
o The argument of the command line option `-I' is now also passed to troff
and grops, specifying a directory to search for files on the command line,
files named in `so' and `psbb' requests, and files named in \X'ps: file'
and \X'ps: import' escapes.
o If option `-V' is used more than once, the commands will be both printed
on standard error and run.
Troff
-----
o Two new read-only, string-valued registers `.m' and `.M' return the
name of the current drawing and background color, respectively.
o New read-only register `.U' which is set to 1 if in safer mode and set
to 0 if in unsafe mode.
o An input encoding file for latin-5 (a.k.a. ISO 8859-9) has been added.
Example use:
groff -Tdvi -mlatin5 my_file > my_file.dvi
Note that some output devices don't support all glyphs of this encoding.
o If the `return' request is called with an argument, it exits twice,
namely the current macro and the macro one level higher. This is
used to define a wrapper macro for `return' in trace.tmac.
o For completeness, two new requests have been added: `dei1' and `ami1'.
They are equivalent to `dei' and `ami', respectively, but the macros
are executed with compatibility mode off (similar to `de1' and `am1').
o New command line option `-I' to specify a directory for files (both
those on the command line and those named in `psbb' requests). This is
also handled by the groff wrapper program.
o Since version 1.19 you can say `.vs 0'. Older versions emit a warning
and convert this to `.vs \n[.V]'.
This hasn't been documented properly. Note that `.vs 0' isn't saved in a
diversion since it doesn't result in vertical motion.
Pic
___
o Dashed and dotted ellipses have been implemented.
Tbl
---
o New key character `x' to make tbl call a user-defined macro on a table
cell. Patch by Heinz-Jürgen Oertel <hj.oertel@surfeu.de>.
Grap2graph
----------
o A new script contributed by Eric S. Raymond <esr@thyrsus.com>. It
converts a grap diagram into a cropped image. Since it uses gs and the
PNM library, virtually all graphics formats are available for output.
[Note that the grap program itself isn't part of the groff package;
see the file MORE.STUFF how to obtain grap.]
Grohtml
-------
o New option `-j' to emit output splitted into multiple files.
Grops
-----
o New command line option `-I' to specify a directory to search for files
on the command line and files named in \X'ps: import' and \X'ps: file'
escapes. This is also handled by the groff wrapper program.
o The default value for the `broken' keyword in the DESC file is now 0.
Grolj4
------
o A new man page `lj4_font(5)' documents how fonts are accessed with
grolj4.
o The built-in fonts for LJ4 and newer PCL 5 devices have been completely
revised, mainly to access as much glyphs as possible. The provided
metric files should be compatible with recent PCL 5 printers also.
Additionally, font description files have been added for the Arial and
Times New Roman family, the MS symbol, and Wingdings fonts.
Afmtodit
--------
o New option `-x' to prevent use of built-in Adobe Glyph List.
Hpftodit
--------
o Completely revised to handle HP TrueType metric files also. See the
hpftodit manual page for more details.
Groffer
-------
o This version is a rewrite of groffer in many parts, but it is kept in
the old single script style.
New options: --text, --mode text, --tty-viewer, --X, --mode X,
--X-viewer, --html, --mode html, --html-view, --apropos-data,
--apropos-devel, --apropos-progs.
New documentation file: README_SH.
Enhancement of the configuration files and the `apropos' handling.
Macro Packages
--------------
o www.tmac: New macro `JOBNAME' to split output into multiple files.
o In mdoc, multiple calls to `.Lb' are now supported in the LIBRARY
section.
Disable the config.status overriding, it causes config.status to be rerun...
FYI an alternative would be to use GNU make!
Changes since last packaged version:
Sed 4.1.1
* preserve permissions of in-place edited files
* yield an error when running -i on terminals or other non regular files
* do not interpret - as stdin when running in in-place editing mode
* fix bug that prevented 's' command modifiers from working
Change list from release notes:
* Synchronized bundled GD library with GD 2.0.23.
* Fixed a bug that prevented compilation of GD extensions against
FreeType 2.1.0-2.1.2.
* Fixed thread safety issue with informix connection id.
* Fixed incorrect resolving of relative paths by glob() in windows.
* Fixed mapping of Greek letters to html entities.
* Fixed a bug that caused an on shutdown crash when using PHP with Apache
2.0.49.
* Fixed a number of crashes inside pgsql, cpdf and gd extensions.
All in all this release fixes over 30 bugs that have been discovered
and resolved since the 4.3.6 release.
Python related changes:
2.6.11: July 5 2004:
- bug fixes: C14N bug serializing namespaces (Aleksey Sanin), testSAX
properly initialize the library (William), empty node set in XPath
(William), xmlSchemas errors (William), invalid charref problem pointed
by Morus Walter, XInclude xml:base generation (William), Relax-NG bug
with div processing (William), XPointer and xml:base problem(William),
Reader and entities, xmllint return code for schemas (William), reader
streaming problem (Steve Ball), DTD serialization problem (William),
libxml.m4 fixes (Mike Hommey), do not provide destructors as methods on
Python classes, xmlReader buffer bug, Python bindings memory interfaces
improvement (with Stéphane Bidoul), Fixed the push parser to be back to
synchronous behaviour.
- improvement: custom per-thread I/O enhancement (Rob Richards), register
namespace in debug shell (Stefano Debenedetti), Python based regression
test for non-Unix users (William), dynamically increase the number of
XPath extension functions in Python and fix a memory leak (Marc-Antoine
Parent and William)
2.6.10: May 17 2004:
- portability: problem with xpath.c on Windows (MSC and Borland), memcmp
vs. strncmp on Solaris, XPath tests on Windows (Mark Vakoc), C++ do not
use "list" as parameter name, make tests work with Python 1.5 (Ed
Davis),
2.6.11: July 5 2004:
- Schemas: a lot of changes and improvements by Kasimier Buchcik for
attributes, namespaces and simple types.
- build fixes: --with-minimum (William Brack), some gcc cleanup
(William), --with-thread-alloc (William)
- portability: Windows binary package change (Igor Zlatkovic), Catalog
path on Windows
- documentation: update to the tutorial (John Fleck), xmllint return code
(John Fleck), man pages (Ville Skytta),
- bug fixes: C14N bug serializing namespaces (Aleksey Sanin), testSAX
properly initialize the library (William), empty node set in XPath
(William), xmlSchemas errors (William), invalid charref problem pointed
by Morus Walter, XInclude xml:base generation (William), Relax-NG bug
with div processing (William), XPointer and xml:base problem(William),
Reader and entities, xmllint return code for schemas (William), reader
streaming problem (Steve Ball), DTD serialization problem (William),
libxml.m4 fixes (Mike Hommey), do not provide destructors as methods on
Python classes, xmlReader buffer bug, Python bindings memory interfaces
improvement (with Stéphane Bidoul), Fixed the push parser to be back to
synchronous behaviour.
- improvement: custom per-thread I/O enhancement (Rob Richards), register
namespace in debug shell (Stefano Debenedetti), Python based regression
test for non-Unix users (William), dynamically increase the number of
XPath extension functions in Python and fix a memory leak (Marc-Antoine
Parent and William)
- performance: hack done with Arjan van de Ven to reduce ELF footprint
and generated code on Linux, plus use gcc runtime profiling to optimize
the code generated in the RPM packages.
use xmlsoft's FTP since GNOME's hasn't got this version
1.1.7: May 17 2004
- build fix: warning about localtime_r on Solaris
- bug fix: UTF8 string tokenize (William Brack), subtle memory corruption,
linefeed after comment at document level (William), disable-output-escaping
problem (William), pattern compilation in deep imported stylesheets
(William), namespace extension prefix bug, libxslt.m4 bug (Edward Rudd),
namespace lookup for attribute, namespaced DOCTYPE name
use xmlsoft's FTP since GNOME's hasn't got this version
2.6.10: May 17 2004
- build fixes: --without-html problems, make check without make all
- portability: problem with xpath.c on Windows (MSC and Borland), memcmp
vs. strncmp on Solaris, XPath tests on Windows (Mark Vakoc), C++
do not use "list" as parameter name, make tests work with Python 1.5
(Ed Davis),
- improvements: made xmlTextReaderMode public, small buffers resizing
(Morten Welinder), add --maxmem option to xmllint, add
xmlPopInputCallback() for Matt Sergeant, refactoring of
serialization escaping, added escaping customization
- bugfixes: xsd:extension (Taihei Goi), assorted regexp bugs
(William Brack), xmlReader end of stream problem, node deregistration with
reader, URI escaping and filemanes, XHTML1 formatting (Nick Wellnhofer),
regexp transition reduction (William), various XSD Schemas fixes (Kasimier
Buchcik), XInclude fallback problem (William), weird problems with DTD
(William), structured error handler callback context (William), reverse
xmlEncodeSpecialChars() behaviour back to escaping '"'
* added and passed tests for converting HTML entities to Unicode
equivalents in illformed feeds (aaronsw)
* added and passed tests for converting character entities to
Unicode equivalents in illformed feeds (aaronsw)
* test for valid parsers when setting XML_AVAILABLE
* make version and encoding available when server returns a 304
* add handlers parameter to pass arbitrary urllib2 handlers (like
digest auth or proxy support)
* add code to parse username/password out of url and send as basic
authentication
* expose downloading-related exceptions in bozo_exception (aaronsw)
* added __contains__ method to FeedParserDict (aaronsw)
* added publisher_detail (aaronsw)
utility to obsolete the standalone xml-i18n-tools package.
Changes since 0.30:
* Usage of [no-xml] option in AC_PROG_INTLTOOL to avoid requiring
XML::Parser Perl module. (Rodney Dawes)
* Support for C#. (Rodney)
* Support for running intltool-update without autoconf, or with
po/Makevars file (as used in gettext). (Danilo Segan)
* New XML merging (compatible with pre-0.27.2 intltool) -- bug
#130802 -- entity decoding and encoding is not done anymore for
XML. (Danilo)
* Multiple-output XML merging now produces simpler files, and sets
"xml:lang" only when translation is used -- bug #46650. (Danilo)
* Support for passing comments to translators from XML and GConf
schema files. (Danilo)
* Many bug fixes (Rodney, Danilo, Callum McKenzie)
Universal Feed Parser is a Python module for downloading and parsing
syndicated feeds. It can handle RSS 0.90, Netscape RSS 0.91, Userland
RSS 0.91, RSS 0.92, RSS 0.93, RSS 0.94, RSS 1.0, RSS 2.0, Atom,
and CDF feeds.
To use Universal Feed Parser, you will need Python 2.1 or later.
Universal Feed Parser is not meant to run standalone; it is a module
for you to use as part of a larger Python program.
Universal Feed Parser is easy to use; the module is self-contained
in a single file, feedparser.py, and it has only one public function,
parse. parse takes a number of arguments, but only one is required,
and it can be a URL, a local filename, or a raw string containing
feed data in any format.
major changes:
ICU 3.0 includes the latest bug fixes, locale/charset updates, and
performance/build/porting enhancements.
- Collation
Collation data is in a separate data tree, allowing for easier
modularization and maintenance.
getFunctionalEquivalent API allows for better caching and UI support.
- Unicode 4.0.1
ICU is updated to the latest version of Unicode standard, which had
significant property changes.
- CLDR 1.1
Updates to CLDR 1.1, with many updates to locale data, and special
emphasis on collation data.
- Formatting
As an aid to migration of traditional C (stdio) and C++ (iostream)
formatting, the POSIX-like input/output library, icuio, is officially
supported.
Significant digits now supported in DecimalFormat, for general use and
%g support.
- RFC822 time zone format support in DateFormat for compatibility.
- Currency formatting/parsing improvements
Allows parsing multiple currencies with one formatter, without knowing the
currency in advance. Much cleaner design allowing extensibility to other
measurement units in the future.
- Regular expressions (C)
The regular expressions framework now features a C API, instead of just C++.
- Locales
Locale canonicalization spec defined and implemented. Provides
interoperability with POSIX and .NET locale IDs, more RFC 3066 support.
- Layout engine
Layout engine now supports using different canonically-equivalent Unicode
forms of the same text: e.g. a + ´ or á. This is especially important for
non-Latin scripts.
- Build Environment
ICU can now build its data library much faster on most platforms.
For a complete list see:
http://oss.software.ibm.com/cvs/icu/~checkout~/icu/readme.html?tag=release-3-0
Package changes:
* remove inclusion of converters/libiconv/buildlink3.mk,
devel/gettext-lib/builtin.mk will take care of that if needed.
Changes since last packaged version (4.0.9):
* // matches the last regular expression even in POSIXLY_CORRECT mode.
* change the way we treat lines which are not terminated by a newline.
Such lines are printed without the terminating newline (as before)
but as soon as more text is sent to the same output stream, the
missing newline is printed, so that the two lines don't concatenate.
The behavior is now independent from POSIXLY_CORRECT because POSIX
actually has undefined behavior in this case, and the new implementation
arguably gives the ``least expected surprise''. Thanks to Stepan
Kasal for the implementation.
* documentation improvements, with updated references to the POSIX.2
specification
* error messages on I/O errors are better, and -i does not leave temporary
files around (e.g. when running ``sed -i'' on a directory).
* escapes are accepted in the y command (for example: y/o/\n/ transforms
o's into newlines)
* -i option tries to set the owner and group to the same as the input file
* `L' command is deprecated and will be removed in sed 4.2.
* line number addresses are processed differently -- this is supposedly
conformant to POSIX and surely more idiot-proof. Line number addresses
are not affected by jumping around them: they are activated and
deactivated exactly where the script says, while previously
5,8b
1,5d
would actually delete lines 1,2,3,4 and 9 (!).
* multibyte characters are taken in consideration to compute the
operands of s and y, provided you set LC_CTYPE correctly. They are
also considered by \l, \L, \u, \U, \E.
* [\n] matches either backslash or 'n' when POSIXLY_CORRECT.
* new option --posix, disables all GNU extensions. POSIXLY_CORRECT only
disables GNU extensions that violate the POSIX standard.
* options -h and -V are not supported anymore, use --help and --version.
* removed documentation for \s and \S which worked incorrectly
* restored correct behavior for \w and \W: match [[:alnum:]_] and
[^[:alnum:]_] (they used to match [[:alpha:]_] and [^[:alpha:]_]
* the special address 0 can only be used in 0,/RE/ or 0~STEP addresses;
other cases give an error (you are hindering portability for no reason
if specifying 0,N and you are giving a dead command if specifying 0
alone).
* when a \ is used to escape the character that would terminate an operand
of the s or y commands, the backslash is removed before the regex is
compiled. This is left undefined by POSIX; this behavior makes `s+x\+++g'
remove occurrences of `x+', consistently with `s/x\///g'. (However, if
you enjoy yourself trying `s*x\***g', sed will use the `x*' regex, and you
won't be able to pass down `x\*' while using * as the delimiter; ideas on
how to simplify the parser in this respect, and/or gain more coherent
semantics, are welcome).
xlhtml is used to convert Microsoft Excel Spreadsheet files into either html
or tab delimitted ASCII. The program can be interfaced with helper scripts
for viewing email attachments. Most use of this program is through the helper
scripts and one would probably rarely resort to using the commandline
interface.
The RSS Abstraction Interface, or RAI (said "ray"), provides an
object-oriented interface to XML::RSS::Parser trees that abstracts
the user from handling namespaces, overlapping and alternate tag
mappings.
It's rather well known that, while popular, the RSS syntax is a
bit of a mess.
RAI provides a single simplified interface that maps one method
call to various overlapping and alternate tags used in RSS feeds.
The interface also abstracts developers from needing to deal with
namespaces. Method names are based on Dublin Core terminology.
XML::RSS::Parser is a lightweight liberal parser of RSS feeds that
is derived from the XML::Parser::LP module the I developed for
mt-rssfeed -- a Movable Type plugin. This parser is "liberal" in
that it does not demand compliance of a specific RSS version and
will attempt to gracefully handle tags it does not expect or
understand. The parser's only requirements is that the file is
well-formed XML and remotely resembles RSS.
A missing type declaration causes compilation failure. The patch
fixes the problem by including the local header file "constant.h"
This closes the referenced PR.
The patch-ac corrects four instances of printf being called with an
uncompilable argument list; the patch follows a recommendation from
Christian Biere <christianbiere@gmx.de>.
The patch-ad corrects a situation where a label at end of block prevents
compilation; the fix adds a null statement as the label's target.
The patch-al fixes a problem where a label appeared at end of block by
supplying a null statement for the label.
The patch-am provides a missing argument list separator (comma) in
four places.
This closes the PR.
dartsdic.cpp:180: error: ISO C++ forbids cast to non-reference type used
as lvalue
The patch corrects the compilation error by rewriting the cast without
changing the meaning of the statement. This closes the PR.
