* Add parameter to displaytime to specify that it is a pubdate,
and in html5 mode, use time tag.
* Add placeholder text in search form (in html5 mode only).
* In html5 mode, use all the nice new semantic tags. Care was taken
to not change the id/class named used in the CSS, so only CSS
that refers to tag types needed to be changed.
* Add ACTIONS variable to page.tmpl, which allows plugins to add arbitrary
links to the action bar without modifying the template further.
(COMMENTSLINK and DISCUSSIONLINK could be folded into this, but
are kept separate for now to avoid breaking modified templates.)
* websetup: Only display Setup button on admins' preferences page.
* graphviz: Fix display of preexisting images in preview mode.
* Fixes a bug in skipping of illegal source files introduced in
3.20100427.
specify DIST_SUBDIR. Fix PR pkg/43249 reported by Robert Elz.
The change was installed TYPOlight's version string only but
binary package has changed. So, bump PKGREVISION.
package gets build with "apache-shared-modules suexec ..." as the options.
Bump package revision for the benefit of users which previously compiled
the package these options and don't have the "suexec" module available.
Problem pointed out by Filip Hajny in private e-mail.
* TMPL_INCLUDE re-enabled for templates read from the templatedir.
(But not in-wiki templates.)
* Version dependency on liburi-perl to >= 1.36; previous versions
did not support building urls from utf-8 strings. Closes: #579713.
* Ikiwiki can be configured to generate html5 instead of the default xhtml
1.0. The html5 output mode is experimental, not yet fully standards
compliant, and will be subject to rapid change.
* htmlscrubber: Allow html5 semantic tags: section, nav, article, aside
hgroup, header, footer, figure, figcaption, time, mark
* htmlscrubber: Also allow some other html5 tags: canvas, progress, meter,
ruby, rt, rp, details, summary, datalist.
* htmlscrubber: Round out html5 video support with the preload
attribute and the source tag.
* htmlscrubber: Allow the html5 form attributes: placeholder, autofocus,
min, max, step, required, autocomplete, novalidate, pattern, list,
and form. (Also the form* override attributes for input and buttons.)
* htmlscrubber: Allow additional misc html5 attributes: reversed,
spellcheck, and hidden.
* template: Fix typo.
Version 2.8.3 (2010-05-01)
--------------------------
- Removed the TinyMCE autoresize plugin
- Fixed: function idnaEncode() encoded the URL multiple times (#1829)
- Fixed: insert tags were not removed in the page description (#1856)
- Fixed: the download element was not hidden if there was no downloadable file (#1853)
- Fixed: ampersand characters were not converted to entities in meta files (#1823, #1831)
- Fixed: image paths were converted to absolute URLs during PDF export (#1843)
- Fixed: search index URLs were incorrect when alias usage is disabled (#1864)
- Fixed: automatically deactivate invalid e-mail addresses while sending newsletters
- Fixed: skipFirst had no effect on news lists without pagination (#1872)
- Fixed: the ModuleWizard and TableWizard were dependent on tl_content (#1883)
- Fixed: the XML news/calendar feeds were not updated if modified via Ajax (#1833)
- Fixed: CSV import failed if the Safe Mode Hack was enabled (#1908)
- Fixed a remote code execution vulnerability in the PDF export function
- Fixed some minor issues
1.) Add missing modules "mod_proxy_scgi.so" and "mod_reqtimeout.so"
if the package is built with shared modules enabled.
This fixes PR pkg/43229 by Ryo HAYASAKA.
2.) Get rid of "PLIST.worker" and use "PLIST_VARS" instead.
3.) Use an option group instead of the "APACHE_MPM" configuration variable
to configure the worker model.
4.) Enable the "apache-shared-modules" options by default. This provides
more flexibility and matches the behaviour of a lot of other
platforms e.g. Solaris or Linux distributions like Ubuntu.
Bump the package revision as the binary package will change by default.
NetBSD Packages Collection.
The Perl 5 module Template::Toolkit::Simple is a simple wrapper
around Template Toolkit. It exports a function called tt which
returns a new Template::Toolkit::Simple object. The object supports
method calls for setting all the Template Toolkit options.
Now you can include customised versions of templates in the source
of your wiki. (For example, templates/page.tmpl.) When these templates
are changed, ikiwiki will automatically rebuild pages that use them.
The --getctime switch is renamed to --gettimes, and it also gets the
file modification time. And it's a lot faster (when using git). But
the really important change is, you don't have to remember to use this
switch. Now ikiwiki will do it when it needs to.
At last, the "tagged()" pagespec only matches tags, not regular wikilinks.
If your wiki accidentially relied on the old, buggy behavior, you might
need to change its pagespecs to use "link()".
Many of your wishes have been answered: Now tag pages can automatically be
created when new tags are used. This feature is enabled by default if you
have configured a tagbase. It can be turned on or off using the
`tag_autocreate` setting.
These changes may also affect some users:
* The `title_natural` sort method (as used by the inline directive, etc)
has been moved to the new sortnaturally plugin, which is not enabled
by default since it requires the Sort::Naturally perl module.
* `TMPL_INCLUDE` is no longer supported in any template used by ikiwiki.
It used to be allowed in certian templates, but not in others.
* The `add_templates` option has been removed from the underlay plugin.
If you used this option, you can instead use templates/ subdirectories
inside underlay directories added by the `add_underlays` option.
Due to the above and other changes, all wikis need to be rebuilt on
upgrade to this version. If you listed your wiki in /etc/ikiwiki/wikilist
this will be done automatically when the Debian package is upgraded. Or
use ikiwiki-mass-rebuild to force a rebuild.
* [ Joey Hess ]
* tag: Automatic creation of tag pages can now be enabled using
the tag_autocreate setting. (David Riebenbauer)
* Customised templates can now be included in the source of wikis
(and also in underlays), and dependencies on them are tracked.
* TMPL_INCLUDE is no longer supported in any template.
* underlay: Removed the add_templates option.
* Add template_depends function to plugin API.
* bzr: Fix bzr log parsing to work with bzr 2.0. (liw)
* comments: Fix missing entity encoding in title.
* txt: Add a special case for robots.txt.
* [ Simon McVittie ]
* Add support for link types, and make the the tagged() pagespec only
match tags, not regular links (a bugfix).
* Rebuild wikis on upgrade to this version to get tag link types recorded
correctly.
* Revamp sorting system; allow new sort methods to be added by plugins,
and add a "sortspec" syntax that can combine, reverse, etc sort methods.
* meta: Add `meta(author)`, `meta(title)` etc sortspecs to allow sorting
by metadata.
* meta: Add optional sortas parameter to author and title meta directives.
This can be used to get names sorted by last name without displaying
them last name first.
* sortnaturally: New plugin; the title_natural sort method has moved here.
* meta: store fields consistently unescaped, and escape on use.
(A wiki rebuild is also needed due to this change.)
* [ Joey Hess ]
* Update dependency for git-core to git transition.
* po: Check that translated underlay directories exist before using them
for master language.
* po: Configuring the same language as master and slave confuses processing;
so filter out such a misconfiguration.
* calendar: Add archive_pagespec, which is used by ikiwiki-calendar to
specify which pages to include on the calendar archive pages.
(The pagespec can still also be specified on the ikiwiki-calendar command
line.)
* pagestats: Class parameter can be used to override default class for
custom styling.
* pagestats: Use style=list to get a list of tags, scaled by use like
in a tag cloud. This is useful to put in a sidebar.
* Rework example blog front page.
* CSS and templates for sidebar changed to use a class, not an id.
* sidebar: Now a sidebar directive can be used to override the sidebar
shown on a page.
* Enable calendar and sidebar in auto-blog.setup.
* sidebar: Add global_sidebars setting.
* conditional: Fix bug that forced "all" mode off by default.
* calendarmonth.tmpl: The month calendar is now put in a sidebar.
* calendar: Improved display of arrows.
* Rename --getctime to --gettime. (The old name still works for
backwards compatibility.)
* --gettime now also looks up last modification time.
* Automatically run --gettime the first time ikiwiki is run on
a given srcdir. (Use --no-gettime to disable.)
* Add rcs_getmtime to plugin API; currently only implemented
for git and svn.
* Optimise --gettime for git, so it's appropriately screamingly
fast. (This could be done for most other backends too.)
* However, --gettime for git no longer follows renames. That would
be slow, and whether a renamed wiki page is the same page is really
an iffy thing.
* Use above to fix up timestamps on docwiki, as well as ensure that
timestamps on basewiki files shipped in the deb are sane.
* autoindex: Switch to using %wikistate instead of abusing
$pagestate{index}.
* bzr: Support rcs_getmtime, and fix rcs_getctime implementation
(Jelmer Vernooij)
* Quite a lot of new optimisations, and one major fix to a recent
performance regression.
* Moved javascript files under the ikiwiki/ directory, to avoid cluttering
the top of the web root. This is another things that requires a wiki
rebuild on upgrade to this version.
* Fix removal of rendered files in rebuild mode.
* Add page() PageSpec, which is like glob() but matches only pages,
not other files.
Packaged by Ryo ONODERA, from pkgsrc-wip.
llgal is a HTML gallery generator for images, videos, text and links.
It is based on igal and adds supports for various types of slides
including movies, text, links and subdirectories and several features
such as the addition of headers and footers in generated HTML files or
the generation of slide captions from JFIF, GIF or EXIF comments.
llgal generates nice web-pages through very simple HTML code that does
not require anything like php or javascript.
Collection.
The Perl 5 module Tatsumaki is a toy port of Tornado for Perl using
Plack (with non-blocking extensions) and AnyEvent.
It allows you to write a web application that does a immediate
response with template rendering, IO-bound delayed response (like
fetching third party API or XML feeds), server push streaming and
long-poll Comet in a clean unified API.
Packages Collection.
The Perl 5 module AnyEvent::HTTP is an AnyEvent user, you need to
make sure that you use and run a supported event loop. It implements
a simple, stateless and non-blocking HTTP client. It supports GET,
POST and other request methods, cookies and more, all on a very low
level. It can follow redirects supports proxies and automatically
limits the number of connections to the values specified in the
RFC.
Packages Collection.
The Perl 5 module Net::Server::Coro implements multithreaded server
for the Net::Server architecture, using Coro and Coro::Socket to
make all reads and writes non-blocking. Additionally, it supports
non-blocking SSL negotiation.
Upstream changes:
2010-01-20 Dave Cross <dave@dacross.(none)>
* Build.PL: Removed version number so it's picked up from lib/WWW/Shorten.pm
2009-12-19 Dave Cross <dave@dave.org.uk>
* Build.PL, MANIFEST: Bumped version number. Removed shorl files from
MANIFEST.
* MANIFEST, lib/WWW/Shorten/Shorl.pm, t/shorl.t: Removed support for shorl.pm
(now in WWW::Shorten::Shorl distribution).
2009-12-15 yappo <yappo@shibuya.pl>
* lib/WWW/Shorten/Shorl.pm: shorl.com was change the request method ( POST to
GET )
2009-11-21 Dave Cross <dave@angel.mag-sol.com>
* lib/WWW/Shorten/NotLong.pm, lib/WWW/Shorten/OneShortLink.pm, t/notlong.t,
t/oneshort.t: Removed support for NotLong and OneShortLink (separate
distributions to follow soon). Bumper to version 2.06.
2009-09-18 Dave Cross <dave@dave.org.uk>
* Build.PL: Bump to version 2.05 for release.
* lib/WWW/Shorten.pm, t/newnames.t: Finish removing all references to
SnipURL.pm.
2009-09-08 Dave Cross <dave@dave.org.uk>
* MANIFEST: Removed snipurl.t from MANIFEST.
* Build.PL, MANIFEST, lib/WWW/Shorten/SnipURL.pm, t/snipurl.t: Removed
WWW::Shorten::SnipURL (API changed, separate dist will follow). Bumped
version to 2.04.
2009-04-11 Dave Cross <dave@dave.org.uk>
* lib/WWW/Shorten.pm: Start using $DEFAULT_SERVICE variable.
* MANIFEST: Removed files from MANIFEST
* lib/WWW/Shorten.pm: Documenation improvements.
* lib/WWW/Shorten/BabyURL.pm, lib/WWW/Shorten/EkDk.pm,
lib/WWW/Shorten/ShortLink.pm, lib/WWW/Shorten/SmLnk.pm,
lib/WWW/Shorten/URLjr.pm, lib/WWW/Shorten/V3.pm, lib/WWW/Shorten/qURL.pm,
t/babyurl.t, t/ekdk.t, t/qurl.t, t/shortlink.t, t/smlnk.t, t/urljr.t, t/v3.t:
Removed a number of modules that supported services that are no longer
around.
Upstream changes:
0.9932 Mon Apr 19 15:23:55 JST 2010
- Enable Lint middleware by default in the development env
- Lint middleware now validates $app on startup
- Fixed documentations on middleware and handlers
2.9.2:
* Fixed problem where where logged in users can peek at trashed posts belonging to other authors.
* Fixed other issues
2.9.1:
* Fixed problem where scheduled posts and pingbacks are not processed correctly due to incompatibilities with some hosts
* Fixed other issues
2.9:
User Features
* Trash status for posts, pages, and comments (includes restore and permanent delete)
* Add support for 'include' and 'exclude' to [gallery] (Gallery Shortcode)
* Allow user registration to be enabled by an XMLRPC client
* Add support for sticky posts to the WXR exporter and importer
* 'rel=canonical' for singular pages
* Scroll back to the same location after saving a file in the Plugin and Theme editors
* Correct comments and remove unnecessary echos from the default themes sidebar template file
* Enable the APP (Atom) attachment file download to work correctly
* Support location of category templates based on 'category-slug' as well as 'category-id' (Ticket 10614)
* Support location of tag templates based on 'tag-id' as well as 'tag-slug' (Ticket 10868)
* Support location of page templates based on 'page-slug' and 'page-id'
* Set "Allow my blog to appear in search engines" to checked in installation
* Don't offer to make a category its own parent
* Remove Sphere from search list
* Minify admin CSS
* Show correct max upload filesize error message
* Add 'rel' attribute to next/previous post links
* Make the default and classic themes comment textareas valid XHTML
* Clean up '.button' and '.button[disabled]' CSS classes, add 'spinner' and 'gray-out' buttons after clicking Publish or Update post
* Fix race condition with autosave when clicking Publish immediately after entering post title
* Add Comments for Pages in the WordPress Default theme
* Define '$content_width' for Kubrick
* Better feedback on publishing of future posts and pages
* Display comments in descending date order, consistently
* Add means of automatically repairing tables
* Press This bookmarklet fixes
* Give plugins and themes simple control over the text displayed at the end of an autogenerated Excerpt
* Don't show "Change Permalinks" button when editing the page set as "Front page"
* Image editing
* Retire BunnyTags importer
* Retire Jerome's keywords importer
* Explain that the permalink is temporary for autosave generated permalinks
* Update SimplePie to 1.2
* Eliminate the redundant and confusing comment threading depth of 1
* Easier Embeds with oEmbed support (see Ticket #10337) (oEmbed discovery disabled by default, use plugin to enable it)
* TinyMCE 3.2.7
* Remove rel='tag' on links in Tag Clouds
* Add a title to the Home link output by wp_page_menu()
* Adjust comment moderation keyboard shortcut keys 'd = trash' or delete depending on the screen
* Show "Draft updated" instead of "Post updated" when saving draft
* Show the login form in a popup when autosave hits the login grace period
* Open View/Preview post in a new window from the link in the Saved/Updated message
* Separate fields for 'image alt' and 'image caption' in Media uploader
* Display better information about broken themes when there is no stylesheet
* Improve situation when tables such as wp_options table were 'corrupt' new installation message was offered. Add means of automatically repairing tables
* Export and import custom taxonomies
* Admin copy improvements
* Don't show page templates in the drop down if they are in a subdirectory
* Make codex link open in a new window
* Change 'Remove' link on widgets to 'Delete' because it doesn't just remove it, it deletes the settings for that widget instance.
Development, Themes, Plugins
* Added 'excerpt_more' filter to wp_trim_excerpt() function, which allow developers to change excerpt '[...]' more string (Ticket 10395)
* Add 'smilies_src' filter so plugins can better add smilies
* Canonical redirects for post name queries
* Allow _wp_get_comment_list() to handle custom comment types
* Return an empty array instead of false for get_children() when no children found
* Add some filters so that HTTP requests can be filtered
* Move plugin update notice output to the plugin specific hook
* Limit wp-mail 'blog by email' checks to every 5 minutes
* Make it much easier to filter contact methods from user profiles
* Allow filtering of get_edit_post_link for custom post_type
* 'get_sample_permalink_html' filter
* Enforce activation key to be a string, reject activation keys that are arrays
* Support for new post types
* Respect custom post_type in queries
* Send Retry-After header when in maintenance mode
* Various WP Filesystem related fixes and documentation
* Add constants for ftp connections timeouts
* Increase timeout on cron-based requests when checking for upgrades
* Don't use has_action() before do_action() in http.php
* Speed up jQuery based scripts
* Use the current user as author for autosave
* Show My Posts as default view on the Edit Posts screen for users without 'edit_others_posts' cap
* Ensure that drafts viewed over XMLRPC have a correct gmt date set
* Pass user id to 'get_' the_author_meta filters
* Move _wp_get_user_contactmethods() into the registrations functions file
* Machine parseable db error codes
* Add global JS vars and actions to the media uploader iframe
* Add JSON compat for PHP < 5.2
* Make option_name the primary key for the options table
* Allow a plugin to do a complete takeover of Post by Email
* Logarithmic scale for tag cloud
* Pass Post ID to the 'get_comments_number' filter
* Always filter the url in the media upload form
* Add a 'the_terms' filter
* is_blog_installed() improvements
* Allow force_ssl_admin() to properly accept false as a value
* Pass logged_in cookie to async-upload and filter the cookie scheme in auth_redirect()
* Add more actions around database add/delete/update operations
* phpDoc for wp_"check|set"_post_lock functions
* Use the old strings which are more translator friendly and add a generic default string to aid re-use by plugins adding post_types
* Filter fields through kses upon display and introduce sanitize_user_object() and sanitize_user_field()
* Use null instead of 0 when setting content length
* Include 'hidden' directories in filesystem dirlist by default
* Pass args array to 'wp_list_pages' filter
* Actions for taxonomy updates
* Key should be 'comment_id' not 'post_id' in comments table
* Add get_delete_post_link () to retrieve delete posts link for post
* Add 'separator' parameter to wp_tag_cloud() and wp_generate_tag_cloud() functions (Ticket 10315)
* Added add_comment_meta() family of functions
* Use a post_parent of 0 instead of -1 to indicate unattached posts
* Improve get_page_hierarchy() function
* Deprecate the_content_rss(), add the_content_feed() and get_the_content_feed(). Convert places that called the_content_rss() with an excerpt length to the_excerpt_rss(). Remove the rss_excerpt_length option. Use the_content_feed() where the_content() was previously used in feeds.
* Add 'pad_counts' argument to wp_dropdown_categories()
* Remove codepress
* Remove the php-gettext library
* Canonical post thumbanils
* Add a filter to the_author_posts_link()
* Merge post.js with page.js and slug.js, optimize categories and tags JS, standardize postboxes IDs and JS
* Introduce register_theme_directory() which takes a wp-content-relative path and will additionally scan it for themes. Plugins can use this to add themes without requiring copying by the user
* Add set_user_role action hook
* Allow theme devs to change attrs (like CSS class) of thumbnail images
* Add wp-post-image CSS class to post images
* Allow for plugins to enhance the number of metadata fields captured from plugin and theme headers
* Merge updated pomo code
* Switch to using NOOP_Translations for untranslated sites
* Improve wptexturize performance
* Provide context to the strings in the Plugin and Theme installers to allow for different grammatical gender
* Fixes for theme subdir support
* Introduce wp_kses_post() and wp_kses_data() for filtering unescaped data
* Add 'orderby=comment_count' argument to query_posts()
* Honor Post Type for Sticky Posts
* Allow querying multiple post types
* Introduce add_theme_support(feature) and current_theme_supports(feature) for announcing and checking theme support for various features
* Introduce require_if_theme_supports()
* Add number of Embed related filters
* Add 'IMAGE_EDIT_OVERWRITE' constant to control edited image save or replace, most useful for setups that have dynamic image resizing
* Add load_child_theme_textdomain() to allow child themes to have their own translation files
* Add sidebar descriptions to sidebar settings and widget admin screen
* Make option_id primary. Add uniques for option_name and autoload
* Allow plugins to override the behaviour of load_textdomain() in a variety of flexible ways
* Mark _c() as deprecated. The new _x() function should be used instead.
* Allow plugins to change the redirect on post/page publishing/submitting
* Standardize on 'user_id' instead of 'user_ID' when passing comment data. Accept either 'user_id' or 'user_ID'. Remove 'user_id' global.
* Filter imported comments
* Introducing set_post_image_size(w, h, crop) so themes can register their special size/crop for canonical post images
* Standardize around "post image" instead of "post thumbnail"
* Allow registering post image support per post type
* Return false from is_paged() if on the first page.
* Check MySQL and PHP versions when auto upgrading
* Add required php and mysql versions to version.php
* Hard code required version in update-core.php
PR pkg/42765
Upstream changes:
NEXT Sat Apr 10 23:10:07 CDT 2010
========================================
[FIXED]
Fixed a declaration in the Movable Type example in
WWW::Mechanize::Examples.
Quiet warnings if %ENV has undef values.
$mech->follow_link() no longer dies with an inappropriate error if
the link is not found.
$mech->click_button() now checks to see if a form is selected.
[INCOMPATIBILITIES]
$mech->form_name() and $mech->form_number() no longer throw warnings
if they can't find the form specified. They still return undef,
though.
[DOCUMENTATION]
More additions to the FAQ.
Pkgsrc changes:
- Add more recommended dependencies
- Makefile cleanup
Upstream changes:
0.20 Fri Apr 16 23:05:10 2010
App::Nopaste::Service::Pastie would generate warnings via
Mechanize due to changes in http://pastie.org HTML (avar)
Add App::Nopaste::Service::Codepeek (arcanez)
Upstream changes:
0.9931 Fri Apr 16 23:52:27 PDT 2010
- replace kyoto.jpg test image file with smaller baybridge.jpg to strip down the tarball size
from 2.5MB to 212KB.
Inspired by PR#42892 from Wen Heping (fixes dependency pattern).
=== 2.0.9 ===
March 5, 2010 (revision 964)
- Fixed URL related bug in gdata.calendar.service. You can now edit and delete
against https URLs.
- Removed hard coded Python version from gdata.contacts. Resolves issue 347.
=== 2.0.8 ===
February 26, 2010 (revision 960)
- Added support for resumable uploads with a sample that shows it's use
with Google Docs. Thanks Eric!
- Added the publish element and copy() method to the DocsClient. Thanks Eric!
- Added support for the JSON-C data format recently released for the YouTube
API.
- Fixed import bug when using cryptomath with Python2.4 and lower.
- Modified gdata.gauth to be usable under Python2.3.
- Fixed errors which affect RetrieveAllEmailLists RetrieveEmailLists
RetrieveAllRecipients RetrieveAllNicknames RetrieveNicknames
RetrieveAllUsers in gdata.apps.service. Resolves issue 340 . Thanks
Aprotim!
- Fixed issues with contacts client returning a raw get and not a ContactEntry
and a few auth_tokens are not being passed into the HTTP request. Resolves
issue 332 . Thanks vinces!
- Fixed issue parsing worksheet feed, entry was the wrong type, corrected to
WorksheetEntry. Resolves issue 343. Thanks Igor!
- Fixed typo in gdata.spreadsheets.client (auth_service). Resolves issue 342.
Thanks to Igor once again!
- Fixed bug in Google Apps APIs UpdateSSOSettings.
- The gdata.data.LinkFinder now looks in both the link and feed_link members
for an ACL link. Resolves issue 339.
- Removed client side check on video type in YouTube service. Rely on the
server to provide an error if the format is not supported. Resolves
issue 338.
- Improved documentation for SpreadsheetsService's UpdateRow method to explain
the use case it addresses and recomment alternatives.
- Changed projecthosting to use the host member with relative URLs.
- Added a get_headers function to be used in Python2.2 and 2.3 since
httplib.HTTPResponse does not have getheaders in those versions of Python.
Resolves issue 335.
- Added a client smoke test to check version and auth related class members
which all gdata.x.client's should have.
- Updated the docs service_test.py to run as part of run_all_tests.py.
=== 2.0.7 ===
January 26, 2010 (revision 937)
- Added gdata.calendar_resource.client to support the management of calendar
resources (like conference rooms) for Google Apps domains. Thanks Vic!
- Updated the sites API client to use version 1.1 of the sites API.
- Added support for 2 Legged OAuth to the GDClient and included a sample
app. Thanks Eric!
- Added a revoke_token method to GDClient to invalidate AuthSub or OAuth
tokens.
- Added a get_record method to the Spreadsheets client to fetch a single
record entry.
- Added includeSuspendedUsers parameter to the apps groups API. Thanks Will
Charles!
- Fixed a client login bug present in the Maps API client. Thanks Roman!
- The client_id property in the YouTube service object is now optional.
- Added optional 'feed' parameter to YouTube service's Query classes.
Resolves issue 323.
- Updated tlslite's use of sha1 to use hashlib if using python2.5 or higher.
Resolves issue 290 . Thanks Jay Lee!
- Added a deadline parameter to gdata.alt.appengine to allow the urlfetch
timeout to be modified to as high as 10 seconds.
- Added GetGeneratorFromLinkFinder to GDataService which simplifies pagination
across the entries in a feed. Resolves issue 325 . Thanks Aprotim!
- Removed the default value for the test domain option.
- Moved auth scopes to gdata.gauth from gdata.service.
=== 2.0.6 ===
December 17, 2009 (revision 914)
- updated the Google Analytics Data Export API to v2. Thanks Nick!
- added support for batch operations on contacts and profiles. Thanks Julian!
- added a DocsQuery class for the Documents List v3 API.
- removed now obsolete ACL classes from gdata.docs.data in favor of using
classes in gdata.acl.data. Thanks Eric!
- fixed bug in querying group membership by email. Thanks Will Charles!
- removed import which are not present in Python2.2 from gdata.service to
maintain support for Python2.2.
- added additional debug information to MockHttpClient.
=== 2.0.5 ===
November 24, 2009 (revision 900)
- added new gdata.contacts.client (and data) which uses version 3.0 of the
contacts API. Thanks Vince!
- added new gdata.docs.client which uses version 3.0 of the Documents List
API. Thanks Eric!
- added new v2 modules for the Analytics API (gdata.analytics.client).
Thanks Nick!
- added conditional get using etags on get_entry. Thanks Eric!
- fixed bug in how Calendar redirect headers are handled when using this
library in an App Engine app.
- added new Calendar XML elements to the gdata.calendar module. Thanks
Michael Ballbach!
- added v2 data model classes for finance, notebook, webmastertools, ACLs,
opensearch and youtube.
- fixed issues with contacts.service when editing profiles. Thanks Julian!
- fixed bug in GDClient's get_next method.
- improved usability in Python2.3, though tests do not run the src modules
should be Python2.3 compatible.
=== 2.0.4 ===
October 15, 2009 (revision 864)
- Added support for the project hosting issue tracker API. Thanks Joe LaPenna!
- Added a get-with-retries method to the v1 service for use with the Google Apps
API to automatically retry. Thanks Takashi Matsuo!
- Revised Google Sites client to use text input when creating a page. Thanks Eric!
- Changed demo calendar in the Google Calendar API example. Thanks Trevor!
=== 2.0.3 ===
October 9, 2009 (revision 856)
- Added support for Google Sites Data API. Thanks Eric!
- Added support for Google Contacts profiles. Thanks Julian Toledo and Pedro
Morais!
- Added support for the Spreadsheets v3 API Tables feed and Records feed.
- Added support for the Google Apps Admin Settings API. Thanks Jay Lee!
- New sample which illustrates use of the Blogger v1 API on App Engine using
OAuth. Thanks Wiktor Gworek!
- Added a new MediaSource for use with v2 data model classes. Thanks Eric!
- The v2 client classes can now force all requests to be made over SSL.
Thanks Eric!
- Revised the test config system to allow test settings to be specified as
command line arguments or prompt the user to enter them interactively.
- Fixed UTF encoding issue in unittest on big endian architectures.
- Fixed XML parsing bug in Spreadsheets data class. Thanks ppr.vitaly!
- Fixed bug in webmaster tools service verification method arguments. Thanks
eyepulp!
- Fixed download URL recognition in Document List Data API.
- Some progress made in re-adding support for Python2.3, not sure yet if
full Python2.3 support is completely feasible. Python 2.4-2.6 should work.
=== 2.0.2 ===
August 20, 2009 (revision 823)
- Added support for the Google Maps Data API. Thanks Roman!
- Added data model classes for the v3 Spreadsheets API. This API is not
yet fully supported by the library and the gdata.spreadsheets package
should be considered experimental until the next release. The v1
gdata.spreadsheet package is unchanged and should still work.
- Auth token lookups when running in App Engine are now memcached which
should improve efficiency. Thanks Marc!
- Fixed a bug with the v2 HTTP proxy client. If behind a proxy, the full
URL is now sent as the request selector. Thanks Dody!
- Removed the gdata.client.GDataClient class which had been deprecated.
=== 2.0.1 ===
July 23, 2009 (revision 805)
- Added support for Secure AuthSub, OAuth with HMAC and with RSA to the
v2 auth code.
- Added v2 data model classes for all XML elements in the gd namespace.
- Usability improvements in the gdata.docs.service module, thanks Eric!
- Made the dependency on elementtree only required if you are using
Python 2.4 and lower. Thanks brosner!
- Setting the ssl member in atom.service.AtomService will now override a URL
which starts with http:. Many thanks to Michael Ballbach for the patch.
- Added a proxy HTTP client which will use proxy environment variables to
send all requests through the desired proxy server if present.
- Added user agent string indicating that this is from the gdata python v2
library.
- Removed unneeded title parameter from the Blogger client's add_comment
method.
- Added an all_tests_local script which runs all tests except those which
would make HTTP requests to remote servers.
- Added the birthday element to contacts entry, the reminder element to
calendar events, and the method attribute to calendar's reminder class.
All with great thanks to Marc!
- The analytics GetData method no longer requires the dimensions argument.
Thanks Jim!
=== 2.0.0 ===
June 29, 2009 (revision 777)
- Released support for version 2 of the Google Data APIs protocol. To use the
new version-aware code, use the gdata.client, atom.data, gdata.data
modules and classes that derive from them. For example, use
gdata.blogger.client.BloggerClient instead of
gdata.blogger.service.BloggerService. High level, service specific classes
are not available yet for other services, so feel free to migrate as they
become available. The v1 service classes are still included so your
existing code should be unaffected.
- Included new Blogger samples to illustrate v2 support, a command line demo
and an app for App Engine.
- Added OAuth sample app which uses App Engine.
- Added the gCal:sequence element to CalendarEventEntry (thanks Anton).
- Added two decorators to diaply warnings when deprecated methods are used.
- Added a login utility for command line samples to share to standardize the
auth process on in a sample program.
- Fixed a bug in v2 XML parsing which prevented multiversion XML classes from
correctly parsing and generating v2 code.
- Fixed v1 HTTP Host header settings, not default ports should now be
included in the Host header.
- Fixed pickle error when an OAuth-RSA token is saved to the App Engine
datastore.
- Fixed missing data in the Document List API unit tests.
- Improved backwards compatibility of v2 data model classes be adding aliases
to v1 functions.
=== 1.3.3 ===
June 5, 2009 (revision 724)
- Added support for the Google Finance API. Thanks Swee Heng!
- The Google Data Service classes now support version 1.0a of the OAuth
protocol (Thanks Eric!).
- Fixed a naming bug in unit tests for atom.service which appeared when using
Python 2.6.x
- Fixed URL parameter propogation when calling GDataService.Delete. The URL
parameters are now preserved.
- Fixed incorrect return type for gdata.FeedLinkFromString.
- Added data model classes for the Atom and AtomPub XML elements which support
version 2 of the Google Data APIs (by means of dynamic AtomPub namespace
switching). The version 2 XML classes are much more backwards compatible
with v1 classes than in previous releases.
=== 1.3.2 ===
May 22, 2009 (revision 711)
- Added support for the Google Analytics API. Thanks Sal Uryasev!
- Added support for the Google Book Search API. Thanks James Sams!
- Improved support for 2 legged OAuth and added a sample app. Thanks Eric
Bidelman!
- Simplified the way an XML response from the server is parsed into the
desired class for the version 2.0 API client (gdata.client.GDClient).
=== 1.3.1 ===
April 23, 2009 (revision 695)
- Fixed issues with setting the developer key in the YouTubeService
constructor.
- For the Document List API (thanks Eric!):
- Added writersCanInvite element
- Fixed a small errors in Document Entry
- Added category label when creating different types of docs
- Fixed DownLoad helpers which shouldn't write a file if server returns
error. Fixes issue 240.
- Added DocumentListEntryFromString converter and using DocumentListEntry
class instead of GDataEntry class to create entries.
- URLs in HTTP requests can now be unicode strings. Resolves issue 233.
- Improvements for the upcoming version 2.0.0 release:
- The v2 auth code now support AuthSub and has been tested on App Engine.
- Unit tests for the v2 client code will default to not use the local file
cache but will make live requests unless cached responses is set to
True in the test configuration module.
- Older v1 unit tests can now use the v2 testing framework, migrated
contacts API tests as a proof of concept. This will come in handy when
testing backwards compatibility for the 2.0 release.
- Improved support for unicode and other character encodings in the v2 XML
core module.
=== 1.3.0 ===
Mar 20, 2009 (revision 665)
- Added support for the Google Health API. (Thanks Eric Bidelman!)
- Added support for the groups management in the Google Apps Provisioning
API. (Thanks to Tony Chen and Oskar Casquero!)
- Added the following new features for the Google Documents List API:
(Thanks again to Eric Bidelman for adding all of these!)
- Folder create, move docs/folders in and out of other folders. (Thanks
Nizam Sayeed!)
- Suppport for modifications to Access Control Lists for documents.
- Export functionality for documents, spreadsheets, and presentations.
- Updated the docs_example.py sample app.
- New new XML elements: resourceID, lastViewed, and lastModifiedBy.
- Added ability to update web clip settings in the Google Apps email settigns
API. (Thanks Takashi Matsuo.)
- Fixed a bug in calendarExample that would cause execution to fail when an
event attendee does not have any attendeeStatus data. (Thanks Trevor
Johns!)
- Fixed AuthSub request URLs which should be https. (Yay Eric!)
- Fixed logic when changing email settings when using the Google Apps API to
only send properties which have been specified. (Thanks Jay Lee!)
- Includes a new and experimental client class (gdata.client.GDClient) which
can be used with version two of the Google Data API protocol. This class
may change in backwards incompatible ways before the 2.0.0 release, so
use at your own risk.
Collection.
The Perl 5 module Starlet provides a standalone HTTP/1.0 server.
It is suitable for running HTTP application servers behind a reverse
proxy and supports the following features: prefork and graceful
shutdown using Parallel::Prefork, hot deployment using Server::Starter,
fast HTTP processing using HTTP::Parser::XS .
Pkgsrc changes:
- adjust dependencies
Upstream changes:
0.2001 Tue Apr 13 21:45:15 PDT 2010
- Fixed the way to set the default Delayed loader
0.2000 Tue Apr 13 20:22:24 PDT 2010
- INCOMPATIBLE: starman executable by default loads the application with Delayed to be safer.
Use --preload-app command line option to preload the application in the master process.
See `starman --help` for details.
0.1007 Tue Apr 13 19:45:59 PDT 2010
- Fixed a bug where Content-Length less response are sent in Keep-Alive without chunked,
choking HTTP/1.0 clients (patspam) #6
Version 2.8.2 (2010-04-13)
--------------------------
- Updated TCPDF to version 4.9.011 (#1802)
- Updated TinyMCE to version 3.3.2 (#1711)
- Updated MooTools More to version 1.2.4.4 (#1697)
- Added: added the white Mediabox theme (#1637)
- Added: DC_Table "oncut_callback" and "oncopy_callback" (#1745)
- Added: added the active record to "ondelete_callbacks" (#1721)
- Added: added class "trail" to the custom navigation module (#1691)
- Added: added the wildcard ##token## to the newsletter subscription
module (#1729)
- Added: added the Russian typolinks translation to TinyMCE (#1744)
- Added: the image resizer now considers the file modification time (#1743)
- Added: added ­ as basic entity (#1768)
- Added: added the "mode" parameter to the image insert tag (#1712)
- Added: added a loadDataContainer hook (#1687)
- Added: added an addComment hook (#1669)
- Fixed: special HTML characters were encoded twice (#1642)
- Fixed: the newsletter reader failed to replace insert tags (#1645)
- Fixed: the subtree navigation did not work in the popup file manager (#1653)
- Fixed: the BBCode parser did not add missing URI protocols (#1658)
- Fixed: PDF exports contained incorrect download links (#1654)
- Fixed: the option "always show in sitemap" did not work for hidden
pages (#1679)
- Fixed: the downloads element did not hide meta.txt files (#1688)
- Fixed: the downloads element did not pass all meta information (#1689)
- Fixed: reverted the changes to the navigation menu (#1635)
- Fixed: the function addToUrl() generated invalid URLs in the front end (#1707)
- Fixed: the Config library did not handle numeric strings correctly (#286)
- Fixed: the Mediabox plugin did not support .jpeg file extensions (#1728)
- Fixed: the file manager allowed invalid file and folder names (#1699)
- Fixed: the news and event reader did not overwrite the page
description (#1702)
- Fixed: regular users were not able to copy/move multiple content
elements (#1715)
- Fixed: the PDF export function did not handle whitespace characters
correctly (#1716)
- Fixed: news headlines appeared twice in link headlines (#1749)
- Fixed: the search module did not remove insert tags from the
keywords string (#1750)
- Fixed: the file tree widget did not validate the mounted paths (#1746)
- Fixed: the task center e-mails contained invalid links (#1753)
- Fixed: the option "show protected pages" also showed guest pages (#1694)
- Fixed: the list wizard did not render correctly if the first item
was empty (#1740)
- Fixed: System::idnaEncode() encoded more than just the domain name (#1748)
- Fixed: the Captcha field was missing class="mandatory" (#1767)
- Fixed: the newsletter recipient importer did not count correctly (#1732)
- Fixed: regular users could not edit comments on unmounted pages (#1659)
- Fixed: the ZipReader class failed to handle empty files (#1804)
- Fixed: the Newsletter module did not embed images correctly (#1799)
- Fixed: parameters passed to the file insert tag were cached by the
Input object (#1818)
- Fixed a possible request forgery vulnerability in the back end
- Fixed some minor issues
