Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
tree: 0f5a1b65f9
Fetching contributors…

Cannot retrieve contributors at this time

236 lines (236 sloc) 10.846 kb
=== Feedtools 0.2.29
* fixed bug with utf-8 content in resolve_relative_uris
=== FeedTools 0.2.28
* fixed major bug with autodiscovery
=== FeedTools 0.2.27
* now uses a new URI class
* full support for IRIs
* uses html5lib instead of HTree
* schema changed to allow additional functionality
* added better support for license information
* added method to allow feeds and feed items to override lazy parsing
* added configuration setting to override lazy parsing
* added serialization feature to allow preparsing with FeedUpdater
* added support for HTTP Basic authentication
* fixed some issues with identifying the correct base uri
* added config path method to the database feed cache
* added default time to live configuration variable
* fixed autodiscovery redirect loops
* improved handling of entities
* no longer causes problems with Mongrel's monkey patch
=== FeedTools 0.2.26
* added support for retarded, ugly Radio redirects
* improved autodiscovery support
* fixed lazy evaluation issue
=== FeedTools 0.2.25
* fixed self reference href bug
* fixed primary link assignment heuristic
* link rel's now default to 'alternate' for atom feeds
* better handling of language attributes
* better support for threaded applications
* generated rss 2.0 now includes author, managingEditor, webMaster elements
* improved handling of author elements
* improved handling of http timeouts
* fixed issue with relative paths in the location header on http redirection
* monkey patches now work with Ruby 1.8.4's version of REXML
* REXML monkey patches moved into their own file
=== FeedTools 0.2.24
* fixed autodiscovery bugs
* updated autodiscovery to handle relative uris
* added per feed configuration
* rewrote http retrieval code
* supports authenticated http proxies
* supports basic http auth
* fixed relative url resolution issues
* corrected db schema files to match migration file
* fixed bug in the save method
* fixed some major http bugs
* updated dependancies
* no longer causes problems with frozen rails
=== FeedTools 0.2.23
* autodiscovery implemented
* now knows a title from a hole in the ground
* now resolves relative urls when possible
* changed default table name to "cached_feeds" to avoid name collisions
* schema now uses "href" instead of "url"
* feed cache is set to nil by default now
* both summary and content elements are generated now
* now supports proxies
* now supports internationalized domain names if libidn is installed
* fixed bug with feed merging referencing a method that was refactored
* no longer dies if uuidtools gem is missing but the UUID class is defined
* updated timestamp handling and generation
* added support for entry sorting on any feed item field
* added support for disabling entry sorting entirely
* fixed issue with itunes categories
* fixed itunes subtitle/summary
* fixed entry assignment bug
* fixed issued/published variable name mix-up
* added support for the payload module
* added support for xhtml:div elements
* dc:date now preempts pubDate
* added better support for the scriptingNews format
* now correctly strips out wrapper div elements from text constructs
* fixed issue with some atom links being incorrectly identified as images
* reorganized some of the helper modules
* made some portions of url normalization case insensitive
* fixed issue with filename handling on Windows
=== FeedTools 0.2.22
* fixed another atom generation error
=== FeedTools 0.2.21
* fixed atom generation error
* added summary element to generated atom xml
* added warning message for improperly set up database tables
=== FeedTools 0.2.20
* fixed some atom namespace bugs
* added a summary method to feed items
* added some support for the apple-wallpapers elements
=== FeedTools 0.2.19
* lousy encoding support (as opposed to none at all)
* xml processing instruction now correctly prefixes generated feeds
* attributes are escaped properly when generating feeds
* uppercase html is no longer sanitized for not being in the whitelist
* added alias method for assigning to entries
* changed the xpath querying to be much, much more DRY
* find_node and find_all_nodes are actually useful now
* full case-insensitivity implemented for the xpath helper methods
* fixed bug in tests where some assertion failures could affect other tests
* fixed bug where the feed item author would sometimes be parsed incorrectly
* fixed bug where the convertLineBreaks element would break feed entries
* default (i.e. preferred) methods will be Atom-style instead of RSS-style
* default feed output format changed to Atom 1.0
* itunes namespace corrected
* fixed images property when dealing with atom
* fixed atom link property
* improved timestamp handling
* whitespace nodes now ignored by REXML
* added option to disable timestamp estimation
* added option to limit time-to-live to some upper maximum
* enclosures included in feed generation
* no longer uses the cache at all for file:/// urls
* changed itunes:keywords to use commas
* testing now excludes cache testing by default
* more tests
=== FeedTools 0.2.18
* no longer ever polls more often than once every 30 minutes
* fixed overlooked improperly refactored enclosure code
* fixed issue with inner_xml incorrectly handling xml comments
* added helper modules
* test cases now implemented using helpers
* fixed issue with timeouts
* fixed stack overflow while estimating timestamps
* fixed some namespace issue with atom
* added base64 decoding support
=== FeedTools 0.2.17
* more fixes for timestamping of feed items
* fixed nil bug in root_node, feed_type, feed_version, build_xml
* build_xml method is now properly unit tested
* fixed issue with setting the Host http header
* namespaces are no longer hard-coded for feed generation
* better url normalization for file: style urls
* generated timestamps are now in UTC
=== FeedTools 0.2.16
* atom feed generation now defaults to atom 1.0
* removed erroneous published element at atom feed-level
* fixed some incompatibilities with ruby 1.8.3
* better support for universal subscription mechanism within feeds
* now handles both atom 0.3 and 1.0 namespaces
* fixed timestamping of feed items
=== FeedTools 0.2.15
* changed so that tidied elements no longer handle entities differently
* tidy now assumes utf-8 encoding instead of ascii
* began the first bit of work on namespace support
* better support for cdatas
* removed "bozo bit" - no need for a parser to moonlight as a validator
=== FeedTools 0.2.14
* fixed bug with feeds not expiring properly
=== FeedTools 0.2.13
* fixed issued with missing namespace parameter for uuid creation
=== FeedTools 0.2.12
* schema change!
* fixed missing cache connection breaking redirects
* fixed redirects creating duplicate cache entry
* exception now raised if cache_only option is set but the cache is disabled
* additional unit tests to verify that the cache works with redirects
* fixed feed expiration bugs with expire!
* xml_data renamed to feed_data
* feed_data_type method added
=== FeedTools 0.2.11
* ruby -w shouldn't produce nearly as many warnings
* you can now force the open method to only pull from the cache
* the global cache_only option has been removed
* nows queries the cache after each redirection
* attribute dictionary stuff removed for being unnecessary and pythonic
* better file structure -- no more 5000 line files
* schema files are formatted better to keep them from looking wierd in rdoc
=== FeedTools 0.2.10
* http error messages now sport a list of redirections to simplify debugging
* automatic table creation removed
* database schema information included in rdoc and package files
* deals with servers returning 404 if passed a User-Agent http header
=== FeedTools 0.2.9
* changed ordering of elements checked for the item content
* added dependancy on uuidtools, uuids used in feed generation
* database feed cache implementation factored out into its own file
* http_fetch changed to use lambda
* fixed invalid rss 1.0/2.0 generation (missing some namespaces)
=== FeedTools 0.2.8
* fixed bug with http headers being loaded from invalid YAML
* fixed uninitialized constant bug (you shouldn't have RUBYOPT set)
* cleaned up inspect methods
=== FeedTools 0.2.7
* added support for Atom 1.0 output
* improved support for Atom parsing and handling
* added no content string option to keep the feed validator happy
* greatly improved handling of timestamps and feed item sorting
* added aliases for the Atom 1.0 elements that were renamed
* feed merging is now threaded
=== FeedTools 0.2.6
* added support for merging feeds
* vastly improved handling of feed creation
=== FeedTools 0.2.5
* fixed multiple rows being created in the cache after a 301 redirection
* fixed broken table creation for postgresql and sqlite
* testing against non-standard feeds
* removed the 'comment_link' method in favor of the 'comments' method
=== FeedTools 0.2.4
* fixed bug in the sqlite table creation query
* greatly improved image support
* improved cdf support (still needs way more work)
* support for text input elements
* now possible to force retrieval from the cache only
* increased the flexibility of the database caching implementation
* feed attributes accessible through the keys and values collections
* minor coding style changes
* fixed really dumb typo in the podcast? and vidlog? methods
* fixed exception from missing titles and descriptions
* now passes all of mark pilgrim's well-formed rss tests
=== FeedTools 0.2.3
* fixed omission of get parameters from http requests
=== FeedTools 0.2.2
* fixed http redirection bug
* fixed several documentation typos
* still more unit tests
* improved support for atom
* minor improvements to the database caching mechanism
* more complete support for rss elements
* major improvements to the handling of tags containing html content
=== FeedTools 0.2.1
* fixed incorrect dependancy on ActiveRecord 1.10.1
* more unit tests
* preliminary support for "bozo-detection"
=== FeedTools 0.2.0
* more complete support for rss, atom, cdf
* modular caching mechanism
* lazy parsing
* HTML sanitization of possibly dangerous fields
* HTML tidy support
* support for podcasts and vidlogs
* corrected handling of http redirection
* made http header information available
* file: protocol support
* custom parsing can be done using the find_node and find_all_nodes methods
=== FeedTools 0.1.0
* basic support for rss, atom, cdf
* basic caching using active record
* support for etags
Jump to Line
Something went wrong with that request. Please try again.