-
Notifications
You must be signed in to change notification settings - Fork 52
Release Notes
TBD
09/17/2015 - The following changes have been added:
- Upgrade to JDK 1.8
08/10/2015 - The following changes have been added:
- Modify HeaderValueFeedEntityTagProcessor to handle multiple header values that are comma separated
07/21/2015 - The following changes have been added:
- When processing categories, the category terms get incorrectly converted to all lower case, causing issues with some category terms that contain alpha numeric ID, whose values are mapped to first class columns in the Database and used for searches/queries. To preserve backwards compatibility, this issue is fixed only for categories whose terms are mapped to first class columns in DB. I.e: those categories are no longer converted to all lower case. The other categories will remain converted to all lower case.
05/15/2015 - The following changes have been added:
- When retrieving feed entries from a datastore the id element of the feed is missing a colon. Made changes so that the id of the feed will be of the form urn:uuid:<uuid>.
10/1/2014 - The following changes have been added:
- The RegexBuilder was removing the backslash(eg:'\d{1,2}') from the resource attribute value of workspace configuration element and feed element in the atom-server.cfg.xml. This was avoiding a valid request path to pass the filter chain. Added a fix to solve the issue.
09/02/2014 - The following changes have been added:
- added a check so that if adapters are mis-configured without the "delimiter" property when "prefixColumnMap" property is used, it will throw exception
08/18/2014 - The following changes have been added:
- invalid characters in the category search used to return 500, now will return 400
08/13/2014 - The following changes have been added:
- search '%2B' used to return 500, now will return 400
07/07/2014 - The following changes have been added:
- provide a feature flag to SelectiveURIJSONFilter class to turn on/off the hashing of HTTP Entity Tag (ETag) header
07/07/2014 - The following changes have been added:
- externalize Provider interceptor, so one can now configure a list of Provider interceptor, so one can one can now configure a list of Provider interceptor, similar to the HTTP Servlet filters. Provider interceptor must implement org.apache.abdera.protocol.server.Filter interface. A Provider interceptor can be used, for example, to provide a feed page in JSON format.
Note that this is a breaking change from previous Atom Hopper release. Previously, returning a feed page in JSON format is automatically done when the proper Accept header is sent by client or if format=json query parameter is sent. Now, one must configure Atom Hopper installation in order to get this behavior. For examples on how to configure this, take a look at: application-context.xml and atom-server.cfg.xml
06/20/2014 - The following changes have been added:
-
externalize Adapter Response Interceptor configuration, so one can now configure a list of response interceptor from the atom-server.cfg.xml file. Adapter Response Interceptor can be used, for example, to build HTTP Entity Tag (ETag) header. Adapter Response Interceptors must implement org.atomhopper.abdera.filter.AdapterResponseInterceptor interface. For examples of this configuration, please take a look at: atom-server.cfg.xml
-
enrich Entity Tag (ETag) processing. You can now configure Atom Hopper with a different ETag processor that appends additional configured tag based on some header values. For examples on how to configure this, take a look at: application-context.xml
05/16/2014 - The following changes have been added:
- add a new feature where pre-configured categories can be saved to their designated database columns, instead of to the categories column of type Postgres array. Using designated database columns to save these categories improves search response time.
- The LogTest servlet has been removed
05/12/2014 - The following changes have been added:
- fix issues with timezone conversion on Australian CentOS/Linux systems returning EST for timezone short name, causing the startingAt query to be incorrectly handled.
03/25/2014 - The following changes have been added:
- fix issues of wrongly calculating the entry found with the specified timestamp, causing the startingAt query to return incorrect entries
03/18/2014 - The following changes have been added:
- add a new query parameter: startingAt to enable searching entries starting at a certain timestamp. For details on the semantic and format of this parameter, please consult the wiki page: https://github.com/rackerlabs/atom-hopper/wiki/Using-the-JDBC-Data-Adapter#search-by-timestamp
- Update logback to version 1.1.1
12/10/2013 - The following changes have been added:
- add a new boolean parameter: enableLoggingOnShortPage. If set to true, then the adapter will log a warning statement to the log file when the adapter returns a page whose number of entries is less than the requested limit parameter.
- add a new class that can report Tomcat connection pool metrics to Graphite. To use this class, you must configure it as a bean in application-context.xml
11/18/2013 - The following changes have been added:
- fix an issue when building pages with incorrect Archived link format
10/29/2013 - The following changes have been added:
- implementation of Archived entries, per RFC 5005
10/1/2013 - The following changes have been added:
- apply feedHeadDelayInSeconds parameter also when building pages with direction=forward
9/18/2013 - The following changes have added:
- fix an issue when configuring feedHeadDelayInSeconds parameter with values other than default values, the values were not honored.
9/13/2013 - The following changes have been added:
- when calculating the "last" page, remove the code that counts for all the entries. In some backend storage, doing count for all entries is an expensive operation and we think it's unnecessary. It was previously done due to some interpretation of the spec. But the spec did not explicitly mention to do count when building the "last" page.
9/3/2013 - The following changes have been added:
- add new optional parameter for JdbcFeedSource: feedHeadDelayInSeconds. This parameter allows Atom Hopper JDBC Adapter to return only entries that are feedHeadDelayInSeconds older than current timestamp. In a busy multi-threaded and multi-process database, timestamp and id are not guaranteed to be in order. I.e: entries with earlier timestamp may get inserted later and the same is with id. This parameter allows Atom Hopper to only return list of entries whose order will not likely to change by late inserts.
7/29/2013 - The following changes have been added:
- add Around Advice that knows how to retry connection failures from Tomcat connection pools to the database backend.
7/18/2013 - The following changes have been added and tested for this release:
- Add Previous and Next links as HTTP "Link" header
5/28/2013 - The following changes have been made:
- Added Regular Expression Support for Feeds within a namespace.
- JDBC Adapter: Switch to use Tomcat connection pool
- JDBC Adapter: Return 409 for duplicate entries
- Fix bug where responses do not have a proper Content-Type
- RPM package fixes
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres and JDBC performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
- Multi-node, high volume use can result in duplicate timestamps, which can result in missing entries during paging, except with the JDBC Adapter.
The following features have been added and tested for this release:
- Adding complex search ability (LDAP format) to the JDBC Adapter
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres and JDBC performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
- Multi-node, high volume use can result in duplicate timestamps, which can result in missing entries during paging, except with the JDBC Adapter.
The following features have been added and tested for this release:
- Adding Yammer Metrics timers to JDBC Adapter
- JDBC adapter refactored to use a mock last link marker in the head call for improved performance on GET of feed head.
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres and JDBC performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
- Multi-node, high volume use can result in duplicate timestamps, which can result in missing entries during paging, except with the JDBC Adapter.
The following features have been added and tested for this release:
- Adding Metrics JMX Counters to track number of entries posted per feed.
- Changed WAR name from ROOT.war to ATOMHOPPER.war in rpm package.
- Fixed New JDBC Adapter to not miss entries while paging if they have the same datelastupdated timestamp.
- Fixed 500 error when adding an entry with a duplicate entry id.
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres and JDBC performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
- Multi-node, high volume use can result in duplicate timestamps, which can result in missing entries during paging, except with the JDBC Adapter.
The following features have been added and tested for this release:
- Fixed New JDBC Adapter to not miss entries while paging if they have the same datelastupdated timestamp.
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres and JDBC performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
- Multi-node, high volume use can result in duplicate timestamps, which can result in missing entries during paging, except with the JDBC Adapter.
- Build 1.2.4 skipped due to build failure.
The following features have been added and tested for this release:
- Addition of a new JDBC Adapter that contains a new auto-incrementing ID column.
- Fixed URL encoding, so that only the correct characters are encoded.
- Fixed missing location header.
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
- Multi-node, high volume use can result in duplicate timestamps, which can result in missing entries during paging, except with the JDBC Adapter.
- Build 1.2.2 skipped due to build failure.
The following features have been added and tested for this release:
- Added atom:published date for entries read from feed
- Multi-category search switched from an AND to an OR search. This is temporary and will be updated in a future release to support AND and OR searches.
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
- URLs with invalid characters cause a 500 internal server error, rather than a 4XX response. This only occurs when certain characters are in the URI, such as 'localhost:8080/}'
This release provides a Postgres data adapter with JDBC support for improved performance.
The following features have been added and tested for this release:
- Postgres data adapter - this adds JDBC support for the Postgresql database as an additional data adapter. This adapter provides:
- Performance improvement
- Support for multiple category search
- Migration data adapter - this adds the ability to migrate from one database schema to another while entries are still being added to the feed.
- Added the ability to override the entry id and date values for new Atom entries
- Log4j has been removed and replaced with logback
- Fixed two issues with the embedded Jetty server and ports (clarifications on ports)
- Fixed three issues using criteria in Hibernate Adapter that caused multiple queries to run
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter; the MongoDB and Postgres performance data adapters work as expected).
- When using the default database schema with Postgresql using a LOB datatype for the entrybody, posting an entry with unicode characters in the content element without explicitly setting the Content-Type header with the UTF-8 charset can cause parsing errors when that entry is requested. (Specifically, content with the right single closed quote character (’) causes the error)
- Multiple 'Content' elements can be inserted in an Atom entry and posted. This appears to be a bug in Apache Abdera (not fully validating the Atom Entry when parsing). We've entered a bug into Abdera's issue tracker: https://issues.apache.org/jira/browse/ABDERA-307
This release reverts the old Hibernate session manager code as it was causing problems with multiple feeds and multiple databases
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter, the MongoDB data adapter works as expected).
This release removes the logback dependencies from the POM.
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter, the MongoDB data adapter works as expected).
This release provides a couple performance improvements.
The following features have been added and tested for this release:
- Updated the HibernateSessionManager class to use an updated Hibernate 4.x method rather than the deprecated one it was using.
- Database reads no longer use transactions.
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter, the MongoDB data adapter works as expected).
The following features have been added and tested for this release:
- C3P0 connection pooling is now enabled by default in application-context.xml
- Fixed issue where an invalid UUID was returning HTTP 500 rather than HTTP 404
- Fixed issue where no default value is supplied for the scheme (atom-server.cfg.xml). It now defaults to HTTP
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (this only affects the Hibernate data adapter, the MongoDB data adapter works as expected).
The following features have been added and tested for this release:
- Updated jar dependencies
- Fixed issue where connections were not being closed properly in the Hibernate data adapter
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- (This only affects the Hibernate data adapter) Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching with Hibernate is an experimental feature)
The following features have been added and tested for this release:
- Added example code to the application-context.xml for connecting to MongoDB using a replica set with authentication
- Changed MongoDB data adapter to use one collection per feed to keep feeds in their own collection
- Added connection pooling settings for C3P0
- Add both the MySQL and Postgresql drivers to be included within the WAR
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- (This only affects the Hibernate data adapter) Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching with Hibernate is an experimental feature)
The following features have been added and tested for this release:
- Ability to force links (feed header links and self references in the Atom entries) to use HTTP or HTTPS, this allows for a configuration where incoming requests are via SSL but get terminated as SSL by a load balancer, etc.
- (FIXED) Issue where an empty POST (no valid Atom XML) returned an HTTP 500 instead of HTTP 422
- (FIXED) Issue where an HTTP 500 occurred when following the previous from the feed head
- (FIXED) Issue with MongoDB last link when using search criteria, the search criteria was not applied so the paging was off
- (FIXED) Issue with the feed self link (on feed head request it was defaulting to forward)
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- (This only affects the Hibernate data adapter) Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching with Hibernate is an experimental feature)
This release is focused on adding a new servlet to ensure logging is functioning and configured correctly.
The following features have been added and tested for this release:
- A new servlet has been created and mapped that points to /logtest in the web.xml file. This allows a quick test to ensure log files are being updated.
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- (This only affects the Hibernate data adapter) Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching with Hibernate is an experimental feature)
This release is focused on updates to the Next, Previous, and First links in the feed head. It also includes in-progress support for a non-RDMS data adapter and full category search.
The following features have been added and tested for this release:
- MongoDB data adapter - this adds support for the MongoDB NoSQL database as an additional data adapter (this is a preview feature, production environments should still be using the Hibernate data adapter)
- Next, Previous and Last links in the feed header have been updated to work more similar to other Atom feeds - this changes how you move through Atom entries in a feed and is more "page" based now.
The following have been removed in this release:
- C3P0 connection pooling has been removed
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- (This only affects the Hibernate data adapter) Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching with Hibernate is an experimental feature)
This release was focused on a bug fix.
- None
- (FIXED) Issue where multiple databases with same feed name were causing Atom entries to persist to the same database
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching is an experimental feature)
This release was focused on a bug fix.
- None
- (FIXED) Issue where the feed header links are getting & values incrementally added
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching is an experimental feature)
This release was focused on testing and bug fixes.
The following features have been added and tested for this release:
- In the feed head "oldest-entry" link has been renamed to "last" to be more Atom compliant
- Added "updated" element to the feed head
- Added "self" link to the feed head
- General code clean-up and adding of new tests
- Atom feed generated by Atom Hopper is now compliant according to the W3C validator
- Renamed the RPM and WAR artifact to have "atomhopper" in the name rather than "ah-war"
- The limit on the feed entries returned must be an integer between 1 and 1000, will default to 25
- Documentation is now provided via a PDF
- (FIXED) Issue where invalid values could be in limit parameter and cause HTTP 500 errors rather than HTTP 400
- (FIXED) Issue where you could pass a UUID that is not valid in the current namespace/feed and it would still be returned
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed (non-critical)
- Feed header links are getting & values incrementally added (non-critical)
- Category searches only work with one category search parameter and return duplicate entries when multiple categories are searched upon (category searching is an experimental feature)
This release is mainly focused on adding the read oldest entry ID functionality to the Atom Hopper code, testing and bug fixes.
The following features have been added and tested for this release:
- Added the ability to get the oldest entry from the feed data
- Modified the RPM build to not overwrite existing configuration files
- Documented how to setup Atom Hopper with multiple databases with multiple feeds
- Log4j.properties file is now read from /etc/atomhopper and the log file stores to /var/log/atomhopper.log
- General RPM clean-up (several scriptlets removed)
- Update parent POM with latest H2 release
- (FIXED) Issue with forward paging not ordering the entries correctly
- (FIXED) Issue with the current marker not being present in backward paging
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
This release is mainly focused on testing and bug fixes.
The following features have been added and tested for this release:
- Changed the file permissions on the RPM build to 644 from 755
- (FIXED) Issue when using the same feed name with differing namespaces
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
This release is mainly focused on packaging updates and bug fixes.
The following features have been added and tested for this release:
- Added a clean up script to remove old entries
- GitHub ReadMe and Wiki were updated with latest information
- (FIXED) The self referencing Atom entry link doesn't support HTTPS configurations
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
This release is mainly focused on packaging updates and bug fixes.
The following features have been added and tested for this release:
- Moved the application.context.xml file to /etc/atom-hopper to keep all config files in a consistent location.
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
- The self link doesn't support HTTPS configurations
These releases involved changes and testing for the build.
Release 0.9.8 is mainly focused on packaging updates and bug fixes.
The following features have been added and tested for this release:
- Updated dependencies with the latest versions
- Added documentation back into build
- Synched POM versions (for Jenkins build machine)
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
Release 0.9.7 is mainly focused on removing the ascending feed order configuration (not compliant with the Atom spec) and getting the RPM ready for field testing.
The following features have been added and tested for this release:
- Removed ASC order configuration
- Modified the RPM to match the custom built Apache Tomcat 7 RPM
- None
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
Release 0.9.6 is mainly focused on bug fixes.
The following features have been added and tested for this release:
- No new features are planned for this release.
- (FIXED) Changing the servlet URL Pattern value from the default of /* causes the Atom entry self links to be incorrect
- (FIXED) When getting the feed page from the Hibernate data adapter there was no query criteria on the feed name (affects scenarios with multiple feeds)
- (FIXED) Criteria to query on feed name was not being applied to the getFeedPage method
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
Improved database read/write speed of Atom entries. Tested against H2, MySQL and Postgresql databases. Some preliminary benchmarks have been completed.
- None
The following bugs are known and will be fixed in future releases:
- The author value for the feed is not displayed
- Changing the Servlet URL Pattern value from the default of /* causes the Atom entry self links to be incorrect
Updates: made category search case insensitive, debian package build, browser UI, and minor bug fixes.
Updates: Minor issue fixed with a single operator on the query string but no search criteria specified, added C3P0 connection pooling to Hibernate.
Updates to Release 0.9. Includes: bug fixes, weak eTag support, the ability to add categories and query/filter on those categories, ordering the Atom feed (LIFO, FIFO), support for JSON feed returns, and the update of the default Jetty server (for the standalone version) to the final Jetty 8.x release.
Updates to Release 0, includes bug fixes.
Release 0, used for initial testing.
The version numbers consist of three parts: major.minor.patch
Major The major number is the first integer in the version string. Changes in the major number typically indicate a significant change in the code base and/or end-user functionality. Major changes are not necessarily backwards compatible.
Minor new features, feature improvements. Patch Bug: bug releases.