- 14 10月, 2014 2 次提交
-
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
- 13 10月, 2014 4 次提交
-
-
由 Till Rohrmann 提交于
-
由 Robert Metzger 提交于
-
由 Stephan Ewen 提交于
-
由 Aljoscha Krettek 提交于
Fix bug in Blob Manager, Cleanup Timer was not cancelled on Shutdown -
-
- 11 10月, 2014 2 次提交
-
-
由 Aljoscha Krettek 提交于
-
由 Aljoscha Krettek 提交于
The java8 Xdoclint:none fix was not correctly transferred when moving the javadoc building to a jekyll plugin.
-
- 10 10月, 2014 3 次提交
-
-
由 Robert Metzger 提交于
-
由 Aljoscha Krettek 提交于
-
由 Aljoscha Krettek 提交于
Also fixes a bug with Java/Scala interop: TupleTypeComparator was only checking for nested Java Tuples and Pojos, not Scala Case classes.
-
- 09 10月, 2014 5 次提交
-
-
由 Fabian Hueske 提交于
This closes #146
-
由 Fabian Hueske 提交于
-
由 Fabian Hueske 提交于
-
由 Robert Metzger 提交于
-
由 Fabian Hueske 提交于
Added Flink logo to JobManager web interface. Updated Run Example Quickstart screenshots Removed old gradients and unused image files from submission client. This closes #145.
-
- 08 10月, 2014 16 次提交
-
-
由 Fabian Hueske 提交于
-
由 twalthr 提交于
This closes #131
-
由 Fabian Hueske 提交于
Added test cases. Minor improvements and clean-ups.
-
由 Artem Tsikiridis 提交于
[FLINK-1076] Extend Hadoop compatibility. Added wrappers for stand-alone Map, Reduce, and CombinableReduce functions.
-
由 Fabian Hueske 提交于
This reverts commit 74dded1c.
-
由 Fabian Hueske 提交于
Tests and documentation included. Also tested on cluster. I hijacked @twalthr's PR #131 to build the documentation of the Hadoop function wrappers on top. Author: Fabian Hueske <fhueske@apache.org> Author: Artem Tsikiridis <artem.tsikiridis@cern.ch> Author: twalthr <info@twalthr.com> Closes #143 from fhueske/hadoopFunctions and squashes the following commits: f1b3f21 [Fabian Hueske] [FLINK-1076] Extended Hadoop Compatibility documentation to cover Hadoop functions f4be4a0 [twalthr] [FLINK-1107] Hadoop Compatibility Layer documented 4a62bdb [Fabian Hueske] [FLINK-1076] Return type determined via type extraction. Added test cases. Minor improvements and clean-ups. df947d9 [Artem Tsikiridis] [FLINK-1076] Extend Hadoop compatibility. Added wrappers for stand-alone Map, Reduce, and CombinableReduce functions.
-
由 Aljoscha Krettek 提交于
This also adds more integration tests, but not all tests of the Java API have been ported to Scala yet.
-
由 Robert Metzger 提交于
-
由 Robert Metzger 提交于
- improved error messages - the composite vs atomic bug aljoscha found - Comparator test for Pojo Comparator enabled - TODO removed - string-based key expression for group sorting fields definition - support for specifying "select all" using * and now also _ (for scala fans) - Exception if user is having multiple fields with the same name in the class
-
由 Robert Metzger 提交于
-
由 Robert Metzger 提交于
-
由 Robert Metzger 提交于
-
由 Gyula Fora 提交于
-
由 Tamas Ambrus 提交于
-
由 szape 提交于
-
由 Fabian Hueske 提交于
This closes #136
-
- 07 10月, 2014 1 次提交
-
-
由 Robert Metzger 提交于
-
- 06 10月, 2014 1 次提交
-
-
由 Daniel Warneke 提交于
Looped through the user code class loader from the envrionment to the tasks. Added timed cleanup task to LibraryCacheManager to remove automatically unreferenced jar files. Moved registration logic out of the IOReadableWritable serialization logic. Updated logger. Made the LibraryCacheManager an object instead of a Singleton. Added proper shutdown of task manager and job manager. Removed nio import to support Java 6 compilation. Removed lock map from BlobLibraryCacheManager. Using global lock for synchronization. Marked StackOverflow code. Removed JavaDocs copyright message. Added getUserCodeClassLoader to AbstractInvokable class. This closes #126 and closes #107.
-
- 04 10月, 2014 1 次提交
-
-
由 ghermann 提交于
Conflicts: flink-addons/flink-streaming/flink-streaming-connectors/pom.xml [streaming] Connectors POM formatting fix
-
- 03 10月, 2014 5 次提交
-
-
由 Fabian Tschirschnitz 提交于
This closes #138
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-
由 Stephan Ewen 提交于
-