[jira] [Commented] (FLINK-1372) TaskManager and JobManager do not log startup settings any more
[ https://issues.apache.org/jira/browse/FLINK-1372?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14279982#comment-14279982 ] Robert Metzger commented on FLINK-1372: --- Yes, switching the logger sounds good. The information is very helpful when debugging issues reported by users. TaskManager and JobManager do not log startup settings any more --- Key: FLINK-1372 URL: https://issues.apache.org/jira/browse/FLINK-1372 Project: Flink Issue Type: Bug Components: JobManager, TaskManager Affects Versions: 0.9 Reporter: Stephan Ewen Assignee: Till Rohrmann Fix For: 0.9 In prior versions, the jobmanager and taskmanager logged a lot of startup options: - Environment - ports - memory configuration - network configuration Currently, they log very little. We should add the logging back in. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (FLINK-655) Rename DataSet.withBroadcastSet(DataSet, String) method
[ https://issues.apache.org/jira/browse/FLINK-655?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280004#comment-14280004 ] Till Rohrmann commented on FLINK-655: - I think the getBroadcastVariable returns a List of the DataSet's elements you have broadcasted. Thus, the return type should at least stay of Collection type. Rename DataSet.withBroadcastSet(DataSet, String) method --- Key: FLINK-655 URL: https://issues.apache.org/jira/browse/FLINK-655 Project: Flink Issue Type: Improvement Components: Java API Reporter: Ufuk Celebi Assignee: Henry Saputra Labels: breaking-api, github-import, starter Fix For: pre-apache To broadcast a data set you have to do the following: ```java lorem.map(new MyMapper()).withBroadcastSet(toBroadcast, toBroadcastName) ``` In the operator you call: ```java getRuntimeContext().getBroadcastVariable(toBroadcastName) ``` I propose to have both method names consistent, e.g. - `withBroadcastVariable(DataSet, String)`, or - `getBroadcastSet(String)`. Imported from GitHub Url: https://github.com/stratosphere/stratosphere/issues/655 Created by: [uce|https://github.com/uce] Labels: enhancement, java api, user satisfaction, Created at: Wed Apr 02 16:29:08 CEST 2014 State: open -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1382][java] Adds the new basic types Vo...
Github user rmetzger commented on the pull request: https://github.com/apache/flink/pull/299#issuecomment-70226685 The change looks good. I would like to see some test cases there. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user rmetzger commented on the pull request: https://github.com/apache/flink/pull/311#issuecomment-70226947 How about names along the lines of Unmodified Fields ? --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user rmetzger commented on a diff in the pull request: https://github.com/apache/flink/pull/311#discussion_r23069617 --- Diff: flink-compiler/src/main/java/org/apache/flink/compiler/dag/BinaryUnionNode.java --- @@ -266,4 +268,44 @@ public void computeOutputEstimates(DataStatistics statistics) { this.estimatedOutputSize = in1.estimatedOutputSize 0 in2.estimatedOutputSize 0 ? in1.estimatedOutputSize + in2.estimatedOutputSize : -1; } + + public static class UnionSemanticProperties implements SemanticProperties { + + @Override + public FieldSet getTargetFields(int input, int sourceField) { + if (input != 0 input != 1) { + throw new IndexOutOfBoundsException(); --- End diff -- How about returning an exception that explains that unions only support input to be 0 or 1. ? --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user rmetzger commented on a diff in the pull request: https://github.com/apache/flink/pull/311#discussion_r23069937 --- Diff: flink-java/src/main/java/org/apache/flink/api/java/functions/SemanticPropUtil.java --- @@ -40,38 +45,108 @@ import org.apache.flink.api.java.functions.FunctionAnnotation.ReadFields; import org.apache.flink.api.java.functions.FunctionAnnotation.ReadFieldsFirst; import org.apache.flink.api.java.functions.FunctionAnnotation.ReadFieldsSecond; +import org.apache.flink.api.java.operators.Keys; +import org.apache.flink.api.java.typeutils.TupleTypeInfo; public class SemanticPropUtil { - private final static String REGEX_LIST = (\\s*(\\d+\\s*,\\s*)*(\\d+\\s*)); - private final static String REGEX_FORWARD = (\\s*(\\d+)\\s*-( + REGEX_LIST + |(\\*))); - private final static String REGEX_LIST_OR_FORWARD = ( + REGEX_LIST + | + REGEX_FORWARD + ); - private final static String REGEX_ANNOTATION = (\\s*( + REGEX_LIST_OR_FORWARD + \\s*;\\s*)*( + REGEX_LIST_OR_FORWARD + \\s*)); + private final static String REGEX_WILDCARD = [\\+ Keys.ExpressionKeys.SELECT_ALL_CHAR+\\+ Keys.ExpressionKeys.SELECT_ALL_CHAR_SCALA+]; + private final static String REGEX_SINGLE_FIELD = [a-zA-Z0-9_\\$]+; + private final static String REGEX_NESTED_FIELDS = (( + REGEX_SINGLE_FIELD + \\.)* + REGEX_SINGLE_FIELD + )(\\.+ REGEX_WILDCARD +)?; + private final static String REGEX_LIST = (( + REGEX_NESTED_FIELDS + ;)*( + REGEX_NESTED_FIELDS + );?); + private final static String REGEX_FORWARD = ((+ REGEX_NESTED_FIELDS +|+ REGEX_WILDCARD +)-( + REGEX_NESTED_FIELDS + |+ REGEX_WILDCARD +)); + private final static String REGEX_FIELD_OR_FORWARD = ( + REGEX_NESTED_FIELDS + | + REGEX_FORWARD + ); + private final static String REGEX_ANNOTATION = (( + REGEX_FIELD_OR_FORWARD + ;)*( + REGEX_FIELD_OR_FORWARD + );?); + + private static final Pattern PATTERN_WILDCARD = Pattern.compile(REGEX_WILDCARD); private static final Pattern PATTERN_FORWARD = Pattern.compile(REGEX_FORWARD); private static final Pattern PATTERN_ANNOTATION = Pattern.compile(REGEX_ANNOTATION); private static final Pattern PATTERN_LIST = Pattern.compile(REGEX_LIST); + private static final Pattern PATTERN_FIELD = Pattern.compile(REGEX_NESTED_FIELDS); - private static final Pattern PATTERN_DIGIT = Pattern.compile(\\d+); - - public static SingleInputSemanticProperties createProjectionPropertiesSingle(int[] fields) { + public static SingleInputSemanticProperties createProjectionPropertiesSingle(int[] fields, CompositeType? inType) + { SingleInputSemanticProperties ssp = new SingleInputSemanticProperties(); - for (int i = 0; i fields.length; i++) { - ssp.addForwardedField(fields[i], i); + + int[] sourceOffsets = new int[inType.getArity()]; + sourceOffsets[0] = 0; + for(int i=1; iinType.getArity(); i++) { + sourceOffsets[i] = inType.getTypeAt(i-1).getTotalFields() + sourceOffsets[i-1]; } + + int targetOffset = 0; + for(int i=0; ifields.length; i++) { + int sourceOffset = sourceOffsets[fields[i]]; + int numFieldsToCopy = inType.getTypeAt(fields[i]).getTotalFields(); + + for(int j=0; jnumFieldsToCopy; j++) { + ssp.addForwardedField(sourceOffset+j, targetOffset+j); + } + targetOffset += numFieldsToCopy; + } + return ssp; } - public static DualInputSemanticProperties createProjectionPropertiesDual(int[] fields, boolean[] isFromFirst) { + public static DualInputSemanticProperties createProjectionPropertiesDual(int[] fields, boolean[] isFromFirst, + TypeInformation? inType1, TypeInformation? inType2) + { DualInputSemanticProperties dsp = new DualInputSemanticProperties(); - for (int i = 0; i fields.length; i++) { - if (isFromFirst[i]) { - dsp.addForwardedField1(fields[i], i); + int[] sourceOffsets1; + if(inType1 instanceof TupleTypeInfo?) { + sourceOffsets1 = new int[inType1.getArity()]; + sourceOffsets1[0] = 0; + for(int i=1; iinType1.getArity(); i++) { + sourceOffsets1[i] = ((TupleTypeInfo?)inType1).getTypeAt(i-1).getTotalFields() + sourceOffsets1[i-1]; + } + } else { +
[jira] [Commented] (FLINK-655) Rename DataSet.withBroadcastSet(DataSet, String) method
[ https://issues.apache.org/jira/browse/FLINK-655?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280030#comment-14280030 ] Fabian Hueske commented on FLINK-655: - Aren't we going for two methods? 1. {{op.withBroadcastSet(DataSetX, String)}} + {{ListX bcs = getRuntimeContext().getBroadcastSet(String)}} 2. {{op.withBroadcastValue(DataSetX, String)}} + {{X bcv = getRuntimeContext().getBroadcastValue(String)}} For the broadcast value, the runtime should check that the dataset has only a single value (implying DOP=1 for the producing operator). Rename DataSet.withBroadcastSet(DataSet, String) method --- Key: FLINK-655 URL: https://issues.apache.org/jira/browse/FLINK-655 Project: Flink Issue Type: Improvement Components: Java API Reporter: Ufuk Celebi Assignee: Henry Saputra Labels: breaking-api, github-import, starter Fix For: pre-apache To broadcast a data set you have to do the following: ```java lorem.map(new MyMapper()).withBroadcastSet(toBroadcast, toBroadcastName) ``` In the operator you call: ```java getRuntimeContext().getBroadcastVariable(toBroadcastName) ``` I propose to have both method names consistent, e.g. - `withBroadcastVariable(DataSet, String)`, or - `getBroadcastSet(String)`. Imported from GitHub Url: https://github.com/stratosphere/stratosphere/issues/655 Created by: [uce|https://github.com/uce] Labels: enhancement, java api, user satisfaction, Created at: Wed Apr 02 16:29:08 CEST 2014 State: open -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user fhueske commented on a diff in the pull request: https://github.com/apache/flink/pull/311#discussion_r23070484 --- Diff: flink-compiler/src/main/java/org/apache/flink/compiler/dag/BinaryUnionNode.java --- @@ -266,4 +268,44 @@ public void computeOutputEstimates(DataStatistics statistics) { this.estimatedOutputSize = in1.estimatedOutputSize 0 in2.estimatedOutputSize 0 ? in1.estimatedOutputSize + in2.estimatedOutputSize : -1; } + + public static class UnionSemanticProperties implements SemanticProperties { + + @Override + public FieldSet getTargetFields(int input, int sourceField) { + if (input != 0 input != 1) { + throw new IndexOutOfBoundsException(); --- End diff -- That's an internal exception. If something fails here, there is something wrong with the optimizer. Nothing a user can solve. Could make that clear though... --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user rmetzger commented on a diff in the pull request: https://github.com/apache/flink/pull/311#discussion_r23070438 --- Diff: flink-java/src/main/java/org/apache/flink/api/java/typeutils/PojoTypeInfo.java --- @@ -45,6 +46,16 @@ */ public class PojoTypeInfoT extends CompositeTypeT{ + private final static String REGEX_FIELD = [a-zA-Z_\\$][a-zA-Z0-9_\\$]*; --- End diff -- Java allows to use any unicode character to be used as field names. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1183] Generate gentle notification mess...
Github user asfgit closed the pull request at: https://github.com/apache/flink/pull/296 --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Commented] (FLINK-1409) Connected datastream functionality broken since the introduction of intermediate results
[ https://issues.apache.org/jira/browse/FLINK-1409?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280035#comment-14280035 ] Ufuk Celebi commented on FLINK-1409: I'll add a test case and look into it right now. Connected datastream functionality broken since the introduction of intermediate results Key: FLINK-1409 URL: https://issues.apache.org/jira/browse/FLINK-1409 Project: Flink Issue Type: Bug Components: Streaming Affects Versions: 0.9 Reporter: Gyula Fora The connected data stream functionality which allows joint transformations on two data streams of arbitrary type is broken since Ufuk's commit which introduces the intermediate results. The problem is most likely in the CoRecordReader which should allow nonblocking read from inputs with different types. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user rmetzger commented on a diff in the pull request: https://github.com/apache/flink/pull/311#discussion_r23070841 --- Diff: flink-compiler/src/main/java/org/apache/flink/compiler/dag/BinaryUnionNode.java --- @@ -266,4 +268,44 @@ public void computeOutputEstimates(DataStatistics statistics) { this.estimatedOutputSize = in1.estimatedOutputSize 0 in2.estimatedOutputSize 0 ? in1.estimatedOutputSize + in2.estimatedOutputSize : -1; } + + public static class UnionSemanticProperties implements SemanticProperties { + + @Override + public FieldSet getTargetFields(int input, int sourceField) { + if (input != 0 input != 1) { + throw new IndexOutOfBoundsException(); --- End diff -- Ah, okay. Then its fine. I saw many helpful exceptions in this change. So I guess the user-facing exceptions are more descriptive. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Commented] (FLINK-1379) add RSS feed for the blog
[ https://issues.apache.org/jira/browse/FLINK-1379?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280036#comment-14280036 ] Max Michels commented on FLINK-1379: Thank you for the hint, Robert. I guess I'm too much rooted in the GitHub world... add RSS feed for the blog - Key: FLINK-1379 URL: https://issues.apache.org/jira/browse/FLINK-1379 Project: Flink Issue Type: Improvement Components: Project Website Reporter: Max Michels Assignee: Max Michels Priority: Minor Attachments: feed.patch I couldn't find an RSS feed for the Flink blog. I think that a feed helps a lot of people to stay up to date with the changes in Flink. [FLINK-391] mentions a RSS feed but it does not seem to exist. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user rmetzger commented on the pull request: https://github.com/apache/flink/pull/311#issuecomment-70230559 Except for the comments and the missing documentation, the change looks good. I can however not really validate the changes in the optimizer. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1395] Add support for JodaTime in KryoS...
Github user rmetzger commented on a diff in the pull request: https://github.com/apache/flink/pull/304#discussion_r23071236 --- Diff: flink-java/src/main/java/org/apache/flink/api/java/typeutils/runtime/KryoSerializer.java --- @@ -185,6 +187,8 @@ private void checkKryoInitialized() { this.kryo.setRegistrationRequired(false); this.kryo.register(type); this.kryo.setClassLoader(Thread.currentThread().getContextClassLoader()); + + kryo.register(DateTime.class, new JodaDateTimeSerializer()); --- End diff -- I would suggest to add some more serializers from `de.javakaffee` .. since we have it already as a dependency, it doesn't hurt to add them. I'm suggesting - `jodatime/JodaIntervalSerializer`, - `guava/ImmutableListSerializer`, - `UnmodifiableCollectionsSerializer`, - `GregorianCalendarSerializer`, - `EnumSetSerializer`, - `EnumMapSerializer`, - BitSetSerializer - serializer for java.util.BitSet - RegexSerializer - serializer for java.util.regex.Pattern - URISerializer - serializer for java.net.URI - UUIDSerializer - serializer for java.util.UUID --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1395] Add support for JodaTime in KryoS...
Github user rmetzger commented on the pull request: https://github.com/apache/flink/pull/304#issuecomment-70231876 Change looks good except for comments. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1406] update Flink compatibility notice
GitHub user mxm opened a pull request: https://github.com/apache/flink/pull/314 [FLINK-1406] update Flink compatibility notice You can merge this pull request into a Git repository by running: $ git pull https://github.com/mxm/flink flink_1406 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/flink/pull/314.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #314 commit b09ab9f644bb21737babddf387045ff953f72135 Author: Max m...@posteo.de Date: 2015-01-16T10:02:47Z [FLINK-1406] update Flink compatibility notice --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1406] update Flink compatibility notice
Github user rmetzger commented on the pull request: https://github.com/apache/flink/pull/314#issuecomment-70234048 Good to merge. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Updated] (FLINK-1406) Windows compatibility
[ https://issues.apache.org/jira/browse/FLINK-1406?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Max Michels updated FLINK-1406: --- Attachment: flink_1406.patch Here is the patch for the how to contribute guide in addition to the pull request. Windows compatibility - Key: FLINK-1406 URL: https://issues.apache.org/jira/browse/FLINK-1406 Project: Flink Issue Type: Improvement Reporter: Max Michels Priority: Minor Attachments: flink_1406.patch The documentation [1] states: Flink runs on all UNIX-like environments: Linux, Mac OS X, Cygwin. The only requirement is to have a working Java 6.x (or higher) installation. I just found out Flink runs also natively on Windows. Do we want to support Windows? If so, we should update the documentation. Clearly, we don't support it at the moment for development. At multiple places, the tests contain references to /tmp or /dev/random/ which are not Windows compatible. Probably it's enough to update the documentation stating that Flink runs on Windows but cannot be build or developed on Windows without Cygwin. [1] http://flink.apache.org/docs/0.7-incubating/setup_quickstart.html -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1389] Allow changing the filenames of t...
Github user StephanEwen commented on the pull request: https://github.com/apache/flink/pull/301#issuecomment-70234744 BTW: How is that done in Hadoop? Can we follow a similar way, to make it easier for users to understand this? --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user fhueske commented on a diff in the pull request: https://github.com/apache/flink/pull/311#discussion_r23073054 --- Diff: flink-java/src/main/java/org/apache/flink/api/java/typeutils/PojoTypeInfo.java --- @@ -45,6 +46,16 @@ */ public class PojoTypeInfoT extends CompositeTypeT{ + private final static String REGEX_FIELD = [a-zA-Z_\\$][a-zA-Z0-9_\\$]*; --- End diff -- Right! Will change that. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1328] Reworked semantic annotations
Github user fhueske commented on the pull request: https://github.com/apache/flink/pull/311#issuecomment-70236149 Thanks for the review! Proposed names for constant field semantic properties: * constant fields (current) * unmodified fields * forwarded fields * preserved fields I leaning towards forwarded fields. Other opinions? --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1147][Java API] TypeInference on POJOs
GitHub user twalthr opened a pull request: https://github.com/apache/flink/pull/315 [FLINK-1147][Java API] TypeInference on POJOs The TypeExtractor now also fully supports generic POJOs and tries to get missing types by using type inference. Functions can look like: ``` MapFunctionPojoWithGenericsLong, T, PojoWithGenericsT,T MapFunctionTuple2E, D, PojoTupleE, D, D MapFunctionPojoTupleE, D, D, Tuple2E, D ``` POJOs can contain fields such as: ``` public PojoWithGenericsZ, Z field; public Z[] field; public Tuple1Z[] field; ``` If you have ideas for other test cases, I'm happy to implement them. You can merge this pull request into a Git repository by running: $ git pull https://github.com/twalthr/flink GenericPojos Alternatively you can review and apply these changes as the patch at: https://github.com/apache/flink/pull/315.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #315 commit 391cc8fde82a38797c0355031f156f8914845f1b Author: twalthr i...@twalthr.com Date: 2015-01-13T22:59:35Z [FLINK-1147][Java API] TypeInference on POJOs --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Commented] (FLINK-1409) Connected datastream functionality broken since the introduction of intermediate results
[ https://issues.apache.org/jira/browse/FLINK-1409?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280224#comment-14280224 ] Gyula Fora commented on FLINK-1409: --- Thanks, let me know if I can help in any way. Connected datastream functionality broken since the introduction of intermediate results Key: FLINK-1409 URL: https://issues.apache.org/jira/browse/FLINK-1409 Project: Flink Issue Type: Bug Components: Streaming Affects Versions: 0.9 Reporter: Gyula Fora The connected data stream functionality which allows joint transformations on two data streams of arbitrary type is broken since Ufuk's commit which introduces the intermediate results. The problem is most likely in the CoRecordReader which should allow nonblocking read from inputs with different types. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1382][java] Adds the new basic types Vo...
Github user twalthr commented on the pull request: https://github.com/apache/flink/pull/299#issuecomment-70255146 There are test cases. I have adapted the existing ones. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Commented] (FLINK-1409) Connected datastream functionality broken since the introduction of intermediate results
[ https://issues.apache.org/jira/browse/FLINK-1409?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280231#comment-14280231 ] Ufuk Celebi commented on FLINK-1409: Thanks :-) I've just looked into it and I know what it is. Connected datastream functionality broken since the introduction of intermediate results Key: FLINK-1409 URL: https://issues.apache.org/jira/browse/FLINK-1409 Project: Flink Issue Type: Bug Components: Streaming Affects Versions: 0.9 Reporter: Gyula Fora The connected data stream functionality which allows joint transformations on two data streams of arbitrary type is broken since Ufuk's commit which introduces the intermediate results. The problem is most likely in the CoRecordReader which should allow nonblocking read from inputs with different types. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: Implement the convenience methods count and co...
Github user StephanEwen commented on a diff in the pull request: https://github.com/apache/flink/pull/210#discussion_r23080920 --- Diff: flink-core/src/main/java/org/apache/flink/api/common/accumulators/ListAccumulator.java --- @@ -0,0 +1,97 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * License); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an AS IS BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + + +package org.apache.flink.api.common.accumulators; + +import java.io.IOException; +import java.io.ObjectInputStream; +import java.io.ObjectOutputStream; +import java.util.ArrayList; +import java.util.List; + +import org.apache.flink.api.common.typeutils.TypeSerializer; +import org.apache.flink.core.memory.DataInputView; +import org.apache.flink.core.memory.DataInputViewStream; +import org.apache.flink.core.memory.DataOutputView; +import org.apache.flink.core.memory.DataOutputViewStream; + +public class ListAccumulatorT implements AccumulatorT, ListT { + + private static final long serialVersionUID = 1L; + + private ArrayListT localValue = new ArrayListT(); + + private TypeSerializerT typeSerializer; + + public ListAccumulator() {} + + public ListAccumulator(TypeSerializerT ser) { + typeSerializer = ser; + } + + @Override + public void add(T value) { + localValue.add(value); + } + + @Override + public ArrayListT getLocalValue() { + return localValue; + } + + @Override + public void resetLocal() { + localValue.clear(); + } + + @Override + public void merge(AccumulatorT, ListT other) { + localValue.addAll(((ListAccumulatorT) other).getLocalValue()); + } + + @Override + public void write(DataOutputView out) throws IOException { + ObjectOutputStream outStream = new ObjectOutputStream(new DataOutputViewStream(out)); + outStream.writeObject(typeSerializer); + outStream.flush(); --- End diff -- Let us change this to `close()` --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: Implement the convenience methods count and co...
Github user StephanEwen commented on a diff in the pull request: https://github.com/apache/flink/pull/210#discussion_r23081185 --- Diff: flink-runtime/src/test/java/org/apache/flink/runtime/AbstractIDTest.java --- @@ -23,8 +23,8 @@ import static org.junit.Assert.fail; import org.junit.Test; - import org.apache.flink.runtime.testutils.CommonTestUtils; +import org.apache.flink.util.AbstractID; --- End diff -- Let us move this test to `flink-core`, with `AbstractID` being now part of the core. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1296] Add sorter support for very large...
Github user StephanEwen commented on the pull request: https://github.com/apache/flink/pull/249#issuecomment-70257415 I think that this one can go into the 0.9 master now. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: Implement the convenience methods count and co...
Github user StephanEwen commented on the pull request: https://github.com/apache/flink/pull/210#issuecomment-70258735 With the scheduler and intermediate data set enhancements coming up for 0.9 soon, this is now quite feasible to use. I suggest to merge it once the inline comments are addressed. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Commented] (FLINK-1098) flatArray() operator that converts arrays to elements
[ https://issues.apache.org/jira/browse/FLINK-1098?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280297#comment-14280297 ] Timo Walther commented on FLINK-1098: - What do you think about a additional method collectEach() in Collector. I know that is just syntactic sugar, but then we could write: {code} text .flatMap((line, out) - out.collectEach(line.toLowerCase().split(\\W+))) .map((word) - new Tuple2(word, 1)) .groupBy(0) .sum(1); {code} flatArray() operator that converts arrays to elements - Key: FLINK-1098 URL: https://issues.apache.org/jira/browse/FLINK-1098 Project: Flink Issue Type: New Feature Reporter: Timo Walther Priority: Minor It would be great to have an operator that converts e.g. from String[] to String. Actually, it is just a flatMap over the elements of an array. A typical use case is a WordCount where we then could write: {code} text .map((line) - line.toLowerCase().split(\\W+)) .flatArray() .map((word) - new Tuple2(word, 1)) .groupBy(0) .sum(1); {code} -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (FLINK-1410) Integrate Flink version variables into website layout
Robert Metzger created FLINK-1410: - Summary: Integrate Flink version variables into website layout Key: FLINK-1410 URL: https://issues.apache.org/jira/browse/FLINK-1410 Project: Flink Issue Type: Bug Components: Project Website Reporter: Robert Metzger The new website layout doesn't use the variables in the website configuration. This makes releasing versions extremely hard, because one needs to manually fix all the links for every version change. The old layout of the website was respecting all variables which made releasing a new version of the website a matter of minutes (changing one file). I would highly recommend to fix FLINK-1387 first. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1382][java] Adds the new basic types Vo...
Github user rmetzger commented on the pull request: https://github.com/apache/flink/pull/299#issuecomment-70269761 Oh, yes .. sorry. I need to be more careful when reviewing pull requests. +1 to merge this. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Created] (FLINK-1411) PlanVisualizer is not working
Till Rohrmann created FLINK-1411: Summary: PlanVisualizer is not working Key: FLINK-1411 URL: https://issues.apache.org/jira/browse/FLINK-1411 Project: Flink Issue Type: Bug Reporter: Till Rohrmann Priority: Minor In the current master, the PlanVisualizer is no longer working. The reason is that the resources folder containing the web resources has been moved to the flink-runtime and flink-clients jar. Maybe we should pick up FLINK-1317 and make the PlanVisualizer accessible through the flink website. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (FLINK-655) Rename DataSet.withBroadcastSet(DataSet, String) method
[ https://issues.apache.org/jira/browse/FLINK-655?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14279683#comment-14279683 ] Henry Saputra edited comment on FLINK-655 at 1/16/15 4:36 PM: -- Currently, the getBroadcastVariable returns List instance instead of just an instance. With this refactor, do we want to change the return type too? was (Author: hsaputra): Currently, the getBroadcastVariable returns List instance of just an instance. With this refactor, do we want to change the return type too? Rename DataSet.withBroadcastSet(DataSet, String) method --- Key: FLINK-655 URL: https://issues.apache.org/jira/browse/FLINK-655 Project: Flink Issue Type: Improvement Components: Java API Reporter: Ufuk Celebi Assignee: Henry Saputra Labels: breaking-api, github-import, starter Fix For: pre-apache To broadcast a data set you have to do the following: ```java lorem.map(new MyMapper()).withBroadcastSet(toBroadcast, toBroadcastName) ``` In the operator you call: ```java getRuntimeContext().getBroadcastVariable(toBroadcastName) ``` I propose to have both method names consistent, e.g. - `withBroadcastVariable(DataSet, String)`, or - `getBroadcastSet(String)`. Imported from GitHub Url: https://github.com/stratosphere/stratosphere/issues/655 Created by: [uce|https://github.com/uce] Labels: enhancement, java api, user satisfaction, Created at: Wed Apr 02 16:29:08 CEST 2014 State: open -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: Update incubator-flink name in the merge pull ...
Github user hsaputra commented on the pull request: https://github.com/apache/flink/pull/313#issuecomment-70284303 Thanks @rmetzger, will merge this today. Not a blocker for 0.8 so I will not merge it to 0.8 branch. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[jira] [Commented] (FLINK-655) Add support for both single and set of broadcast values
[ https://issues.apache.org/jira/browse/FLINK-655?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14280590#comment-14280590 ] Henry Saputra commented on FLINK-655: - I change the summary to reflect the new task to do for this JIRA Add support for both single and set of broadcast values --- Key: FLINK-655 URL: https://issues.apache.org/jira/browse/FLINK-655 Project: Flink Issue Type: Improvement Components: Java API Reporter: Ufuk Celebi Assignee: Henry Saputra Labels: breaking-api, github-import, starter Fix For: pre-apache To broadcast a data set you have to do the following: ```java lorem.map(new MyMapper()).withBroadcastSet(toBroadcast, toBroadcastName) ``` In the operator you call: ```java getRuntimeContext().getBroadcastVariable(toBroadcastName) ``` I propose to have both method names consistent, e.g. - `withBroadcastVariable(DataSet, String)`, or - `getBroadcastSet(String)`. Imported from GitHub Url: https://github.com/stratosphere/stratosphere/issues/655 Created by: [uce|https://github.com/uce] Labels: enhancement, java api, user satisfaction, Created at: Wed Apr 02 16:29:08 CEST 2014 State: open -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (FLINK-655) Add support for both single and set of broadcast values
[ https://issues.apache.org/jira/browse/FLINK-655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Henry Saputra updated FLINK-655: Summary: Add support for both single and set of broadcast values (was: Rename DataSet.withBroadcastSet(DataSet, String) method) Add support for both single and set of broadcast values --- Key: FLINK-655 URL: https://issues.apache.org/jira/browse/FLINK-655 Project: Flink Issue Type: Improvement Components: Java API Reporter: Ufuk Celebi Assignee: Henry Saputra Labels: breaking-api, github-import, starter Fix For: pre-apache To broadcast a data set you have to do the following: ```java lorem.map(new MyMapper()).withBroadcastSet(toBroadcast, toBroadcastName) ``` In the operator you call: ```java getRuntimeContext().getBroadcastVariable(toBroadcastName) ``` I propose to have both method names consistent, e.g. - `withBroadcastVariable(DataSet, String)`, or - `getBroadcastSet(String)`. Imported from GitHub Url: https://github.com/stratosphere/stratosphere/issues/655 Created by: [uce|https://github.com/uce] Labels: enhancement, java api, user satisfaction, Created at: Wed Apr 02 16:29:08 CEST 2014 State: open -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[GitHub] flink pull request: [FLINK-1369] [types] Add support for Subclasse...
GitHub user fhueske opened a pull request: https://github.com/apache/flink/pull/316 [FLINK-1369] [types] Add support for Subclasses, Interfaces, Abstract Classes This PR rebased PR #236 to the current master. Some tests were failing and I had a closer look. The original PR handled interfaces and abstract classes without member variables as POJO types. However, POJO types without members cannot be handled correctly and do also not have members that can be referenced as keys or fields. I changed the logic such that interfaces and abstract classes without members are handled as GenericTypes. I'm not so familiar with the TypeExtractor, so it would be good if someone else could have a quick look. You can merge this pull request into a Git repository by running: $ git pull https://github.com/fhueske/flink aljoscha-subclass-types Alternatively you can review and apply these changes as the patch at: https://github.com/apache/flink/pull/316.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #316 commit 8f20c0df045db820f69aa18ced8658055e131081 Author: Aljoscha Krettek aljoscha.kret...@gmail.com Date: 2014-11-26T12:27:06Z [FLINK-1369] [types] Add support for Subclasses, Interfaces, Abstract Classes. - Abstract classes with fields are handled as POJO types. - Interfaces and abstract classes without fields are handled as generic types. This closes #236 This closes #316 --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] flink pull request: [FLINK-1369] [types] Add support for Subclasse...
Github user twalthr commented on the pull request: https://github.com/apache/flink/pull/316#issuecomment-70348260 I will take a look at it. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---