You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hudi.apache.org by GitBox <gi...@apache.org> on 2020/04/26 20:15:28 UTC

[GitHub] [incubator-hudi] pratyakshsharma opened a new pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

pratyakshsharma opened a new pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566


   ## *Tips*
   - *Thank you very much for contributing to Apache Hudi.*
   - *Please review https://hudi.apache.org/contributing.html before opening a pull request.*
   
   ## What is the purpose of the pull request
   
   HoodieDeltaStreamer create SchemaProvider instance and delegate to DeltaSync for periodical sync. However, default implementation of SchemaProvider does not refresh schema, which can change due to schema evolution. DeltaSync snapshot the schema when it creates writeClient, using the SchemaProvider instance or pick up from source, and the schema for writeClient is not refreshed during the loop of Sync.
   
   ## Brief change log
   
   - Introduced a function to refresh schema after every run of DeltaSync.
   
   ## Verify this pull request
   
   *(Please pick either of the following options)*
   
   This pull request is a trivial rework / code cleanup without any test coverage.
   
   *(or)*
   
   This pull request is already covered by existing tests, such as *(please describe tests)*.
   
   (or)
   
   This change added tests and can be verified as follows:
   
   *(example:)*
   
     - *Added integration tests for end-to-end.*
     - *Added HoodieClientWriteTest to verify the change.*
     - *Manually verified the change by running a job locally.*
   
   ## Committer checklist
   
    - [ ] Has a corresponding JIRA in PR title & commit
    
    - [ ] Commit message is descriptive of the change
    
    - [ ] CI is green
   
    - [ ] Necessary doc changes done or have another open PR
          
    - [ ] For large changes, please consider breaking it into sub-tasks under an umbrella JIRA.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] vinothchandar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
vinothchandar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-624114147


   @pratyakshsharma  @afilipchik IIUC using the Confluent Avro Kafka decoders etc will integrate with SR and fetch and decode using the latest schema for us, which we will use as the schema for the write as well... There is another PR tracking this/fixing this.. (a lot of these schema PR interplay quite a bit :))
   
   
   On the initial suggestion, @pratyakshsharma I was merely suggesting a better contract for `SchemaProvider` where `getSourceSchema()` is support to return the latest source schema as of that time, not a cached copy based on what was fetched in the constructor.. Existing schema providers do a mix of these.. 
   
   Filebased/JdbcBased fetch the schema once in constructor and keep serving.. whereas SchemaRegistry/RowBased fetch again when `getSourceSchema()` is called.. So no need to create the schemaRegistryProvider instance every run, simply call `getSourceSchema()` every run? 
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426184790



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaSet.java
##########
@@ -0,0 +1,44 @@
+/*
+ * 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.hudi.utilities.schema;
+
+import java.io.Serializable;
+import java.util.HashSet;
+import org.apache.avro.Schema;
+import org.apache.avro.SchemaNormalization;
+
+import java.util.Set;
+
+/**
+ * Tracks already processed schemas.
+ */
+public class SchemaSet implements Serializable {
+
+  private final Set<Long> processedSchema = new HashSet<>();

Review comment:
       Personally I feel you do not have many schema evolutions in production environment. So not removing old schemas should not create any issues. @bvaradar to take the final call. 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426185251



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaRegistryProvider.java
##########
@@ -81,11 +66,22 @@ private static Schema getSchema(String registryUrl) throws IOException {
 
   @Override
   public Schema getSourceSchema() {
-    return schema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);
+    try {
+      return getSchema(registryUrl);
+    } catch (IOException ioe) {
+      throw new HoodieIOException("Error reading source schema from registry :" + registryUrl, ioe);
+    }
   }
 
   @Override
   public Schema getTargetSchema() {
-    return targetSchema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);
+    String targetRegistryUrl = config.getString(Config.TARGET_SCHEMA_REGISTRY_URL_PROP, registryUrl);
+    try {
+      return getSchema(targetRegistryUrl);

Review comment:
       Should not be a problem I believe since sourceSchema is used for reading while targetSchema is used for writing. Any issues if at all should be handled in HoodieAvroUtils when we try to rewrite the record with targetSchema. 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] codecov-io commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
codecov-io commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-619623233


   # [Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=h1) Report
   > Merging [#1566](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=desc) into [master](https://codecov.io/gh/apache/incubator-hudi/commit/19ca0b56296bf89c611406bb86c3626f87d562ee&el=desc) will **increase** coverage by `0.03%`.
   > The diff coverage is `100.00%`.
   
   [![Impacted file tree graph](https://codecov.io/gh/apache/incubator-hudi/pull/1566/graphs/tree.svg?width=650&height=150&src=pr&token=VTTXabwbs2)](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=tree)
   
   ```diff
   @@             Coverage Diff              @@
   ##             master    #1566      +/-   ##
   ============================================
   + Coverage     71.82%   71.86%   +0.03%     
   - Complexity      294      296       +2     
   ============================================
     Files           383      383              
     Lines         16549    16555       +6     
     Branches       1663     1664       +1     
   ============================================
   + Hits          11887    11897      +10     
   + Misses         3930     3927       -3     
   + Partials        732      731       -1     
   ```
   
   
   | [Impacted Files](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=tree) | Coverage Δ | Complexity Δ | |
   |---|---|---|---|
   | [...apache/hudi/utilities/deltastreamer/DeltaSync.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvRGVsdGFTeW5jLmphdmE=) | `73.23% <100.00%> (+0.78%)` | `39.00 <1.00> (+2.00)` | |
   | [...i/utilities/deltastreamer/HoodieDeltaStreamer.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvSG9vZGllRGVsdGFTdHJlYW1lci5qYXZh) | `80.37% <100.00%> (+0.37%)` | `11.00 <0.00> (ø)` | |
   | [...ache/hudi/common/fs/inline/InMemoryFileSystem.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS1jb21tb24vc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvY29tbW9uL2ZzL2lubGluZS9Jbk1lbW9yeUZpbGVTeXN0ZW0uamF2YQ==) | `89.65% <0.00%> (+10.34%)` | `0.00% <0.00%> (ø%)` | |
   
   ------
   
   [Continue to review full report at Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=continue).
   > **Legend** - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta)
   > `Δ = absolute <relative> (impact)`, `ø = not affected`, `? = missing data`
   > Powered by [Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=footer). Last update [19ca0b5...dfaa70d](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=lastupdated). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments).
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426184828



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/HoodieDeltaStreamer.java
##########
@@ -464,21 +464,25 @@ private void shutdownCompactor(boolean error) {
      */
     protected Boolean onInitializingWriteClient(HoodieWriteClient writeClient) {
       if (cfg.isAsyncCompactionEnabled()) {
-        asyncCompactService = new AsyncCompactService(jssc, writeClient);
-        // Enqueue existing pending compactions first
-        HoodieTableMetaClient meta =
-            new HoodieTableMetaClient(new Configuration(jssc.hadoopConfiguration()), cfg.targetBasePath, true);
-        List<HoodieInstant> pending = CompactionUtils.getPendingCompactionInstantTimes(meta);
-        pending.forEach(hoodieInstant -> asyncCompactService.enqueuePendingCompaction(hoodieInstant));
-        asyncCompactService.start((error) -> {
-          // Shutdown DeltaSync
-          shutdown(false);
-          return true;
-        });
-        try {
-          asyncCompactService.waitTillPendingCompactionsReducesTo(cfg.maxPendingCompactions);
-        } catch (InterruptedException ie) {
-          throw new HoodieException(ie);
+        if (null != asyncCompactService) {

Review comment:
       +1




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-629443388


   @pratyakshsharma : Just rebased and did some cleanup.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426187093



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaRegistryProvider.java
##########
@@ -81,11 +66,22 @@ private static Schema getSchema(String registryUrl) throws IOException {
 
   @Override
   public Schema getSourceSchema() {
-    return schema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);

Review comment:
       It is already handled in the code, in case different schema is returned, writeClient is recreated and new avro schemas are registered with spark configuration.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] codecov-io edited a comment on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
codecov-io edited a comment on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-619623233


   # [Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=h1) Report
   > Merging [#1566](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=desc) into [master](https://codecov.io/gh/apache/incubator-hudi/commit/19ca0b56296bf89c611406bb86c3626f87d562ee&el=desc) will **increase** coverage by `0.03%`.
   > The diff coverage is `100.00%`.
   
   [![Impacted file tree graph](https://codecov.io/gh/apache/incubator-hudi/pull/1566/graphs/tree.svg?width=650&height=150&src=pr&token=VTTXabwbs2)](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=tree)
   
   ```diff
   @@             Coverage Diff              @@
   ##             master    #1566      +/-   ##
   ============================================
   + Coverage     71.82%   71.86%   +0.03%     
   - Complexity      294      296       +2     
   ============================================
     Files           383      383              
     Lines         16549    16555       +6     
     Branches       1663     1664       +1     
   ============================================
   + Hits          11887    11897      +10     
   + Misses         3930     3927       -3     
   + Partials        732      731       -1     
   ```
   
   
   | [Impacted Files](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=tree) | Coverage Δ | Complexity Δ | |
   |---|---|---|---|
   | [...apache/hudi/utilities/deltastreamer/DeltaSync.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvRGVsdGFTeW5jLmphdmE=) | `73.23% <100.00%> (+0.78%)` | `39.00 <1.00> (+2.00)` | |
   | [...i/utilities/deltastreamer/HoodieDeltaStreamer.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvSG9vZGllRGVsdGFTdHJlYW1lci5qYXZh) | `80.37% <100.00%> (+0.37%)` | `11.00 <0.00> (ø)` | |
   | [...ache/hudi/common/fs/inline/InMemoryFileSystem.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS1jb21tb24vc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvY29tbW9uL2ZzL2lubGluZS9Jbk1lbW9yeUZpbGVTeXN0ZW0uamF2YQ==) | `89.65% <0.00%> (+10.34%)` | `0.00% <0.00%> (ø%)` | |
   
   ------
   
   [Continue to review full report at Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=continue).
   > **Legend** - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta)
   > `Δ = absolute <relative> (impact)`, `ø = not affected`, `? = missing data`
   > Powered by [Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=footer). Last update [19ca0b5...dfaa70d](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=lastupdated). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments).
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] vinothchandar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
vinothchandar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-620199568


   @afilipchik interested in taking a run at this? :) 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-620679844


   > if we redid the schema provider implementations such that the schema is read each time from SR (schema registry)
   
   Do you have some plan around how to achieve this? @vinothchandar 
   
   I was thinking of maintaining the subject and version information from the very beginning with us (which should be easy to get from the schema-registry url provided by the user), and then compare the same by fetching the serialized schema id from incoming messages. Whenever we experience there is some version update, we can refresh the schemas at that point and continue consuming the incoming messages. This would need writing our own custom AvroDeserializer. 
   
   Also the above plan only works for the combination of AvroKafkaSource and SchemaRegistryProvider. Thoughts? 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] afilipchik commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
afilipchik commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426074416



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaSet.java
##########
@@ -0,0 +1,44 @@
+/*
+ * 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.hudi.utilities.schema;
+
+import java.io.Serializable;
+import java.util.HashSet;
+import org.apache.avro.Schema;
+import org.apache.avro.SchemaNormalization;
+
+import java.util.Set;
+
+/**
+ * Tracks already processed schemas.
+ */
+public class SchemaSet implements Serializable {
+

Review comment:
       should we add serialVersionUID? If it is not specified and anything in the class imports is shaded -> it will affect autogenerated one. Which causes issues if there is more than 1 version of the class in the classpath which was shaded differently. 

##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaRegistryProvider.java
##########
@@ -81,11 +66,22 @@ private static Schema getSchema(String registryUrl) throws IOException {
 
   @Override
   public Schema getSourceSchema() {
-    return schema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);
+    try {
+      return getSchema(registryUrl);
+    } catch (IOException ioe) {
+      throw new HoodieIOException("Error reading source schema from registry :" + registryUrl, ioe);
+    }
   }
 
   @Override
   public Schema getTargetSchema() {
-    return targetSchema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);
+    String targetRegistryUrl = config.getString(Config.TARGET_SCHEMA_REGISTRY_URL_PROP, registryUrl);
+    try {
+      return getSchema(targetRegistryUrl);

Review comment:
       it might result in target schema != source schema when targetRegistryUrl is not specified as schema might change between getSourceSchema, getTargetSchema calls. Is it a problem? 

##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/Compactor.java
##########
@@ -59,6 +60,10 @@ public void compact(HoodieInstant instant) throws IOException {
           "Compaction for instant (" + instant + ") failed with write errors. Errors :" + numWriteErrors);
     }
     // Commit compaction
-    compactionClient.commitCompaction(instant.getTimestamp(), res, Option.empty());
+    writeClient.commitCompaction(instant.getTimestamp(), res, Option.empty());
+  }
+
+  public void updateWriteClient(HoodieWriteClient writeClient) {

Review comment:
       is it used anywhere? 

##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaSet.java
##########
@@ -0,0 +1,44 @@
+/*
+ * 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.hudi.utilities.schema;
+
+import java.io.Serializable;
+import java.util.HashSet;
+import org.apache.avro.Schema;
+import org.apache.avro.SchemaNormalization;
+
+import java.util.Set;
+
+/**
+ * Tracks already processed schemas.
+ */
+public class SchemaSet implements Serializable {
+
+  private final Set<Long> processedSchema = new HashSet<>();

Review comment:
       will this grow indefinitely? How would we remove old schema? 

##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaRegistryProvider.java
##########
@@ -81,11 +66,22 @@ private static Schema getSchema(String registryUrl) throws IOException {
 
   @Override
   public Schema getSourceSchema() {
-    return schema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);

Review comment:
       is it called only once during a run? Will it be an issue if it is called more than once and slightly different schema is returned? 

##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/HoodieDeltaStreamer.java
##########
@@ -464,21 +464,25 @@ private void shutdownCompactor(boolean error) {
      */
     protected Boolean onInitializingWriteClient(HoodieWriteClient writeClient) {
       if (cfg.isAsyncCompactionEnabled()) {
-        asyncCompactService = new AsyncCompactService(jssc, writeClient);
-        // Enqueue existing pending compactions first
-        HoodieTableMetaClient meta =
-            new HoodieTableMetaClient(new Configuration(jssc.hadoopConfiguration()), cfg.targetBasePath, true);
-        List<HoodieInstant> pending = CompactionUtils.getPendingCompactionInstantTimes(meta);
-        pending.forEach(hoodieInstant -> asyncCompactService.enqueuePendingCompaction(hoodieInstant));
-        asyncCompactService.start((error) -> {
-          // Shutdown DeltaSync
-          shutdown(false);
-          return true;
-        });
-        try {
-          asyncCompactService.waitTillPendingCompactionsReducesTo(cfg.maxPendingCompactions);
-        } catch (InterruptedException ie) {
-          throw new HoodieException(ie);
+        if (null != asyncCompactService) {

Review comment:
       would be great to have some documentation on why it is done this way.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] afilipchik commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
afilipchik commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-620394679


   I think it is not enough. Example: when consuming from kafka, schema might change midway. Example: we are reading 10000 messages, schema will be fetched on start. Starting from 9000 message new column is added, but we will not see it. As a result we just lost data. 
   
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma edited a comment on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma edited a comment on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-620661941


   > I think it is not enough. Example: when consuming from kafka, schema might change midway. Example: we are reading 10000 messages, schema will be fetched on start. Starting from 9000 message new column is added, but we will not see it. As a result we just lost data.
   
   Yeah true but that was not the purpose of this PR :)


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-620661941


   > I think it is not enough. Example: when consuming from kafka, schema might change midway. Example: we are reading 10000 messages, schema will be fetched on start. Starting from 9000 message new column is added, but we will not see it. As a result we just lost data.
   
   Yeah true. that was not the purpose of this PR but I guess I should include it as well :)


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426187139



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaSet.java
##########
@@ -0,0 +1,44 @@
+/*
+ * 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.hudi.utilities.schema;
+
+import java.io.Serializable;
+import java.util.HashSet;
+import org.apache.avro.Schema;
+import org.apache.avro.SchemaNormalization;
+
+import java.util.Set;
+
+/**
+ * Tracks already processed schemas.
+ */
+public class SchemaSet implements Serializable {
+

Review comment:
       @bvaradar to chime in here. 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] codecov-io edited a comment on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
codecov-io edited a comment on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-619623233


   # [Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=h1) Report
   > Merging [#1566](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=desc) into [master](https://codecov.io/gh/apache/incubator-hudi/commit/83796b3189570182c68a9c41e57b356124c301ca&el=desc) will **decrease** coverage by `0.01%`.
   > The diff coverage is `69.81%`.
   
   [![Impacted file tree graph](https://codecov.io/gh/apache/incubator-hudi/pull/1566/graphs/tree.svg?width=650&height=150&src=pr&token=VTTXabwbs2)](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=tree)
   
   ```diff
   @@             Coverage Diff              @@
   ##             master    #1566      +/-   ##
   ============================================
   - Coverage     71.80%   71.79%   -0.02%     
   - Complexity     1087     1095       +8     
   ============================================
     Files           385      387       +2     
     Lines         16591    16645      +54     
     Branches       1669     1675       +6     
   ============================================
   + Hits          11913    11950      +37     
   - Misses         3949     3962      +13     
   - Partials        729      733       +4     
   ```
   
   
   | [Impacted Files](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=tree) | Coverage Δ | Complexity Δ | |
   |---|---|---|---|
   | [.../hudi/utilities/schema/SchemaRegistryProvider.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL3NjaGVtYS9TY2hlbWFSZWdpc3RyeVByb3ZpZGVyLmphdmE=) | `0.00% <0.00%> (ø)` | `0.00 <0.00> (ø)` | |
   | [...a/org/apache/hudi/client/AbstractHoodieClient.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS1jbGllbnQvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvY2xpZW50L0Fic3RyYWN0SG9vZGllQ2xpZW50LmphdmE=) | `75.00% <50.00%> (-3.95%)` | `6.00 <0.00> (ø)` | |
   | [...i/utilities/keygen/TimestampBasedKeyGenerator.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2tleWdlbi9UaW1lc3RhbXBCYXNlZEtleUdlbmVyYXRvci5qYXZh) | `58.82% <61.11%> (-0.16%)` | `7.00 <1.00> (+2.00)` | :arrow_down: |
   | [...i/utilities/deltastreamer/HoodieDeltaStreamer.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvSG9vZGllRGVsdGFTdHJlYW1lci5qYXZh) | `78.89% <70.00%> (-1.11%)` | `11.00 <0.00> (ø)` | |
   | [...apache/hudi/utilities/deltastreamer/DeltaSync.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvRGVsdGFTeW5jLmphdmE=) | `72.93% <76.92%> (+0.48%)` | `40.00 <2.00> (+3.00)` | |
   | [.../client/embedded/EmbeddedTimelineServerHelper.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS1jbGllbnQvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvY2xpZW50L2VtYmVkZGVkL0VtYmVkZGVkVGltZWxpbmVTZXJ2ZXJIZWxwZXIuamF2YQ==) | `80.00% <80.00%> (ø)` | `0.00 <0.00> (?)` | |
   | [...apache/hudi/utilities/deltastreamer/Compactor.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL2RlbHRhc3RyZWFtZXIvQ29tcGFjdG9yLmphdmE=) | `68.75% <80.00%> (-8.18%)` | `3.00 <0.00> (ø)` | |
   | [...in/scala/org/apache/hudi/IncrementalRelation.scala](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS1zcGFyay9zcmMvbWFpbi9zY2FsYS9vcmcvYXBhY2hlL2h1ZGkvSW5jcmVtZW50YWxSZWxhdGlvbi5zY2FsYQ==) | `72.41% <100.00%> (-0.17%)` | `0.00 <0.00> (ø)` | |
   | [...va/org/apache/hudi/utilities/schema/SchemaSet.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS11dGlsaXRpZXMvc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvdXRpbGl0aWVzL3NjaGVtYS9TY2hlbWFTZXQuamF2YQ==) | `100.00% <100.00%> (ø)` | `3.00 <3.00> (?)` | |
   | [...ache/hudi/common/fs/inline/InMemoryFileSystem.java](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree#diff-aHVkaS1jb21tb24vc3JjL21haW4vamF2YS9vcmcvYXBhY2hlL2h1ZGkvY29tbW9uL2ZzL2lubGluZS9Jbk1lbW9yeUZpbGVTeXN0ZW0uamF2YQ==) | `79.31% <0.00%> (-10.35%)` | `0.00% <0.00%> (ø%)` | |
   | ... and [5 more](https://codecov.io/gh/apache/incubator-hudi/pull/1566/diff?src=pr&el=tree-more) | |
   
   ------
   
   [Continue to review full report at Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=continue).
   > **Legend** - [Click here to learn more](https://docs.codecov.io/docs/codecov-delta)
   > `Δ = absolute <relative> (impact)`, `ø = not affected`, `? = missing data`
   > Powered by [Codecov](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=footer). Last update [25e0b75...741db83](https://codecov.io/gh/apache/incubator-hudi/pull/1566?src=pr&el=lastupdated). Read the [comment docs](https://docs.codecov.io/docs/pull-request-comments).
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r428209317



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaRegistryProvider.java
##########
@@ -81,11 +66,22 @@ private static Schema getSchema(String registryUrl) throws IOException {
 
   @Override
   public Schema getSourceSchema() {
-    return schema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);
+    try {
+      return getSchema(registryUrl);
+    } catch (IOException ioe) {
+      throw new HoodieIOException("Error reading source schema from registry :" + registryUrl, ioe);
+    }
   }
 
   @Override
   public Schema getTargetSchema() {
-    return targetSchema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);
+    String targetRegistryUrl = config.getString(Config.TARGET_SCHEMA_REGISTRY_URL_PROP, registryUrl);
+    try {
+      return getSchema(targetRegistryUrl);

Review comment:
       yes, target schema is allowed to be different than source schema due to transformations and this is fine.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r428206400



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaSet.java
##########
@@ -0,0 +1,44 @@
+/*
+ * 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.hudi.utilities.schema;
+
+import java.io.Serializable;
+import java.util.HashSet;
+import org.apache.avro.Schema;
+import org.apache.avro.SchemaNormalization;
+
+import java.util.Set;
+
+/**
+ * Tracks already processed schemas.
+ */
+public class SchemaSet implements Serializable {
+
+  private final Set<Long> processedSchema = new HashSet<>();

Review comment:
       I think this is similar in scope to how sparkConf maintains avro schemas. In continuous mode, we reuse the same spark session. I think this is ok.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] vinothchandar commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
vinothchandar commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r416319065



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/DeltaSync.java
##########
@@ -162,18 +162,23 @@ public DeltaSync(HoodieDeltaStreamer.Config cfg, SparkSession sparkSession, Sche
     this.fs = fs;
     this.onInitializingHoodieWriteClient = onInitializingHoodieWriteClient;
     this.props = props;
-    this.schemaProvider = schemaProvider;
 
     refreshTimeline();
-
     this.transformer = UtilHelpers.createTransformer(cfg.transformerClassNames);
     this.keyGenerator = DataSourceUtils.createKeyGenerator(props);
-
-    this.formatAdapter = new SourceFormatAdapter(
-        UtilHelpers.createSource(cfg.sourceClassName, props, jssc, sparkSession, schemaProvider));
-
     this.conf = conf;
+    refreshSchemaProvider(schemaProvider);
+  }
 
+  /**
+   * Very useful when DeltaStreamer is running in continuous mode.
+   * @param schemaProvider
+   * @throws IOException
+   */
+  public void refreshSchemaProvider(SchemaProvider schemaProvider) throws IOException {

Review comment:
       right.. embedded server re-use is important. let's scratch my idea. 
   
   




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-629430390


   > @pratyakshsharma : I updated this PR to address comments in the interest of reducing the review cycle time.
   
   I went through the changes. Looks good. I guess we can close it then or is there anything else to be done here? @bvaradar 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-629167984


   @pratyakshsharma : I updated this PR to address comments in the interest of reducing the review cycle time. 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r428210549



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaSet.java
##########
@@ -0,0 +1,44 @@
+/*
+ * 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.hudi.utilities.schema;
+
+import java.io.Serializable;
+import java.util.HashSet;
+import org.apache.avro.Schema;
+import org.apache.avro.SchemaNormalization;
+
+import java.util.Set;
+
+/**
+ * Tracks already processed schemas.
+ */
+public class SchemaSet implements Serializable {
+

Review comment:
       I will add serialVersionUUID. Usually, serialversion mismatch is a clue to an underlying problem which is package version mismatch 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] vinothchandar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
vinothchandar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-620376125


   Zooming out a bit, this seems like an artifact on how the different schema providers are implemented right... Most of them fetch the schema in the constructor (design choice to fail fast and early when there were no continuous mode) and `schemaProvider.getSourceSchema()` just returns this initial value.. if we redid the schema provider implementations such that the schema is read each time from SR (schema registry) (like how the JDBCSchemaProvider does).. This problem will naturally be fixed and is a more elegant design IMO..
   
    


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r416220182



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/DeltaSync.java
##########
@@ -162,18 +162,23 @@ public DeltaSync(HoodieDeltaStreamer.Config cfg, SparkSession sparkSession, Sche
     this.fs = fs;
     this.onInitializingHoodieWriteClient = onInitializingHoodieWriteClient;
     this.props = props;
-    this.schemaProvider = schemaProvider;
 
     refreshTimeline();
-
     this.transformer = UtilHelpers.createTransformer(cfg.transformerClassNames);
     this.keyGenerator = DataSourceUtils.createKeyGenerator(props);
-
-    this.formatAdapter = new SourceFormatAdapter(
-        UtilHelpers.createSource(cfg.sourceClassName, props, jssc, sparkSession, schemaProvider));
-
     this.conf = conf;
+    refreshSchemaProvider(schemaProvider);
+  }
 
+  /**
+   * Very useful when DeltaStreamer is running in continuous mode.
+   * @param schemaProvider
+   * @throws IOException
+   */
+  public void refreshSchemaProvider(SchemaProvider schemaProvider) throws IOException {

Review comment:
       @pratyakshsharma : It looks like refreshSchemaProvider  not only refreshes schema-provider but also recreates Source and setup WriteClient
   
   @vinothchandar : Recreating DeltaSync each run would require to handle embedded timeline server reuse and async compaction triggering differently.  Another option is to have explicit refreshSchema() API in SchemaProvider (with default implementation (for compatibility) and implementing refresh in existing Schema Provider implementation) and have delta-streamer call this ? Let me know your thoughts on this ?
   
   
   




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r416694602



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/DeltaSync.java
##########
@@ -162,18 +162,23 @@ public DeltaSync(HoodieDeltaStreamer.Config cfg, SparkSession sparkSession, Sche
     this.fs = fs;
     this.onInitializingHoodieWriteClient = onInitializingHoodieWriteClient;
     this.props = props;
-    this.schemaProvider = schemaProvider;
 
     refreshTimeline();
-
     this.transformer = UtilHelpers.createTransformer(cfg.transformerClassNames);
     this.keyGenerator = DataSourceUtils.createKeyGenerator(props);
-
-    this.formatAdapter = new SourceFormatAdapter(
-        UtilHelpers.createSource(cfg.sourceClassName, props, jssc, sparkSession, schemaProvider));
-
     this.conf = conf;
+    refreshSchemaProvider(schemaProvider);
+  }
 
+  /**
+   * Very useful when DeltaStreamer is running in continuous mode.
+   * @param schemaProvider
+   * @throws IOException
+   */
+  public void refreshSchemaProvider(SchemaProvider schemaProvider) throws IOException {

Review comment:
       > It looks like refreshSchemaProvider not only refreshes schema-provider but also recreates Source and setup WriteClient
   
   Do you see any side effects of doing this? @bvaradar 
   
   > have delta-streamer call this ? 
   
   This call will happen exactly at the same point where I am calling refreshSchemaProvider in delta-streamer, right? 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] bvaradar commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
bvaradar commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r428208406



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/schema/SchemaRegistryProvider.java
##########
@@ -81,11 +66,22 @@ private static Schema getSchema(String registryUrl) throws IOException {
 
   @Override
   public Schema getSourceSchema() {
-    return schema;
+    String registryUrl = config.getString(Config.SRC_SCHEMA_REGISTRY_URL_PROP);

Review comment:
       this is called in every run and if we detect schema change, we register and recreate write client.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] vinothchandar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
vinothchandar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-624114815


   >Also the above plan only works for the combination of AvroKafkaSource and SchemaRegistryProvider. Thoughts?
   
   All for improving AvroKafkaSource/SR combo, since its heavily used.. but our framework needs to be improved generally.. 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] pratyakshsharma commented on a change in pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
pratyakshsharma commented on a change in pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#discussion_r426184652



##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/Compactor.java
##########
@@ -59,6 +60,10 @@ public void compact(HoodieInstant instant) throws IOException {
           "Compaction for instant (" + instant + ") failed with write errors. Errors :" + numWriteErrors);
     }
     // Commit compaction
-    compactionClient.commitCompaction(instant.getTimestamp(), res, Option.empty());
+    writeClient.commitCompaction(instant.getTimestamp(), res, Option.empty());
+  }
+
+  public void updateWriteClient(HoodieWriteClient writeClient) {

Review comment:
       Yes, in HoodieDeltaStreamer. 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [incubator-hudi] vinothchandar commented on pull request #1566: [HUDI-603]: DeltaStreamer can now fetch schema before every run in continuous mode

Posted by GitBox <gi...@apache.org>.
vinothchandar commented on pull request #1566:
URL: https://github.com/apache/incubator-hudi/pull/1566#issuecomment-628749732


   @bvaradar this and #1518 are again related.. Can you take both of these home ? 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org