[GitHub] orc pull request #283: ORC-378
GitHub user sershe opened a pull request: https://github.com/apache/orc/pull/283 ORC-378 You can merge this pull request into a Git repository by running: $ git pull https://github.com/sershe/orc master-378 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/orc/pull/283.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 #283 commit 47f1aed7608ba276ff228f8094efde0c2c6514ee Author: sergey Date: 2018-06-15T02:31:19Z commit 1 ---
[GitHub] orc pull request #222: ORC-310 better error handling for codec
Github user sershe commented on a diff in the pull request:
https://github.com/apache/orc/pull/222#discussion_r173943752
--- Diff: java/core/src/java/org/apache/orc/impl/RecordReaderUtils.java ---
@@ -306,15 +330,24 @@ public void releaseBuffer(ByteBuffer buffer) {
@Override
public DataReader clone() {
+ if (this.file != null) {
+throw new UnsupportedOperationException(
+"Cannot clone a DataReader that is already opened");
+ }
try {
-return (DataReader) super.clone();
+DefaultDataReader clone = (DefaultDataReader) super.clone();
+// Make sure we don't share the same codec between two readers.
+clone.codec = OrcCodecPool.getCodec(clone.compressionKind);
--- End diff --
super.clone() is Object.clone(), so no
---
[GitHub] orc pull request #222: ORC-310 better error handling for codec
Github user sershe commented on a diff in the pull request:
https://github.com/apache/orc/pull/222#discussion_r171728763
--- Diff: java/core/src/java/org/apache/orc/impl/ReaderImpl.java ---
@@ -472,8 +473,9 @@ public static OrcTail extractFileTail(ByteBuffer
buffer, long fileLength, long m
.setPostscript(ps)
.setFooter(footer)
.setFileLength(fileLength);
+ isCodecError = false;
} finally {
--- End diff --
existing code
---
[GitHub] orc pull request #222: ORC-310 better error handling for codec
Github user sershe commented on a diff in the pull request:
https://github.com/apache/orc/pull/222#discussion_r171728816
--- Diff: java/core/src/java/org/apache/orc/impl/WriterImpl.java ---
@@ -645,6 +645,8 @@ public void
appendUserMetadata(List userMetadata) {
return ReaderImpl.deserializeStats(builder.getStatisticsList());
}
+ // TODO: remove this
--- End diff --
Removed the attribute, clarified the TODO
---
[GitHub] orc issue #222: ORC-310 better error handling for codec
Github user sershe commented on the issue: https://github.com/apache/orc/pull/222 Updated the JIRA with motivation. I will rename the flag; there isn't a better way primarily because of the TODOs you are asking me to remove - we don't have access to information about errors if the user and the return-er of the codec are two different pieces of code at different times. There's also no good way to check the state of the codec to see if it's valid or not. Hadoop ZLib codec may just throw NPE in reset when it's invalid, so the "Safely" catch block will handle that case. However it's nice to give user an opportunity to be proactive about this, esp. in the case of the get-use-return usage pattern ---
[GitHub] orc pull request #222: ORC-310 better error handling for codec
GitHub user sershe opened a pull request: https://github.com/apache/orc/pull/222 ORC-310 better error handling for codec You can merge this pull request into a Git repository by running: $ git pull https://github.com/sershe/orc orc-310 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/orc/pull/222.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 #222 commit d9727cb36c9b1e9f1c045e29d928401a680062cb Author: sergey Date: 2018-03-01T00:00:51Z ORC-310 better error handling for codec ---
[GitHub] orc pull request #102: ORC-166 : add codec pool to ORC; make sure end is cal...
GitHub user sershe opened a pull request: https://github.com/apache/orc/pull/102 ORC-166 : add codec pool to ORC; make sure end is called on underlyin⦠â¦g codecs (Sergey Shelukhin) You can merge this pull request into a Git repository by running: $ git pull https://github.com/sershe/orc master-166 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/orc/pull/102.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 #102 commit dd6cab70e37c7eb0ba2c7dbac78d463b1f33c873 Author: Sergey Shelukhin Date: 2017-03-21T02:13:06Z ORC-166 : add codec pool to ORC; make sure end is called on underlying codecs (Sergey Shelukhin) --- 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] orc pull request: ORC-55. Resynchronize storage-api and orc with t...
Github user sershe commented on the pull request: https://github.com/apache/orc/pull/25#issuecomment-221338899 +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. ---
