Skip to content

Conversation

@sahilTakiar
Copy link
Contributor

@sahilTakiar sahilTakiar commented Jun 12, 2019

HDFS-14478: Add libhdfs APIs for openFile

Exposes the openFile API via libhdfs, adds new tests to test_libhdfs_ops.c. For the most part, the libhdfs API mirrors the Java API (e.g. the async open file is modeled in a similar fashion to java.util.concurrent.Future).

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@steveloughran
Copy link
Contributor

I dont see any tests for trying to open a file which doesnt exist. When you do that. the failure is postponed until the completion is evaluated, so you need to add a test there to make sure the failure is escalated properly

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

shanthoosh pushed a commit to shanthoosh/hadoop that referenced this pull request Oct 15, 2019
….scala and SystemConfig.scala to Java (apache#955)

* moving FileSystemCheckpointManagerConfig.scala and SystemConfig.scala to java

* renaming JavaSystemConfig to SystemConfig now that the scala version is gone
@hadoop-yetus

This comment has been minimized.

@smengcl
Copy link
Contributor

smengcl commented Jul 31, 2020

Thanks @Sahil for the patch.

Would you rebase the commit on the latest trunk? so we can get a test run against trunk.

@steveloughran
Copy link
Contributor

I should warn that in #2046 I'm adding opt/must to take a long as well; we just missed that out in the past. Needed so that applications like yours can pass in the known length of the file (split limit is fine) and we can skip the HEAD from that alone. hive needs that

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

@hadoop-yetus

This comment has been minimized.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants