Apache9 commented on PR #7432:
URL: https://github.com/apache/hbase/pull/7432#issuecomment-3491429070

   > Thanks, that is one option. But can we update the documentation of 
`Scan#setLimit` to say:
   > 
   > > ... When used with `TableInputFormat` or `TableSnapshotInputFormat`, 
this limit is applied locally to each split.
   > 
   > From a user’s perspective, even having a per-split limit is still an 
improvement over the current behavior.
   > 
   > As for introducing a new dedicated option, I'd personally prefer fixing 
and reusing the existing interface. As long as we clearly document its 
behavior, I think it's better than introducing another configuration parameter 
for users to discover. HBase already has too many little-known parameters 
buried deep in the source code.
   
   We'd better follow the same pattern with TableSnapshotInputFormat, where we 
introduce a configuration to set limit per split, instead of passing it through 
Scan. This is a more clear solution. Not all users will look deeply into the 
javadoc and we introduce different meanings when using Scan limit through 
normal scan and map reduce job if go with your solution, which may confuse our 
users...


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to