PDGGK commented on PR #10532:
URL: https://github.com/apache/seatunnel/pull/10532#issuecomment-5079292953

   Thanks @DanielLeens — the conflict is real, and it turned out to be worth 
looking at before resolving it.
   
   The conflict is entirely in `AbstractJdbcCatalog.java`, and it exists 
because that half of this PR has since
   landed on `dev` independently: #11133 (`80c003c45`) implements the same 
close-all-connections behaviour —
   aggregate the failures with `addSuppressed`, clear the map, rethrow — and 
ships its own
   `AbstractJdbcCatalogTest`. Issue #10530 is closed as completed accordingly. 
So resolving the conflict by
   keeping my version would just recreate code that is already on `dev`. I'm 
dropping `AbstractJdbcCatalog.java`
   and `AbstractJdbcCatalogCloseTest.java` from this PR instead.
   
   The other half is still outstanding. On current `dev`, 
`AbstractReadStrategy#resolveArchiveCompressedInputStream`
   still builds `new 
GzipCompressorInputStream(hadoopFileSystemProxy.getInputStream(path))` in the 
`case GZ:` branch
   without try-with-resources, while `ZIP`, `TAR` and `TAR_GZ` right next to it 
all use it — so the stream is leaked
   once per split. That's #10531, still open.
   
   @zhangshenghang — before I rebase and rescope this one, could you say which 
PR you'd like to carry the GZ fix?
   You have #11209 open for the same thing, so there's no point in both of us 
pushing near-identical changes:
   
   - If you'd rather land #11209, say so and I'll close this PR right away — no 
objection at all, and I'll move the
     `AbstractReadStrategyGzCloseTest` case over there as a suggestion if it's 
useful.
   - If you'd rather this one carried it, I'll rebase onto current `dev` and 
cut it down to the two connector-file
     files (the fix plus the regression test), and retitle accordingly.
   
   For what it's worth both branches are equally out of date against `dev` 
right now — #11209 is also showing
   conflicts — so whichever one we keep needs the same rebase. Happy either 
way; just want to avoid duplicating your
   work a second 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.

To unsubscribe, e-mail: [email protected]

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

Reply via email to