abhinav-phi opened a new pull request, #19986:
URL: https://github.com/apache/hudi/pull/19986

   describe the issue this pull request addresses
   
   issue 19883 tracks repeated spark archive downloads in integration tests
   this follows the merged dlcdn fast path and preserves its archive fallback
   
   summary and changelog
   
   restore the spark tarball using an exact cache key containing the runner 
operating system and complete archive path
   skip the download only on an exact cache hit
   retain the tarball after extraction and save it only after successful 
integration tests on a master push
   exclude spark archive cache keys from the scheduled cache purge so 
subsequent runs can reuse them
   no prefix restore keys and no cache writes from pull request runs
   
   impact
   
   warm runs can avoid downloading the archive from apache
   cold runs still use the existing dlcdn and archive fallback
   no changes to spark versions or integration test commands
   no hosted cache speedup measurement is claimed
   
   risk level
   
   low
   this changes ci download and cache cleanup behavior
   spark cache generations remain subject to github eviction
   maintainer agreement is needed on the narrow cache cleanup exception
   
   documentation update
   
   none
   
   validation
   
   eight local behavior tests passed using the shell extracted from the 
modified workflows with a small tar fixture and mocked curl and maven commands
   cases cover an exact hit a miss cdn failure with archive success both 
sources failing a corrupt cached archive mixed cache keys only spark keys and 
an empty cache list
   both workflow files parse as yaml
   bot workflow passes actionlint with external shellcheck and pyflakes checks 
disabled
   scheduled workflow reports only the same preexisting checkout version error 
reproduced on the base commit
   whitespace validation passed
   independent static review found no blocking defects
   full hudi integration tests and hosted github cache restore and save have 
not been run locally
   
   contributor checklist
   
   contributor guide read
   context provided above
   focused local behavior checks performed
   
   ai disclosure
   
   ai assistance was used for implementation local validation and this 
description
   


-- 
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