laskoviymishka commented on code in PR #2482:
URL: https://github.com/apache/iceberg-go/pull/2482#discussion_r4537061687
##########
table/arrow_scanner.go:
##########
@@ -1966,20 +1998,9 @@ func (as *arrowScan) prepareToRead(ctx context.Context,
file iceberg.DataFile, i
entry := as.fileReadPlanCache.entry(file.FilePath())
Review Comment:
Should the delete sets be part of this cache key? The plan from
`readPrunedSchema` now depends on `deleteSets`, but the entry is keyed by
`file.FilePath()` only. With two hand-built `ReadTasks` for the same file, one
with a nested missing-field equality delete and one without, the scan errored
in 5 of 6 runs, depending on which task filled the entry first. `PlanFiles`
never builds that mix, and `main` errors there too.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
