[
https://issues.apache.org/jira/browse/TIKA-1930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15226967#comment-15226967
]
Tim Allison commented on TIKA-1930:
-----------------------------------
Is there a way to tell the underlying parser to avoid caching the files or to
clean them up? Or do we need to create our own temp directory into which we
place the temp file and then wipe the full directory including the gbx/ncx
files after a parse?
> Clean up resources from grib parser
> -----------------------------------
>
> Key: TIKA-1930
> URL: https://issues.apache.org/jira/browse/TIKA-1930
> Project: Tika
> Issue Type: Bug
> Reporter: Tim Allison
> Priority: Minor
>
> The grib parser isn't currently set up to delete the temp file correctly.
> That part is a trivial fix.
> The underlying library also generates temp files ending in "gbx9" and "ncx2"
> that we aren't currently cleaning up. I've tried variations on
> {{DiskCache.cleanCache(0, new StringBuilder());}} with no luck.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
