[
https://issues.apache.org/jira/browse/TIKA-1948?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15235825#comment-15235825
]
Tim Allison commented on TIKA-1948:
-----------------------------------
I ran this proposed change against ~325k pdfs. There were very modest gains,
with no downsides. We extracted more content (and it looked like good content,
not mojibake) for 40 files. Any objections to adding this?
> Catch exceptions per page in PDFParser
> --------------------------------------
>
> Key: TIKA-1948
> URL: https://issues.apache.org/jira/browse/TIKA-1948
> Project: Tika
> Issue Type: Improvement
> Reporter: Tim Allison
> Assignee: Tim Allison
> Priority: Minor
>
> In a discussion with [~tilman] somewhere(???), I think he observed that we
> weren't doing a try/catch for each page. If there's an exception in an early
> page, it might still be possible to extract text from later pages in a
> problematic PDF.
> With very minimal modifications we could add a try/catch per page, store the
> caught exceptions, and then throw the first caught exception after the parse
> finishes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
