[
https://issues.apache.org/jira/browse/TIKA-4891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18114644#comment-18114644
]
Tim Allison commented on TIKA-4891:
-----------------------------------
Bot reports:
{noformat}
(217 KB, producer Skia/PDF m86, i.e. Chrome's print-to-PDF). It came from the
SAFEDOCS zip 1008.zip, so it is Common
Crawl material and fine to attach. What Tilman will see in it: obj 73 0 R
/S /TD /A << /O /Table /Headers [ (node00000605) ] >>
obj 75 0 R /S /TD /A << /O /Table /Headers [ (node00000606) ] >> The
/Headers array holds COSStrings, and PDStandardAttributeObject.getArrayOfString
casts each element to COSName, so PDTableAttributeObject.getHeaders()
throws ClassCastException. The spec says the entries are byte strings
(element IDs). Skia writes it this way
for every table cell, so any Chrome-printed tagged table trips it.
{noformat}
> Improve handling of PDF/UA structural tags/marked content
> ---------------------------------------------------------
>
> Key: TIKA-4891
> URL: https://issues.apache.org/jira/browse/TIKA-4891
> Project: Tika
> Issue Type: Task
> Reporter: Tim Allison
> Priority: Major
> Attachments: 0001869.pdf
>
>
> PDF/UA includes structural markup. We hacked out a standalone handler for
> this back in 1.x but haven't touched it in years.
>
> We should modernize our handling of structural tags and eventually consider
> turning that on by default. That decision will be based on evaluation on
> 1000s of PDFs. This is not a default switch to be taken lightly.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
