[
https://issues.apache.org/jira/browse/TIKA-1544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14309977#comment-14309977
]
Tim Allison edited comment on TIKA-1544 at 2/6/15 9:45 PM:
-----------------------------------------------------------
Y, I'm not sure.
I have hesitation about changing the original logic, because I'm not sure why
if(inParagraph) was added...maybe for properly closing formatting?
I wonder if
{noformat}
} else {
lazyStartParagraph();
out.endElement("p");
}
{noformat}
would be more general and handle the formatting stuff properly?
We could do an A/B test against govdocs1 and see what happens. :)
was (Author: [email protected]):
Y, I'm not sure. If there are consecutive {{\par}}s, we aren't calling
lazyStartParagraph, and I don't think we should.
I have hesitation about changing the original logic, because I'm not sure why
if(inParagraph) was added...maybe for properly closing formatting?
I wonder if
{noformat}
} else {
lazyStartParagraph();
out.endElement("p");
}
{noformat}
would be more general and handle the formatting stuff properly?
> empty lines are not preserved
> -----------------------------
>
> Key: TIKA-1544
> URL: https://issues.apache.org/jira/browse/TIKA-1544
> Project: Tika
> Issue Type: Bug
> Affects Versions: 1.6
> Environment: Windows 8, Java 1.8
> Reporter: mortee
> Priority: Minor
> Attachments: preserve_new_lines_in_rtf.patch, testRTFNewlines.rtf
>
>
> I'm trying to extract the text content from RTF documents. The files contain
> empty lines (two or more consecutive paragraph-end marks), on which the
> further processing relies to tell apart different parts of the text. But
> unfortuantely Tika (with --text switch) eliminates all those empty lines,
> instead of preserving them.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
