You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tika.apache.org by "Gerard Bouchar (JIRA)" <ji...@apache.org> on 2018/07/31 15:41:00 UTC
[jira] [Commented] (TIKA-2700) The HTML parser should parse the
contents of the title tag as raw text, not HTML
[ https://issues.apache.org/jira/browse/TIKA-2700?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16563851#comment-16563851 ]
Gerard Bouchar commented on TIKA-2700:
--------------------------------------
The bug is in tagsoup, that tika uses. Unfortunately, the library seems to be completely unmaintained (and it doesn't have any tests).
I made [a pull request|https://github.com/jukka/tagsoup/pull/4] to them, but I doubt they react anytime soon.
Are there plans to move to a better HTML parsing library by default ?
> The HTML parser should parse the contents of the title tag as raw text, not HTML
> --------------------------------------------------------------------------------
>
> Key: TIKA-2700
> URL: https://issues.apache.org/jira/browse/TIKA-2700
> Project: Tika
> Issue Type: Bug
> Reporter: Gerard Bouchar
> Priority: Major
> Attachments: title.html
>
>
> The current HTML parser in tika fails to extract the correct document title when it contains at least one unescaped '<' character.
>
> For instance, in the following HTML document:
> {code:html}
> <html><title>title with a <b>tag</b> in it</title><body></body></html>
> {code}
> the extracted title is
> {code}
> title with a
> {code}
> Browsers however respect the [html parsing specification|https://www.w3.org/TR/2011/WD-html5-20110113/tokenization.html#parsing-main-inhead], and display this title as
> {code}
> title with a <b>tag</b> in it
> {code}
> (with a literal _<b>_)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)