[jira] Updated: (TIKA-334) HtmlParser should use CharsetDetector whenever no charset is specified via meta http-equiv tag

["Ken Krugler (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/TIKA-334?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ken Krugler updated TIKA-334:
-----------------------------

    Attachment: TIKA-334.patch

> HtmlParser should use CharsetDetector whenever no charset is specified via meta http-equiv tag
> ----------------------------------------------------------------------------------------------
>
>                 Key: TIKA-334
>                 URL: https://issues.apache.org/jira/browse/TIKA-334
>             Project: Tika
>          Issue Type: Improvement
>    Affects Versions: 0.5
>            Reporter: Ken Krugler
>         Attachments: TIKA-334.patch
>
>
> Currently the HtmlParser will just call TagSoup to parse, without specifying a charset, if no charset is passed in via metadata.
> TagSoup uses the platform encoding in this case, which is often going to be wrong.
> The right thing to do is to first check for a charset specified by a meta tag. If that doesn't exist, then create a CharsetDetector. If there's a charset in the incoming meta-data, use that to call setDeclaredEncoding().

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.