[
https://issues.apache.org/jira/browse/COCOON-2297?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12902345#action_12902345
]
Nico Verwer commented on COCOON-2297:
-------------------------------------
I am afraid the patch is not correct. The line that now reads
tidy.setShowWarnings(getLogger().isWarnEnabled() && ((wChar == 'f') || (wChar
== 'n') || (wChar == '0')));
should have been
tidy.setShowWarnings(getLogger().isWarnEnabled() && ((wChar == 't') || (wChar
== 'y') || (wChar == '1')));
Otherwise, the meaning of show-warnings is inverted. Sorry for the confusion.
Are there still committers for Cocoon 2.1 who are willing to pick this up?
> Character encoding does not follow JTidy properties
> ---------------------------------------------------
>
> Key: COCOON-2297
> URL: https://issues.apache.org/jira/browse/COCOON-2297
> Project: Cocoon
> Issue Type: Bug
> Components: Blocks: HTML
> Affects Versions: 2.1.11
> Reporter: Nico Verwer
> Attachments: HTMLTransformer.patch
>
>
> The text that HTMLTransformer sends to JTidy is always encoded according tot
> the platform default encoding, by calling text.getBytes() without an encoding
> parameter. JTidy does not follow the platform default encoding, but has its
> own default. It is possible to change JTidy's input encoding in the
> properties file.
> The patch uses the encoding specified by JTidy's configuration.
> The result is that HTMLTransformer handles UTF-8 or other encodings
> correctly, so you don't get Chinese characters where you expected a
> diacritical mark.
> While I was changing the code, I also changed the logging settings. They now
> take the settings in the JTidy configuration into account.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
