[jira] Commented: (JCR-2012) BufferedStringValue corrupts non ISO-8859-1 characters on large Strings

["Thomas Mueller (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/JCR-2012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12680490#action_12680490 ] 

Thomas Mueller commented on JCR-2012:
-------------------------------------

I'm working on this problem as well.

There is another bug: the reported length is wrong.

My patch is larger. It also improves performance and simplified the code. I will post a patch later today.

> BufferedStringValue corrupts non ISO-8859-1 characters on large Strings
> -----------------------------------------------------------------------
>
>                 Key: JCR-2012
>                 URL: https://issues.apache.org/jira/browse/JCR-2012
>             Project: Jackrabbit Content Repository
>          Issue Type: Bug
>          Components: jackrabbit-core
>    Affects Versions: core 1.4.5
>         Environment: Sun JDK 1.6, Win 2000
>            Reporter: Henryk Paluch
>            Priority: Critical
>         Attachments: BufferedStringValue.java.encFix.diff, BufferedStringValueTest.java
>
>
> When storing,retrieving large String values (for example large "sv:property" named "content" - which contains text of paragraphs) then non-ISO-8859-1 characters are lost.
> This is caused becaus of improper handling of Temporary files in BufferedStringValue - they use Readers/Writers without specifying encoding - so national characters could be lost if system wide encoding does not support them.
> Pending attachments:
> - JUnit test Case
> - Proposed fix

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