[ 
https://issues.apache.org/jira/browse/JCR-2007?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Thomas Mueller updated JCR-2007:
--------------------------------

    Description: 
Both Session.importXML and Workspace.importXML don't work correctly in some 
cases.

Importing very large foreign language (for example, Chinese) text properties 
could fail on some platforms. BufferedStringValue buffers very large string to 
a temporary file. It uses the platform default encoding to read and write the 
text.

BufferedStringValue is relatively slow on some systems when importing large 
texts or binary data because of using FD().sync().

If a string contains a carriage return (\r), this character was truncated on 
some platforms.

If a string contains a characters < 32 excluding newline (\n) and tab (\t) - 
for example form feed (\f) - the imported string was base64 encoded.


  was:BufferedStringValue buffers very large string to a temporary file. It 
uses the platform default encoding to read and write the text. Usually this is 
not a problem, however importing very large foreign language (for example, 
Chinese) text properties could fail on some platforms.

        Summary: Importing strings with special characters fails  (was: 
Importing very large strings could fail in some situations)

> Importing strings with special characters fails
> -----------------------------------------------
>
>                 Key: JCR-2007
>                 URL: https://issues.apache.org/jira/browse/JCR-2007
>             Project: Jackrabbit Content Repository
>          Issue Type: Bug
>          Components: jackrabbit-core
>            Reporter: Thomas Mueller
>            Assignee: Thomas Mueller
>
> Both Session.importXML and Workspace.importXML don't work correctly in some 
> cases.
> Importing very large foreign language (for example, Chinese) text properties 
> could fail on some platforms. BufferedStringValue buffers very large string 
> to a temporary file. It uses the platform default encoding to read and write 
> the text.
> BufferedStringValue is relatively slow on some systems when importing large 
> texts or binary data because of using FD().sync().
> If a string contains a carriage return (\r), this character was truncated on 
> some platforms.
> If a string contains a characters < 32 excluding newline (\n) and tab (\t) - 
> for example form feed (\f) - the imported string was base64 encoded.

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

Reply via email to