[
https://issues.apache.org/jira/browse/HTTPCORE-229?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12887777#action_12887777
]
Oleg Kalnichevski commented on HTTPCORE-229:
--------------------------------------------
Igor,
My bad. I responded without having analysed the issue properly. I assumed you
were using HttpCore NIO, which is not the case. The LoggingSessionInputBuffer
class is apparently from HttpClient package. This goes to show that providing
relevant details such an exception stack tract does help.
By the way, you are most likely getting this number of characters because
HttpCore has not been configured to use UTF-8 for message elements. Per default
HttpCore assumes it is interacting with sane, standards compliant servers and
therefore it uses ASCII charset as required by the HTTP protocol.
Having said that, I'll make sure LoggingSessionInputBuffer does not choke on
such malformed headers.
Oleg
> IOOBE in case of UTF-8 encoded http element(headers)
> ----------------------------------------------------
>
> Key: HTTPCORE-229
> URL: https://issues.apache.org/jira/browse/HTTPCORE-229
> Project: HttpComponents HttpCore
> Issue Type: Bug
> Components: HttpCore
> Environment: Ubuntu 10.4. java 6
> Reporter: Igor Mikhailuk
> Priority: Minor
> Fix For: 4.1
>
>
> When there is the line in the http response header like this:
> Location: /слово/
> encoded in UTF-8 AbstractSessionInputBuffer#readLine(final CharArrayBuffer
> charbuffer) return 22(bytes count in this line).
> But I beleve it should return 17 (actual chars count).
> So. When we are trying to handle such server responce we got IOOBE in
> LoggingSessionInputBuffer#readLine(final CharArrayBuffer buffer).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@hc.apache.org
For additional commands, e-mail: dev-h...@hc.apache.org
