|Summary:||squirrelmail cannot handle handle multibyte characters in attachment.|
|Product:||Red Hat Enterprise Linux 4||Reporter:||Issue Tracker <tao>|
|Component:||squirrelmail||Assignee:||Warren Togami <wtogami>|
|Status:||CLOSED ERRATA||QA Contact:|
|Fixed In Version:||RHSA-2006-0668||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2006-09-26 12:26:39 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
|Cloudforms Team:||---||Target Upstream Version:|
Comment 1 Issue Tracker 2006-06-08 10:43:35 UTC
Description of problem: After update squirrelmail to squirrelmail-1.4.6-5.el4, it can't handle handle multibyte characters in the bese64 attachment correctly. Version-Release number of selected component (if applicable): squirrelmail-1.4.6-5.el4 How reproducible: Always Steps to Reproduce: 1. accesss http://dhcp-82.brisbane.redhat.com/webmail/src/webmail.php Actual results: Please see Actual_results_01.png or Actual_results_02.png Expected results: Please see Expected_results.png Additional info: squirrelmail-1.4.3a-9.EL4 works fine, but the latest version can not treat multibyte characters in the bese64 attachment. It seems that a bug in formatBody() of functions/mime.php. It passes the wrong args against translateText(). Or $body_message->header->getParameter('charset'), it may not work correctly. Kind regards, Fuchi This event sent from IssueTracker by jnansi [Support Engineering Group] issue 95106
Comment 9 Warren Togami 2006-06-08 16:31:38 UTC
Copied Fuchi's mail to another RHEL4 server and tried to read it in both versions of Squirrelmail, with Japanese language mode set for the interface, and Firefox web browser. The results that I'm seeing indicate that even the previous version wasn't working properly. http://people.redhat.com/wtogami/temp/squirrelmail-1.4.3a-9.EL4.png The garbage text where the attachment text should be viewable is encoded in UTF-8, while the rest of the Japanese text in both the interface and body are EUC-JP. If you view this in either Firefox or Konqueror, you can read the attachment text *ONLY* if you force the encoding to UTF-8. But then you can't read anything else on the page, because everything else is EUC-JP encoded and it becomes garbage. http://people.redhat.com/wtogami/temp/squirrelmail-1.4.6-5.el4.png In the case of the newer squirrelmail, it does fail to display UTF-8 where the attachment text is. Key questions here: 1) *WHY* has the Squirrelmail web interface always been in EUC-JP? This doesn't match the encoding of most incoming mail (ISO-2022-JP) and it isn't UTF-8 or ASCII compatible. 2) It appears that the "Expected result" screenshot above is using Safari. Does Safari behave differently in your tests from Firefox in rendering different parts of the page in different encodings? In my tests the browsers cannot handle this mixed encoding display that the old version of squirrelmail did.
Comment 10 Warren Togami 2006-06-08 16:39:43 UTC
To be clear, it seems to me that between 1.4.3a and 1.4.6, the view attachment behavior went from one broken behavior to another broken behavior. Is this not the case?
Comment 11 Issue Tracker 2006-06-14 04:51:16 UTC
translateText() is added new arg in test5.patch. It is not good, since translateText() is called the other main php, like view_text.php. This event sent from IssueTracker by fuchi issue 95106
Comment 13 Warren Togami 2006-07-10 17:24:57 UTC
http://people.redhat.com/wtogami/temp/squirrelmail-1.4.7-2.el4.noarch.rpm Believed to be fixed in FC and this test package.
Comment 30 Red Hat Bugzilla 2006-09-26 12:26:39 UTC
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2006-0668.html