[
https://issues.apache.org/jira/browse/SHINDIG-445?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
John Hjelmstad updated SHINDIG-445:
-----------------------------------
Attachment: httpresp.patch
Potential solution attached. I haven't thought through all the implications, so
it'd be great to get input from anyone else.
> Invalid Date: header from remote server shouldn't cause null pointer exception
> ------------------------------------------------------------------------------
>
> Key: SHINDIG-445
> URL: https://issues.apache.org/jira/browse/SHINDIG-445
> Project: Shindig
> Issue Type: Improvement
> Components: Gadget Rendering Server (Java)
> Reporter: John Hjelmstad
> Priority: Minor
> Attachments: httpresp.patch
>
>
> I've recently encountered a server which returns an invalid HTTP/1.1 Date
> header:
> Date: Thu, 10 Jul 2008 22:54:20 EDT
> Per (http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.3.1), Date
> must be specified in GMT. Shindig's implementation is in line with this
> requirement.
> Shindig Java's implementation causes an NPE with this input due to the
> following code:
> DateUtil.java:
> public static Date parseDate(String dateStr) {
> try {
> return rfc1123DateFormat.parseDateTime(dateStr).toDate();
> } catch (Exception e) {
> // Don't care.
> return null;
> }
> }
> Used here:
> return DateUtil.parseDate(date).getTime();
> I'm inclined to give poor caching characteristics to such inputs rather than
> NPE altogether, which offers a poor/confusing developer experience, in
> particular returning 0 when the Date returned from parseDate is invalid.
> Thoughts?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
