Hi,

On Tue, Aug 11, 2015 at 4:06 PM, Martijn Dashorst <
[email protected]> wrote:

> I've recently poured some time (together with the help of Emond) in
> figuring out why one particular javascript file could not be loaded.
> It appeared that at least 3 bugs were present in Wicket that surfaced
> in our apps that were ported to 7.0.0
>
> From the existing code we have concluded that these bugs also exist in
> Wicket 6, but that they don't typically surface due to some unknown
> code path. We were only able to trigger the bug in one particular code
> path where a strict resource match prevented loading of a js file,
> while directly loading the js file after startup used the non-strict
> matching (WICKET-5968)
>
> Other bugs we encountered were:
>
>  - WICKET-5966 ResourceUtils.getLocaleFromFilename can't handle
> minimized resources well
>
> and
>
> - WICKET-5967 Unable to load i18n minified js
>
> Now the latter two have some behavior changes. Should we backport those
> two?
>

Since they are bugs I think they should be fixed in 6.x too.

My only concern is about the hardcoded usage of '.min.js' as the only
possible extension for minified resources. [1]
But I don't see how this could be improved so that each resource could use
its own naming scheme.

I am +1 to downport the commits and:
- add entry to the Wicket 7.x migration page, section "Behavior changes".
Many people are still on 6.x so this new entry may help them later
- add a note about this to the release announcement of 6.21.0/7.1.0


>
> WICKET-5968 can be backported without problem AFAIK.
>
> WDYT?
>
> Martijn
>


1.
https://issues.apache.org/jira/browse/WICKET-5967?focusedCommentId=14681425&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14681425

Reply via email to