hi igor,
so i am a little confused as to why we are talking about transparent resolvers, and why gerolf did not apply that code himself...
i tried to verify whether or not is related to the transparent resolver - it seems at last it's connected to them cause i couldn't produce a case where it happens without them. (though i can't claim it can't happen)
maybe gerolf didn't want to step into ground where others may have more insight - that's the reason i filed this to jira.
personally i do not like to change code if it "maybe" helps,
it helps, so it's not 'maybe' in this case.
especially if it is markup parsing code which is a pretty fragile area in wicket.
right, i very well understand - this has to be carefully reasoned. i didn't find any problems with this workaround with my code yet - a test on a second codebase would be helpful though.
if there are no objections i will ask gerolf to apply this patch (and remove it again should there be evidence that it breaks anything on another project).
hopefully there will be another way with 1.5 to cope with transparent resolvers so this only is a matter of time 'til wicket is more robust on those use cases.
best regards, --- jan. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
