[Product-Developers] collective.flowplayer (3.0b4/5): weird behavior with safari and collective.flowplayer.js

[David Hostetler]

collective.flowplayer (3.0b4) was failing to embed itself in safari (4.04).
It just wouldn't show up, and viewing the page source would reveal only the
original markup as seen in the plone editor.

Naro was on #plone and did a quick investigation and suggested removing the
'@@' from the '@@collective.flowplayer.js' entry in portal_javascripts.

Oddly enough -- that worked.   And naro scurried off to package up v3.0b5
which included this 'fix'.

But here's where it gets weird:  adding '@@' back to the portal_javascript
entry didn't rebreak the player.  It still showed up, even though the entry
had been reverted back to the original value.

And it still worked after a zeoclient restart.

I discovered this part after naro had left #plone.   Thinking that the fix
was merely the removal of '@@', he uploaded v3.0b5 which changed the
portal_javascript entry to just be 'collective.flowplayer.js'.

And things get a little weirder:  I installed 3.0b5  (on a test instance -
not the same one on which I'd done the '@@' change manually).  And indeed
the portal_javascript entry reflected the new 3.0b5 setting, as just
'collective.flowplayer.js'.    Only now flowplayer is broken again!  It
doesn't show up.

So I manually ADD '@@' to the portal_javascript entry.   And voila -
flowplayer shows up again!!   So I manually remove '@@'....  and flowplayer
STILL shows up.


What the heck is going on?   It's like something doesn't quite hook up right
initially, but touching the portal_javascript entry 'unsticks' whatever is
wrong, after which it doesn't seem to matter whether the '@@' is there or
not.

I don't know enough about browserviews or javascript to imagine what might
explain this.  I merely had it explained to me by naro that
collective.flowplayer.js itself comes from a browserview, it isn't a plain
js file (like, say, flowplayer.min.js).


Anyway -- I wanted to get this logged so that naro (and maybe Martin) could
see it and offer up an explanation or perhaps a more legitimate fix, since I
don't think 3.0b5 does what it was expected to do.


regards,

-David

David Hostetler
negativesum at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.plone.org/pipermail/plone-product-developers/attachments/20091211/f2f4cbd1/attachment.html>