[rfc][icedtea-web] fix some nastyness in reproducers

[Jiri Vanek]

On 03/09/2015 02:42 PM, Jie Kang wrote:
>
>
> ----- Original Message -----
>> Hello!
>>
>
> Hello,
>
> Patch seems fine.
>
>> During todays debugging I found few nasty things in reproducers:
>> - invlalid (useless) urls
>>     - please do not wont me to fix the new line \> - I was just sedding those
>
> Hahah... I was going to say, fix the new lines, till I read this. Still, is it that hard? I don't even see that many... Can I push a fix for the new lines if you don't want to do it?

Ok. Feel free to push my chnageset. with this change .  TY!
>
>>     jnlps....
>> - "" codebases (sigh!)
>
>  From [1] 'Best Practices' section, I think it is better for us to use empty/no codebase instead of "." codebase. What are the benefits of changing them all to use "."? Afaict, the tests all work.

Well....  in case of itw... not so obviously:
http://icedtea.classpath.org/hg/icedtea-web/rev/c6af2f50a95e?revcount=120
>
> [1] http://docs.oracle.com/javase/7/docs/technotes/guides/jweb/applet/codebase_determination.html
>
>> - and JS tests with disabled opera. It seems to me like opera have improved,
>> and most of the tests is now not failing.
>
> Sure. When you say 'most' can you be more specific? Do you know which reproducers still fail on opera and which don't?

All I tested passed. (Bout 10). I had not tried the rest.

Still I think this exclusion should be gone...  Maybe tune the knowntofail annotation to determine browser rather then this hack.


ty!

J.