[rfc][icedtea-web] Fix CacheReproducerTest

[Andrew Azores]

On 08/30/13 07:06, Jacob Wisor wrote:
> Even at the risk this may be perceived as nagging, but you may want to consider replacing those "/" String literals with File.separator, especially in places where the string concatinations are derived from system properties. It is probably no problem to pass slashes ( / ), backslashes ( \ ), or any other system specific file separator to File's constuctors or objects because File converts any paramters into URLs internally anyway. But when manipulating or comparing raw strings one may get undefined or different results. Or, one should make sure to use URLs only and stick to those when manipulating or comparing strings.
Mm yea, this makes sense I suppose. Hopefully it shouldn't affect
anything to be using the '/' char like you said, but, better safe than
sorry.

> Another thing that perhaps should be considered are systems that do not - for what ever reason - implement the XDG specification. What should be IcedTea-Web's expected behavior in this case? Should it simply force XDG onto that system or should it rather fall back to the previously employed simple scheme (System.getProperty("user.home") + File.separator + ".icedtea")?
>
> Regards,
> Jacob
(changed font formatting of this mail because it came out looking,
frankly, atrocious in Thunderbird on my end. There was also some strange
extra email address added to the Cc when I hit reply-all. Please find a
better mail client, Jacob...)

I don't know honestly, I didn't do the thinking behind the XDG-spec
patch :). In any case if you're worried about that behaviour, this
reproducer test is not really the place to be fixing it. I think this
reproducer should really just be looking for the cache where it's been
specified by the user, or in the default location if the user hasn't
changed that, and not worrying about whether or not that cache location
is XDG-spec compliant or not.

On that note - Jiri, wouldn't using the value from Defaults not take
into account the user changing the cache location using itweb-settings?
Perhaps it makes more sense to use DeploymentConfiguration here to
ensure that the reproducer test is actually testing using the cache dir
that the user expects to use as a cache dir? This way if you've never
changed the cache location in itweb-settings then you get the same path
as taking it from Defaults would give you, but it does still also
respect if you've changed the cache location.

Thanks,

-- 
Andrew A

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20130830/7c653d9a/attachment.html