[Mediawiki-l] Http::get() fails on some URLs

[Platonides]

Yaron Koren wrote:
> Hi,
> 
> My extension External Data uses Http::get() to retrieve the contents of a
> URL. Unfortunately, it turns out that this function fails on some URLs,
> returning an HTTP return code of "0". Here's an example of a URL for which
> it doesn't work:
> 
> http://mouse.brain-map.org/GeneExpression/Hippocampal+region/1.xml
> 
> Http::get() uses PHP's cURL library when possible (and on my server, it's
> possible), and I assume that's what's causing the problem: when I replace
> the call to Http::get() with a call to file_get_contents(), it works fine.
> Does anyone know what the exact problem is? (It might be related to the fact
> that the page is in XML format, although other XML pages work.) Should I
> just use file_get_contents() instead?
> 
> Thanks,
> Yaron

Try increasing $wgHTTPTimeout
It's a big file, so it's probably timeouting before fully transferring
it, as it happened here. file_get_contents doesn't have a timeout,
unless specifically set with a context.