This class is not for direct initialization, and
new MWHttpRequest(...)->execute() doesn't work at all.
Change-Id: I557723c218cd8e8f27283d01141c890263da095d
@fixme is simply not recognized by doxygen whereas @todo is used to
generate a nice ... todo list!!
Change-Id: If956c0a164373126ce48b791d45c56962034eecd
* respHeaders contains arrays for each header (should've been
respHeaders['content-length'][0] instead of respHeaders['content-length'])
using getResponseHeader() instead.
* This also fixes InstantCommons, which was broken by this (bug 36653).
* Also cleaned up a bit of w/s in foreign repo code while in the area.
Change-Id: I429b0a36618cc2b873077516e9c1023dc5b7d693
We detect such prematurely ended request when there's a Content-Length
greater than the body we got. A strict comparison would easily fail
when there was a content encoding layer, but hopefully we won't hit
a compressed request _larger_ than the original content.
See http://thread.gmane.org/gmane.org.wikimedia.mediawiki/39622
for breakage caused by truncated HTTP replies.
Change-Id: I71418424730d46a781bde5cbfda8038457ec79c5
examining MWHttpRequest::method always expects its value to be
fully uppercased.
Yes, this should not be a problem if MWHttpRequest is accessed
via HTTP class, but that will not always be guaranteed.
Resolves bug 36137.
Change-Id: Ic20e1b99dcb56b8a11cea50293ba44022e564de9
Really, I'm not sure about the usefulness of exploding with a MWException anytime we can't do an external request, but at least we can stop the installer from exploding :)
Tested fix with PHPUnit Tests that aren't currently executed by
CruiseControl b/c they're marked “broken” till someone comes up with
some reasonable unit tests that will work for everyone.
r65152 switched upload-by-URL ($wgAllowCopyUploads) to use Http / MwHttpRequest class instead of CURL directly.
While this is mostly nice, it switched from saving large files directly to a temp file to buffering them in memory, causing large files to fail when they hit the PHP memory limit.
Fix uses MwHttpRequest's callback capability to override the read handler; now appending it to the temporary file as we go, and can confirm that largish files work again; was able to upload a 64mb .ogv that previously didn't work for me: http://prototype.wikimedia.org/tmh/images/b/b2/File-Arborophila_brunneopectus_pair_feeding_-_Kaeng_Krachan.ogv
Also expanded the documentation on MwHttpRequest::setCallback to clarify the function parameters and return value for the callback (which currently matches the low-level CURL handler's callback directly).
Note that the non-CURL implementation doesn't abort the read if the callback doesn't return the expected number of bytes, but this is an immediate fatal end of request on the Curl backend. May want further cleanup.
regex, but also accepts ftps because both cURL and php support it. It no longer accepts thing like 'foo http://bar bax'
which was my main concern
Note the previous regex kind of looks more restrictive, but is not since saying "anything not containing a space
optionally followed by anything not containing a bunch of characters including a space" is the same as saying anything
with no spaces. See also r83296. This obviously doesn't catch all cases, but I personally think its sufficient.
At the very least it is a very significant improvement over the previous version that caught almost nothing.
