[PATCH try2 1/4] urlmon/tests: Declare the bind flags as a global variable in url tests.

Misha Koshelev mk144210 at bcm.edu
Mon Aug 6 23:08:21 CDT 2007


Sorry for try2, only change is in patch 3. Just realized I could make it slightly simpler, there
was an unnecessary check for HTTP_TEST when checking the return codes, and also the unk == NULL when
hres == MK_S_ASYNCHRONOUS is no longer a todo_wine, not because it's actually fixed (this will be
the case when we unroll the http hack) but rather b/c the test is fixed to properly only catch unk == NULL 
when we are binding asynchronously and have not yet began receiving data (the case for, say, http protocol 
with the wine about url, but not for http protocol with the post test url or for the other asynchronous 
protocols tested). This does have the unfortunate effect of not really being able to discriminate a
properly implemented asynchronous protocol from one (like the current hack in git) that is completely
synchronous, but as even the "asynchronous" protocol on native can completely finish binding before
BindToStorage returns (it is a timing issue, just try replacing WINE_ABOUT_URL with the posttest url from
protocol.c), there is actually no way to test this in the general case.
---
 dlls/urlmon/tests/url.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 2b80e99ad2657a6f8dff2875df874fce95ad02fa.diff
Type: text/x-patch
Size: 1137 bytes
Desc: not available
Url : http://www.winehq.org/pipermail/wine-patches/attachments/20070806/1ed7faf5/2b80e99ad2657a6f8dff2875df874fce95ad02fa.bin


More information about the wine-patches mailing list