| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
| |
If there is a issue loading the usual .pr* files do not crash qmake.
Print out a useful warning if MAKEFILE_GENERATOR is not defined by a
loaded .pr* file.
Reviewed-by: mariusSO
|
| |
|
|
| |
Reviewed-by: Trust Me
|
| |
|
|
|
|
|
|
| |
having system-dependent bahavior is Not Nice (TM), and given that we do
only printf()s of whole lines anyway, this should have no performance
impact whatsoever.
Reviewed-by: mariusSO
|
| |
|
|
|
|
|
|
|
|
|
|
| |
behavior on all versions of Windows).
setvbuf handles its arguments differently on Unix and Windows.
Windows uses the size parameter when the given buffer is NULL, which
appears to violate C89/C99. Giving a size parameter of 0 with a NULL
buffer caused qmake to crash on Windows Vista Business x64.
Windows also can't set line buffering with setvbuf _at all_ according to
the MSDN documentation, so don't bother calling it on Windows.
|
| |
|
|
|
|
|
|
| |
(which is unbuffered)
Task-number: 254786
Merge-request: 610
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
| |
|
|
|
|
|
|
| |
SUBDIRS target
they don't have initially, either.
Reviewed-by: mariusSO
|
| |
|
|
|
|
|
|
| |
the top-level one doesn't have one, either, so this lead to inconsistent
behavior depending on whether the pro was processed directly or from a
SUBDIRS target.
Reviewed-by: mariusSO
|
| |
|
|
| |
Reviewed-by: mariusSO
|
| |
|
|
|
|
| |
seems it was never used since the initial checkin
Reviewed-by: mariusSO
|
| |
|
|
| |
Reviewed-by: mariusSO
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
This fixes a problem where qmake would not compile when it was built
from it's .pro file because this method uses the pre-compiled header.
This header was causing a compile error in qlocale.cpp because
qtextstream.h was included and this includes qlocale.h. The problem in
qlocale.cpp was that it uses a define called QLOCALE_CPP to enable
extra functions in the class declaration, but the pre-compiled header
was preventing the qlocale.h from being re-processed and therefore the
function was never compiled in.
Reviewed-by: Marius Storm-Olsen
|
| |
|
|
|
|
|
|
|
|
| |
Only the Solution Generator was using the fixFilename() function, so
under some circumstances the solution filename wouldn't find the
correct vcproj file to include. This created a problem with
network-chat.vcproj.
Task-number: 254772
Reviewed-by: joao
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
DESTDIR=./ and using Windows, nmake and shadow builds.
qmake would canonicalize the DESTDIR of "./" to "". Then it would check
if the original DESTDIR ended with "/", and if so, append it to the new
DESTDIR, resulting in a DESTDIR of "/" - the root of the current drive.
Don't do that.
This bug doesn't occur with in-source builds because qmake detects that
the source and build directories are the same directory and replaces
the DESTDIR of "./" with "" before it reaches the buggy code.
Autotest: included
Reviewed-by: Lincoln Ramsay
|
| |
|
|
|
|
| |
foo == QString() should be foo.isNull(). Fixes 7 warnings in the Norwegian Breakfast Network
Reviewed-by: Samuel
|
| |
|
|
|
|
| |
version from a Linux machine.
Reviewed-by: trustme
|
| |
|
|
|
|
|
|
|
|
|
|
| |
When generating Windows Makefiles, qmake writes out a .rc file for each
of debug and release (unless you've limited to just one build type).
When doing a clean shadow build, the first .rc file is written into a
directory that does not exist but the code was not handling the error
case. The fix does 2 things.
1) Attempt to create the destination directory if we can't write the file.
2) Die with an error if we still can't write the file after doing #1.
Reviewed-by: Marius Storm-Olsen
|
| |
|
|
|
|
|
|
| |
In the function initCompilerTool we handled QMAKE_CXXFLAGS twice for
every configuration (debug / release).
The call of parseOptions before the if clause is enough.
Reviewed-by: mariusSO
|
| |
|
|
|
| |
doesn't seem to be a terribly popular feature, given that nobody noticed
this yet ...
|
| |
|
|
|
|
|
| |
Removed dead code and simplified conditionals. This should not otherwise
change behavior or output of qmake in any way.
Reviewed-by: mariusSO
|
| |
|
|
|
|
|
|
|
|
| |
And some indentation fixes in qmake itself.
Fixes indentation; TRUE => true; FALSE => false; #includes and #defines
cleanup; removes dead code; comments... and somewhere along the way
marked a function static.
Reviewed-by: mariusSO
|
| |
|
|
|
|
|
| |
When generating nmake makefiles, the same inference rules were
generated several times.
Reviewed-by: mariusSO
|
| | |
|
| |
|
