Fix wrong library name in wx_add_sample() for webrequest.
Also prefer to disable wxWebRequest if no backends for it are available,
for consistency with the other libraries.
It was never defined because USE_WIN32 wasn't set at the time it was
tested. Fix this by just always defining this option, even though it
only makes sense for MSW.
Also fix the test for winhttp.h, it needs to include windows.h first, as
otherwise compiling the header would always fail.
Using build system libraries for a different host doesn't make sense and
can be actively harmful, so set PKG_CONFIG_LIBDIR to prevent pkg-config
from finding them (it, or PKG_CONFIG_PATH, can still be set to some
host-appropriate directory manually before/when running configure).
This obviates the need for the changes in the previous commit, so revert
it.
This is unnecessary, it can be protected and we can initialize
wxWebRequest::m_headers directly in its ctor instead of using this
function (which also simplifies code and makes it impossible to forget
to do this).
Use BeforeFirst() when we only need to find the first colon instead of
wxSplit().
No real changes (except for pathological case when there is no colon at
all, which wasn't handled correctly by the original code and still
isn't, but in a slightly different way).
They're not necessary to use this class and we may consider exporting
them later, possibly with a better API and more tests, if really needed.
Also do change their API slightly by leaving only a single function and
returning the value instead of using an out parameter for it to make it
simpler to use.
Define wxWebSession::ms_defaultSession and ms_factoryMap in the
implementation file to avoid having to make the otherwise unnecessary
wxStringWebSessionFactoryMap type public.
No real changes.
The former can be called from the derived class ctors while the latter
only needs to be called from wxWebRequest itself, so just make it a
friend: this is not ideal, but still better than leaving this public and
simpler than any alternatives.
It doesn't make much sense to specify the conversion here, it would
ideally be taken from the response Content-Type header itself and
currently is just assumed to be UTF-8 anyhow.
Also implement fallback to Latin-1 to avoid losing the data entirely if
it's not in UTF-8.
Default ctor was leaving pointers uninitialized which was dangerous, so
merge it with the other ctor to ensure that we always set them to NULL.
Also make m_response const as it can't be changed after creating the
event.
It's not used anywhere else with constructors taking more than one
(non-optional) argument and is not really useful with them (and not
useful at all until C++17), so don't use it here neither for
consistency.
No real changes.
Using --with is more appropriate for libcurl and WinHTTP as they
introduce dependencies on external libraries and is done for
NSURLSession too for consistency.
And there doesn't seem to be any real reason to make these options names
longer by including "webrequest" in their names, they're clear and
unambiguous enough even without it.
This is mostly useful for disabling the use of WinHTTP, e.g. because
only libcurl-based backend will be used or because WinHTTP headers or
libraries are not available and so using it would fail in any case.
This is simpler and better (supports cross-compiling, manual override,
...) than doing it manually.
Also disable wxWebRequest completely under Unix if libcurl was not
found, as it's the only backend which can be used on non-MSW/Mac
platforms.
Check that the stream is valid, if specified at all, and return false if
it isn't -- or if no size was specified and determining stream size
failed.
Check for SetData() success in the test to provide better diagnostics in
case the file it uses is not found (as is the case when running the test
from another directory, for example).
Also pass wxSharedPtr<> by const reference instead of by value to avoid
unnecessary copies.
Show the condition which failed for wxASSERT() (as opposed to
wxASSERT_MSG()), as otherwise the error message didn't show any
information at all, making diagnosing the problem impossible.
Also show the assert location, as this can be useful too and there
doesn't seem to be any reason not to do it.
This provides the same information as wxGetCpuArchitectureName(), but is
consistent with the other wxPlatformInfo accessors, which also
correspond to the existing wxGetXXX() functions, and makes it more
convenient to use this information in the application code as now
everything is available via wxPlatformInfo instead of having to use
different functions for different pieces.
Calling SystemParametersInfoForDpi() with LOGFONTA parameter corrupted
the stack because this function only accepts LOGFONTW, i.e. it wrote
twice as much data into the provided buffer as the caller expected.
This could be fixed by allocating a temporary wide buffer and converting
back, but this doesn't seem to be worth it, so just disable the use of
the DPI-specific function in the ANSI build.
Closes#19002.
This results in various problems, including both the problems with
handling keyboard input (see #16840) and, even less expectedly,
graphical issues with flickering (see #18462).
Fixing or working around these issues doesn't seem to be easily
possible, but we can at least warn people about the problem.
Closes https://github.com/wxWidgets/wxWidgets/pull/2138
Add the #pragma to link with shlwapi in the same object file which uses
this function, as even though the same pragma was already present in
other files, they could be completely discarded by the linker resulting
in link errors about unresolved StrCmpLogicalW() later.
Also move the MinGW-specific function declaration to the top of the
file, near the inclusion of the header which would normally declare it.
The rect offset was correctly changed in 04a7a3f150 (Fix drawing outside
of the splitter sash under macOS, 2020-12-05), but its size was not. Fix
it too now.
See #18861.
As we don't use any special value for the inactive caption background,
use the same value as for normal caption text for inactive caption text
too.
This is almost certainly not ideal, but at least allows the tab labels
to be readable.
Closes#18988.
