As reported by Moritz, the Qt toolkit (and maybe a few other ones) are
using the content of this property, but WindowMaker did not set it properly
so for examples the menus in KDE could fail to display in multi-screen
configuration (probably because the toolkit is using the property to make
sure the menu stays in the visible area).
The original code just assumed it was safe to use the usable area of the
1st screen in the list, but unfortunately it is not that simple.
The new code calculates a rectangle that contains the usable region from
all the screens, so the toolkit gets a more accurate value.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As found by Douglas Torrance, when the 'configure' script is generated
using v2.68 of autoconf then it gets wrongly generated due to a regression
in the handling of names in AS_VAR_PUSHDEF, and crashes with this kind of
sibylline messages:
checking CFLAGS for -Wtrampolines... ./configure: line 11916: wm_cv_c_check_compopt_Werror_trampolines=no, unknown: command not found
This patch adds a check on autoconf version to ensure the problem will get
caught as soon as possible.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, this macro is not used anywhere in the 2 files. Because
the code also suggests that they should be inherited from any other place
anyway, remove them from the file to prevent them from being used in the
future.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Apparently, there was a time when it was though to add a label below (?)
the icons of the list of panels, but this has never been finished and the
unfinished code to do this was even removed in commit
0e01e69205
So, as it is unlikely that the "feature" will return, and if it did the
font should be handled in a better way, this patch removes the definition
of the constants with the names for the font used for those labels.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Include a comment to explain what their aim is, and why WindowMaker is not
making use of them. They are not simply removed because it is informative
for future contributors that we know about them and we purposedly decided
to not use them.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
This parameter is given always the same value, so there is not reason to
keep it. The code is updated to take this value into account, and the
related constants are removed.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
These 2 constants have been defined at the beginning of the project, but
where never used later. As the use of the associated constants suggests
that they are not bringing anything useful, this patch simply removes them.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The goal is to explain the reason why we don't do anything about them, so
people looking at the code in the future will know why it is this way. The
expected side effect is also to silent warnings [-Wunused-macros] from gcc.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, this constant is not used anywhere in the code. It seems
to have been prepared for a scroll animation when the switchpanel would be
too big to fit the screen, but the code looks like to never have been
activated, and was cleaned away in commit
6d08aa22d8
If the feature were to be re-added, it would be better to use something
proportional to the menu scroll speed parameter that can be configured by
user anyway.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, this macro is not used anywhere, that is probably
because the code have been relying on the function "IsDoubleClick" for a
very long time now.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
For code maintainability, it is better to have a single definition of the
constant than many ones dispatched in many places. In addition, we try to
count on the constant M_PI that can be defined by the header <math.h> if
possible because it may have the best accuracy for the platform.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, the 4 constant the define the corner positions are not
used anywhere in the code, and it looks like it has always been like this.
As it is a cost for maintainability, get rid of them.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, a few constants which contain sizes for buffers are not
used anywhere in the code. This is probably due to code cleanup, where
these buffers have been removed or where size is automatically calculated
by using 'sizeof' instead to reduce risk of bugs.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
There is a header that defines the stucts and constants to support GNUstep
specific things, and it is already used somewhere else in WINGs, so instead
of duplicating locally the stuff we re-use the header, it makes code easier
to maintain.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The group of XDND_* macro do not bring a lot in term of clarity, so it is
probable that they will get away someday, but at current time as 2 of them
are not used anywhere they get removed to prevent their use in the future.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
This macro was define a *very* long time ago, in commit
d98f1fa645
but was not used at that time, and have never been used anywhere since
then.
As the macro does not look like a good idea for performance anyway, get rid
of it before anyone could get tempted to use it.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
It seems the code had the possibility to load a default icon, probably to
be used on the windows if the main program does not provide any.
This code however have never been enabled, probably because it is better to
not provide an icon and let the window manager use its own.
This patch then removes that dead code, but keeps the image as part of the
WINGs resource for the case where an application would have been using it.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, this macro is never used in the code; when no font is
defined by user the 'paintButton' code already falls back to the
'normalFont' of the screen, and this code could not need the macro anyway.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, this constant is not used anywhere, and it looks like
it have always been the case. As there's no reason to specifically add a
border to this widget, remove the constant.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, the height is never used and the code make use of the
SCROLLER_WIDTH constant anyway, so this patch updates the code to directly
use that constant instead of intermediate values that just adds noise for
code maintainability.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by gcc, the macro is not used, that is probably because it is
totally ok to never provide a title from X point of view, so there is now
reason to fall back to that empty constant.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Because the use of trampolines can be a lot of problems, this patch changes
the use of the flag to:
- make it an error if possible, so coders are forced to not use them;
- always enable them, not just for debug, so we increase the probability
to get them caught.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Strict prototype are better for portability and to avoid bugs because it
makes sure the compiler has the information to properly validate the
arguments given when a function is called.
This flag however need special care when checking for it, because the
declaration for 'main' generated by autoconf cannot be a strict prototype
so the detection would always see the flag as failing.
This patch handles this by creating a dedicated macro for this detection
which uses a good prototype because we're in a case where it is possible,
so the detection will not always fail; it also makes sure to add the flag
to CFLAG only at the end, to avoid falsely crashing any further test done
in the configure script.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
This flag is always detected as unsupported because every time autoconf is
generating a test program for any feature, it puts at the beginning of the
test source all the '#define' that have been detected so far, which is what
we expect normally.
But for this option, as we cannot reasonably make a dummy use of every
macro, the warning triggers and falsely gets autoconf into thinking it does
not work.
This patch creates a dedicated macro (WM_CFLAGS_GCC_OPTION_UNUSEDMACROS)
for this flag, which works around the problem by having no '#define' in the
test source. It also adds a new macro WM_CFLAGS_GCC_OPTION_POSTPONED
because if the flag works, we still cannot add it to the compilation
command because it could fail all further tests done, so the macro will add
it to CFLAGS only at the end.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Apparently, some pixels from the close button on the window image were
missing from the Non-opaque image, they have been added in the XPM file
which was then converted to TIFF with ImageMagick's command:
convert -depth 8 -compress lzw xpm/nonopaque.xpm tiff/nonopaque.tiff
to have a file similar to what was there before
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by Coverity, the PLStrings created to store the information on
the texture leak. This is due to the fact that they are created with a
refCount of 1, then the PLArray in which they are placed increments that
count, so at list destruction the count would return to 1 instead of 0,
meaning the PLStrings won't be freed.
This patch release the PLStrings once after adding them to the PLArray so
the count will go back to 1, which means they will be properly freed when
the PLArray will be released.
Took opportunity to remove the call to WMRetainPropList on the titem-prop
because it artificially increases the refCount but this is already done
when adding to the PLArray.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Recent version of perl seem to report incorrect stuff in regular expression
and this seems to be fixed in the kernel reference file, so this patch
brings the fix to our (older) version so we won't get that spurious message
when running the script.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
A macro can be a source of problems, because the compiler has no type on
the arguments to make checks. Using an inline function allows to do those
checks, meaning clearer error messages, it provides clear info in case of
name collision, it is easier to maintain (no need for the hacky '\' for
multi-lines) and the scope of visibility can be controlled more easily (no
need for #undef).
The macro store_modifier had to face a slight change because its 2nd
parameter is used as a reference, which is now clearly visible in the
prototype of the function.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
A macro can be a source of problems, because the compiler has no type on
the arguments to make checks. Using an inline function allows to do those
checks, meaning clearer error messages, it provides clear info in case of
name collision, it is easier to maintain (no need for the hacky '\' for
multi-lines) and the scope of visibility can be controlled more easily (no
need for #undef).
Took opportunity to change a 0 to the constant NoSymbol which is the name
defined by X for this case and another to NULL which is the right way to
set a null pointer in C.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
A few files are generated using custom commands, which so far did not
follow the silent rule as the compilation stuff.
This patch adds the needed stuff so they will also be silent if the user
wants so, leading to a cleaner build process where warning/error messages
are more visible.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The function is called only if wsmap is not null, and the function is not
modifying its value so it won't become null. Removed the checks to keep the
code as simple as possible for maintainability.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The definition in the local header was not correct; it works because gcc is
tolerant to this kind of errors but other compilers are not. The
declaration was creating a local variable in each file that call header,
and because it is not static gcc's linker will merge them. Other compilers
will at best complain for duplicate symbol, and at worst silently duplicate
the variable so it will not work as expected.
The variable is now moved to the existing structure meant for global
variables, so now the code is really clear about using a global variable
instead of a static/local one.
Took opportunity to add some missing 'static' attributes to some variables.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
This patch is adding the core code needed to run the workspace pager.
Currently when called, a new frame presenting mini workspace images
are displayed at the bottom of the screen above other windows.
When clicking on one of the miniatures, the workspace is switched and
the frame is closed.
Drawbacks: a screenshot of the workspace is used for the miniatures.
This screenshot is taken only when a workspace switched event occurs.
First, it means that the workspace switching process can be longer than
usual. Secondly, updated minitatures can only be available when the
workspace is "opened" (cause windows have to be mapped to be able to
be copied).
So when wmaker is (re)started or when for example a window is moved to
another workspace the corresponding miniature is NOT updated.
(I did not find a clean and easy way to do so, feel free to share if
you have some ideas)
This path is adding the functions and variables needed for workspace pager preferences.
An option to enable/disable the workspace pager, and one to set a default background image.
This patch is adding a checkbox in the expert zone to disable
completely the workspace pager and add an entry in keyshortcut
preference to set the shortcut used to open the pager with the
action "Open workspace pager".
The default frame background can be configured by setting an optional
variable "WorkspaceMapBack" from WindowMaker conf file as in:
WorkspaceMapBack = (tpixmap, "/tmp/testme.png", gray20);
or
WorkspaceMapBack = (solid, "#2c2482");
We have a few directories with source codes that we tell configure to
prepare, but we do not actually want them built during normal operations
(tests and examples only).
However, there are some special targets brought by automake which still
need to see them, so this patch adds these directories to the list, but
only for these rules, we keep them unvisited by the normal build process.
The wanted side effect of this is that now "make distcheck" works untill
the end as expectable.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The code is making use of a few of the libm functions, but it looks like
gcc adds automatically the libm dependency (either by trying to be smart or
as an inherited dependency?).
Apparently, when compiling with clang-3.5 the function 'round' still needs
the use of math library (the others do not seem to), so this patch adds it
to the list of link libraries, which is more portable.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
There are a few cases in which nested functions are an helpful way to write
code, as this is explained in 'script/nested-func-to-macro.sh'. However,
some compiler do not support them (like clang), so this patch proposes an
elegant solution, where developers can get the benefit of them, but for
users they are automatically converted to C macro if needed.
The advantage of this solution is that we keep the code simple, there is no
hack in the source (like #ifdef and code duplication), yet still having the
full advantages.
The translation is done according to what have been detected by configure
(see the WM_PROG_CC_NESTEDFUNC macro) so that user has nothing to do.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The memory was allocated with wmalloc, so for consistency it should be
freed using wfree. This could be a problem if the user compiled with
support for Boehm GC, or if we later decide to add support for other malloc
libraries.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by Coverity, the function handles the case where the file to be
copied is not found by properly warning the user and the deleting the
currently built theme directory, but then it continued executing the file
copy that would crash on the null pointer.
This patch just adds the missing return that will avoid the crash.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Two WMColor were created to draw the background for the title on icons for
iconified windows, in the Appearance panel. As pointed by Coverity, these
colors were not released after use, which this patch fixes.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
The name of the color is stored in a newly created PLString which leaks.
This is due to the fact that they are created with a refCount of 1, then
the PLArray in which they are placed increments that count, so at list
destruction the count would return to 1 instead of 0, meaning the
PLString won't be freed.
This patch properly calls WMReleasePropList after addition to the list, so
that the count goes back to 1, which means it will be automatically freed
when the PLArray will be released.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
As pointed by Coverity, if dock is still null at this point then the
function will return via the previous check, because it is not possible to
have icon_pos >= 0 if dock is null.
This patch removes the check because it complicate the code which is not
recommended for maintainability.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Using a variable to store the return value that will be used later is not a
good idea, because it forces to track everywhere in the function when
needing to work on the function.
This patch removes the variables and places explicit return if each case,
so on first look it is clear where the code stops.
It also fixes a bug where the function would handle an event but still
returns False (meaning the event was not treated), whose root cause was
coming from the complexity brought by the variable.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
Coverity complain that there can be security issues because the variable
'i' is being modified using untrusted data (coming from a file). This is
probably pessimistic, because in the present case we're talking with the
kernel.
Using the correct signedness for the variable should however keep us safe,
and (I hope) make Coverity happy.
Took opportunity to include an error message in case of read problem
because it can help to debug.
Signed-off-by: Christophe CURIS <christophe.curis@free.fr>
