The lengths are unsigned according to the specification. Passing
negative values can lead to data corruption.
Signed-off-by: Matthieu Herrb <matthieu@herrb.eu>
Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
cbb59d172 ('Braille: Fix typing quickly') broke the default lookup that
translates Braille keysym patterns to Braille Unicode patterns since it
rightfully clears brl_committing, but then we do not have it any more to
fill brl_committed.
This change saves the committed pattern so we can return it in the
default lookup.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
On finishing releasing Braille keys, we should clear the just-commited
pattern, to reset the state to initial state, and avoid having to wait for
0.3s before typing the next pattern.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Tested-by: Jean-Philippe Mengual <jpmengual@hypra.fr>
The _XimCacheStruct structure is followed in memory by two strings containing
fname and encoding. The memory was accessed using the last member of the
structure `char fname[1]`. That is a lie, prohibits us from using sizeof and
confuses checkers. Lets declare it properly as a flexible array, so compilers
don't complain about writing past that array. As bonus we can replace the
XOffsetOf with regular sizeof.
Fixes GCC8 error:
In function 'strcpy',
inlined from '_XimWriteCachedDefaultTree' at imLcIm.c:479:5,
inlined from '_XimCreateDefaultTree' at imLcIm.c:616:2,
inlined from '_XimLocalOpenIM' at imLcIm.c:700:5:
/usr/include/bits/string_fortified.h:90:10: error: '__builtin_strcpy'
forming offset 2 is out of the bounds [0, 1] [-Werror=array-bounds]
return __builtin___strcpy_chk (__dest, __src, __bos (__dest));
Caused by this line seemingly writing past the fname[1] array:
imLcIm.c:479: strcpy (m->fname+strlen(name)+1, encoding);
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
You can save a bit of code. The is no need to check XFree arguments bring free_fontdataOM in line with other free function and check for NULL arg
Signed-off-by: harms wharms@bfs.de
Incorrect parameter usage with sizeof. Earlier passed argument FontData
will be 4 bytes always as its a pointer hence the change is needed and
FontDataRec should be used for memset.
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
If we hit the depth limit, filename leaks. Move the depth check up before we
allocate filename.
Introduced in 226622349a.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
The Compose format has a feature which allows specifying certain
modifiers must (or must not) be present with a given keysym in the
sequence.
The grammar in imLcPrs.c and the Compose man page both do not match what
the code actually does (see the handling of the variables
`modifier_mask` and `modifier` in parseline() in imLcPrs.c, which are
eventually matched as `ev->state & modifier_mask == modifier`).
Also explicitly list the accepted modifier names, since they are
not standard (e.g. "Ctrl" instead of "Control").
Signed-off-by: Ran Benita <ran234@gmail.com>
Signed-off-by: James Cloos <cloos@jhcloos.com>
* Do not use variables before checked for NULL.
* remove some superfluid spaces (Mark Kettenis)
Signed-off-by: Harms <wharms@bfs,de>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
* Do not use variables before checked for NULL.
Signed-off-by: Harms <wharms@bfs,de>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
See http://sourceware.org/bugzilla/show_bug.cgi?id=10948
Currently, if the locale is UTF-8, no CJK fonts are installed, and someone
does XCreateFontSet() with a font name of "*", we end up asking the server
to list the (non-existent) fonts 11 times for each CJK encoding, which can
take a while.
A * wildcard can match multiple components in a XLFD name in XListFonts(),
so there's no need to try adding more than one to get a match.
We do try once with a leading '*-' in case the fontname isn't a full
well-formed XLFD name, maybe even that isn't needed?
(See also http://invisible-island.net/xterm/xterm.faq.html#slow_menus)
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
C89 or bust! This was documented as being needed for "only Lynx,
Linux-libc5, OS/2" and has never been enabled in modular builds,
since none of those platforms have had anyone step up to add support
since the X11R7 conversion to autotools.
Mostly performed with unifdef -UX_LOCALE, followed by removal of files
left without any purpose, and manual cleanup of remaining references.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Left one cast behind that is necessary to change from const char *
to char * in src/xlibi18n/lcCharSet.c.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
imRm.c: In function '_XimSetICMode':
imRm.c:2419:37: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
imRm.c:2420:30: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
lcGenConv.c: In function 'byteM_parse_codeset':
lcGenConv.c:345:13: warning: cast discards '__attribute__((const))' qualifier from pointer target type [-Wcast-qual]
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Fix bogus timestamp generted by XIM due to uninitialized
data field. Also set appropriate serial, too.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=39367
Signed-off-by: Chiaki ISHIKAWA <ishikawa@yk.rim.or.jp>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
XSetICFocus() and XUnsetICFocus() are both asynchronous events.
This is a pretty stupid idea: those functions may undo certain
settings on the client side for which requests from the server
may still be in the queue unprocessed. Thus things may be set
in the wrong order ie instead of set -> unest it will be unset -> set.
Moreover there is no way for either the client or the server to
cause the event queue to be flushed - which is pretty bad as
XIM is bidirectional.
The scenario is as follows:
Two ICs are created:
ic1 = XCreateIC(im,
XNInputStyle, XIMPreeditCallbacks | XIMStatusCallbacks,
XNClientWindow, window,
XNPreeditAttributes, preedit_attr,
XNStatusAttributes, status_attr,
NULL);
ic2 = XCreateIC(im, XNInputStyle,
XIMPreeditNothing | XIMStatusNothing,
XNClientWindow, window, NULL);
Then the focus is removed from ic2:
XUnsetICFocus(ic2);
If SCIM is used as the input server it will send a bunch of requests
following an XCreateIC(). One of the requests registers a key release
filter. XUnsetICFocus() unsets both key press and release filters.
Since it is asynchronous, the input server requests to register key
press and release filters may not have been processed, when XUnsetICFocus()
is called. Since there is no explicite way for client programs to enforce
the request queue to be flushed explicitely before an X[Set/Unset]ICFocus()
call it would be safest to make those two calls synchronous in the sense
that they ensure the request queue has been handled before they execute.
The easiest way to do this from Xlib is thru a call to XGetICValues()
which sends a request to the server and subsequently reads the queue
from the server to the client. This will cause all outstanding requests
in the queue to be read and handled.
This is an ugly hack and this could be fixed directly in the client,
however it seems to be easier to fix Xlib than to fix numerous clients.
This problem arose since there is no well documented way how to handle
and synchronize XIM requests and not all input servers send requests
when an IC is created.
This has been discussed extensively in:
https://bugzilla.novell.com/show_bug.cgi?id=221326
Signed-off-by: Egbert Eich <eich@freedesktop.org>
When synthesized key events are sent on commit XIM sets the 'fabricated'
flag so that the keypress handler knows that these were not real events.
This also happens when committing due to the loss of focus. However in this
case the keypress/release filters which consume and unset this flag are no
longer in the filter chain.
So the flag is erronously set when a real keyboard event is received after
focus has been regained. So the first event is wrongly treated as a
fabricated key in the keypress handler which will at the same time reset
the flag so the second key event is treated correctly.
This fix only sets the flag when at least one of the keyboard filters is in
place.
How to reproduce this bug: run scim, choose a Japanese input method start
two instances of xterm: start typing in one xterm (this should pop up an
IM window). Without comitting (hitting 'enter') move focus to the other
xterm, then move focus back. Start typing again. The first character will
be committed immediately without popping up an input window.
With this fix this behavior is gone.
See also: https://bugzilla.novell.com/show_bug.cgi?id=239698
Signed-off-by: Egbert Eich <eich@freedesktop.org>
It seems useless to do that since the code tests for both source
length and destination to be non-zero. This fixes a cut'n'paste
problem in xterm where the paste length was limited to 1024 (BUFSIZ)
in button.c.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
When trying to process file paths the tokens %H, %L, & %S are expanded
to $HOME, the standard compose file path & the xlocaledir path.
If enough of these tokens are repeated and values like $HOME are set to
very large values, the calculation of the total string size required to
hold the expanded path can overflow, resulting in allocating a smaller
string than the amount of data we'll write to it.
Simply restrict all of these values, and the total path size to PATH_MAX,
because really, that's all you should need for a filename path.
Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
Called from _XimCreateDefaultTree() which uses getenv("XCOMPOSEFILE")
to specify filename.
If the size of off_t is larger than the size of unsigned long (as in
32-bit builds with large file flags), a file larger than 4 gigs could
have its size truncated, leading to data from that file being written
past the end of the undersized buffer allocated for it.
While configure.ac does not use AC_SYS_LARGEFILE to set large file mode,
builders may have added the large file compilation flags to CFLAGS on
their own.
size is left limited to an int, because if your Xim file is
larger than 2gb, you're doing it wrong.
Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
parseline() can call _XimParseStringFile() which can call parseline()
which can call _XimParseStringFile() which can call parseline() ....
eventually causing recursive stack overflow and crash.
Limit is set to a include depth of 100 files, which should be enough
for all known use cases, but could be adjusted later if necessary.
Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
Check the provided buffer size against the amount of data we're going to
write into it, not against the reported length from the ClientMessage.
Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Matthieu Herrb <matthieu.herrb@laas.fr>
MUSTCOPY seems to have only been defined in <X11/Xmd.h> when building for
CRAY, to handle missing some sizes of integer type.
(mostly performed with unifdef, followed by some manual cleanup of
spacing/indenting in the remaining code)
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Has never been converted to build in modular builds, so has been unusable
since X11R7.0 release in 2005. DNETCONN support was removed from xtrans
back in 2008.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Has never been converted to build in modular builds, so has been unusable
since X11R7.0 release in 2005. All known platforms with TLI/XTI support
that X11R7 & later releases run on also have (and mostly prefer) BSD
socket support for their networking API.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Private is a struct member name in mingw-w64 <winioctl.h>, causing this
useless define in a private header to break the build.
Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
Reviewed-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Currently, only non-modifier keys (actually, keysyms) can be part of a compose
sequence, and they are matched against the defined compose sequences at the
time the key is pressed. The patch allows to use modifier keys an well, but
matches them on key release, and only if no other key has been pressed after
the modifier.
Releasing a non-matched modifier during an ongoing compose sequence only aborts
the sequence if any modifier release would have matched. In particular, if no
compose sequences with modifiers are specified, the compose mechanism works
exactly as without this patch.
Even if modifiers are part of a compose sequence, they are not filtered. This
is because modifiers affect the keyboard state no matter what we do here and,
therefore, filtering them only could confuse clients.
The purpose is this extension to the compose mechanism is to allow to make
better use of keys in convenient reach for touch typing.
Signed-off-by: Andreas Wettstein <wettstein509@solnet.ch>
Signed-off-by: James Cloos <cloos@jhcloos.com>
NEED_SYNC_REPLY flag should be in Xim not in Xic.
Because the focused Xic can be changed before sending sync reply.
After focused Xic changed, the new Xic doesn't have NEED_SYNC_REPLY
flag enabled, so libX11 doesn't send XIM_SYNC_REPLY packet.
This patch adds sync reply flag to Xim and removes sync reply
from Xic.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=7869
Signed-off-by: Choe Hwanjin <choe.hwanjin@gmail.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
In case of error, XSetICValues() must return the first argument
that failed to be set.
But in some error paths, it returns False, which is converted to NULL,
so the function returns OK in case of error.
Signed-off-by: Yann Droneaud <yann@droneaud.fr>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
As stated in man page (XOpenIM) and Xlib documentation (chapter 13.5.3),
XGetIMValues() and XSetImValues() "returns the name of the first argument
that could not be obtained."
But currently,
err = XGetIMValues(im, "invalid", &arg, NULL);
returns &arg instead of "invalid".
This patch fixes https://bugs.freedesktop.org/show_bug.cgi?id=12897
Signed-off-by: Yann Droneaud <yann@droneaud.fr>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
