summaryrefslogtreecommitdiff
path: root/symbols.c
AgeCommit message (Collapse)Author
2024-05-04Symbols: resize the syms array when removing NoSymbolHEADmasterMatthieu Herrb
On OpenBSD, with malloc_conf=S, the recallocarray() code detects a size mismatch in ResizeKeyGroup when numLevels[group] doesn't match the allocated size of syms[group]. Should fix https://gitlab.freedesktop.org/xorg/app/xkbcomp/-/issues/25 Signed-off-by: Matthieu Herrb <matthieu@herrb.eu> Part-of: <https://gitlab.freedesktop.org/xorg/app/xkbcomp/-/merge_requests/24>
2023-01-03Mark more pointers as constAlan Coopersmith
Some suggested by cppcheck, others by manual code inspection Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2023-01-03Use unsigned ints when shifting to create bitmasksAlan Coopersmith
symbols.c:1057:28: portability: Shifting signed 32-bit value by 31 bits is implementation-defined behaviour. See condition at line 1049. [shiftTooManyBitsSigned] radio_groups |= (1 << (tmp.uval - 1)); ^ symbols.c:1049:41: note: Assuming that condition 'tmp.uval>32' is not redundant if ((tmp.uval < 1) || (tmp.uval > XkbMaxRadioGroups)) ^ symbols.c:1057:28: note: Shift radio_groups |= (1 << (tmp.uval - 1)); ^ symbols.c:1057:28: warning: Either the condition 'tmp.uval>32' is redundant or there is signed integer overflow for expression '1<<(tmp.uval-1)'. [integerOverflowCond] radio_groups |= (1 << (tmp.uval - 1)); ^ symbols.c:1049:41: note: Assuming that condition 'tmp.uval>32' is not redundant if ((tmp.uval < 1) || (tmp.uval > XkbMaxRadioGroups)) ^ symbols.c:1057:28: note: Integer overflow radio_groups |= (1 << (tmp.uval - 1)); ^ Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Remove unnecessary casts from bzero() callsAlan Coopersmith
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Remove unnecessary casts from memcpy() callsAlan Coopersmith
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Remove unnecessary checks for NULL pointers before calling free()Alan Coopersmith
Not needed in C89 and later Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Replace uFree() with direct free() callsAlan Coopersmith
All these wrappers did was mess with types and add a test for NULL pointers that isn't needed in C89 and later. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Replace uTypedRecalloc() with direct recallocarray() callsAlan Coopersmith
Retains uRecalloc() as a fallback for platforms without recallocarray() Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Replace uCalloc() and uTypedCalloc() with direct calloc() callsAlan Coopersmith
All these wrappers did was mess with types. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Replace uAlloc() and uTypedAlloc() with direct malloc() callsAlan Coopersmith
All these wrappers did was mess with types. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11Variable scope reductionsAlan Coopersmith
Some found by cppcheck, some found by manual code inspection Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-10Remove register keyword from variable declarationsAlan Coopersmith
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-07-11suppress four more warnings when 'warningLevel' is zeroBenno Schulenberg
This addresses issue #20 some more. Reported-by: Vincent Lefevre Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2022-07-11suppress the "Could not resolve" warning when 'warningLevel' is zeroBenno Schulenberg
Two years ago, commit c8cfca25ab changed an error to a warning but forgot to add a check for 'warningLevel'. This partially addresses issue #20. Reported-by: Vincent Lefevre Signed-off-by: Benno Schulenberg <bensberg@telfort.nl>
2021-11-28Fix "upercase" typoAlan Coopersmith
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2021-01-21Downgrade the warning for missing symbols to infoPeter Hutterer
In the interest of maintainability, it's easier to include as many keycodes as possible and then have the symbols mapping specific to the layout. This is particularly true for evdev where the kernel takes care of device-specifics and every keyboard has the same set of keycodes anyway. So let's downgrade this from a warning to a mere info, virtually every keyboard right now triggers this warning for a number of keys. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-01-21Replace WARN[1-9], ERROR[1-9], etc. with their unnumbered versionPeter Hutterer
Those macros date back to when varargs weren't a thing but they've been #defined to the same value for 17 years now. Patch generated with: for action in WARN INFO ERROR ACTION FATAL WSGO; do sed -i "s/${action}[1-9]/${action}/g" `git ls-files` done Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2021-01-21Remove trailing whitespacesPeter Hutterer
Let's clean this up so I don't have to fight vim and git in removing them. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2020-07-23Fix spelling/wording issuesAlan Coopersmith
Found by using: codespell --builtin clear,rare,usage,informal,code,names Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2020-06-05Don't pretend unresolved symbols are an errorPeter Hutterer
Whenever xkeyboard-config is newer than xorgproto, or libX11 just hasn't yet been rebuilt against the protocol we end up with some unresolved symbols. That's not an error, it just happens. Let's downgrade this to a warning instead. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2015-11-09When overriding a key, adjust also its number of levels (#57242).Benno Schulenberg
Specifying an explicit key type when overriding a key should adjust the number of levels to that of the specified type. This gets rid of the age-old warning of the right Alt key being ONE_LEVEL but having two symbols assigned. Fixes bug #57242 <http://bugs.freedesktop.org/show_bug.cgi?id=57242>. Signed-off-by: Benno Schulenberg <bensberg@justemail.net> Reviewed-by: Ran Benita <ran234@gmail.com> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2015-11-04symbols: increase the warning level for shortening a key typePeter Hutterer
Any use of the german keyboard layout or anything else using level3(ralt_switch) produces the warning Type "ONE_LEVEL" has 1 levels, but <RALT> has 2 symbols Ignoring extra symbols This warning is meaningless, the default definition for RALT comes from pc(pc105) including altwin(meta_alt) which defines it as TWO_LEVEL and { Alt_R, Meta_R }. Including level3(ralt_switch) correctly sets it as ONE_LEVEL but now we get the warning. That makes users file bug reports and we can't be having that, can we? Up the warning level for this case to one past the default. The warning is only useful when you're testing a new keyboard layout or trying to fix an actual bug with missing symbols. And then you can run xkbcomp with -w10. No need to spam the log on every server startup. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Hans de Goede <hdegoede@redhat.com>
2013-11-01Fix many const char * warnings from gccAlan Coopersmith
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
2011-12-28Delete redundant redeclarations of extern Atoms in symbols.cAlan Coopersmith
Fixes gcc warnings of: symbols.c:43: warning: redundant redeclaration of 'tok_ONE_LEVEL' tokens.h:99: warning: previous declaration of 'tok_ONE_LEVEL' was here symbols.c:44: warning: redundant redeclaration of 'tok_TWO_LEVEL' tokens.h:100: warning: previous declaration of 'tok_TWO_LEVEL' was here symbols.c:45: warning: redundant redeclaration of 'tok_KEYPAD' tokens.h:102: warning: previous declaration of 'tok_KEYPAD' was here Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2011-11-11Silence -Wparentheses-equalityJeremy Huddleston
warning: equality comparison with extraneous parentheses [-Wparentheses-equality] Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2010-06-15Lazy keysym parsing (avoid XStringToKeysym)Daniel Stone
Instead of calling XStringToKeysym on every keysym we parse, store it as a string until we need to store it in an actual keymap. Signed-off-by: Daniel Stone <daniel@fooishbar.org>
2008-08-22Don't overwrite previously assigned type indices with useless ones.Peter Hutterer
The default type for a key with no levels is ONE_LEVEL. Let's not overwrite the kt_index with this default type if we have previously assigned a real type. Reproduceable by running setxkbmap -layout "ru(phonetic),us", the first group is assigned ONE_LEVEL and shift stops working. Red Hat Bug #436626 <https://bugzilla.redhat.com/show_bug.cgi?id=436626>
2008-08-21More comments.Peter Hutterer
2008-08-18Add some explanatory commentsPeter Hutterer
2008-08-12Indent fixes.Peter Hutterer
indent -cbi 0 -nprs -nut -npcs -i4 -bli 0 *.c *.h
2008-08-12Remove RCS tags.Peter Hutterer
2007-09-23Fixed a bunch of const correctness bugs.Tilman Sauerbeck
2004-04-23Merging XORG-CURRENT into trunkXORG-6_7_99_904XORG-6_7_99_903XORG-6_7_99_902XORG-6_7_99_901XORG-6_7_99_2XORG-6_7_99_1XACE-SELINUX-MERGEEgbert Eich
2004-03-14Importing vendor version xf86-4_4_99_1 on Sun Mar 14 00:26:39 PST 2004xf86-4_4_99_1Egbert Eich
2004-03-03Importing vendor version xf86-4_4_0 on Wed Mar 3 04:09:24 PST 2004xf86-4_4_0Egbert Eich
2004-02-26readding XFree86's cvs IDsxf86-4_3_99_903Egbert Eich
2004-02-26Importing vendor version xf86-4_3_99_903 on Wed Feb 26 01:21:00 PST 2004Egbert Eich
2003-11-25XFree86 4.3.99.16 Bring the tree up to date for the Cygwin folksxf86-4_3_99_901xf86-4_3_99_16Kaleb Keithley
2003-11-14XFree86 4.3.0.1xf86-4_3_0_1PRE_xf86-4_3_0_1Kaleb Keithley
2003-11-14R6.6 is the Xorg base-lineXORG-MAINKaleb Keithley
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-24 18:11:00 +0000