Skip to content
GitLab
  1. 16 Jan, 2013 1 commit
  3. 05 Jan, 2013 1 commit
  5. 05 Nov, 2012 1 commit
  7. 01 Nov, 2012 1 commit
    • Daniel Stone's avatar
      Reset scan state when opening a new file · a68c5b7b
      Daniel Stone authored 
      When we opened a new file after encountering a parse error, we weren't
resetting our scan state, and continued to read out of the old buffer
rather than that of the new file.

Fixes a regression introduced in 9887842e

 ('Use fread() instead of
getc()').

Test case:
    xkbcomp -lfhlpR "path/to/xkeyboard-config.git/symbols/*"
The Makefiles in that tree would cause parse errors, the resulting directory
file is incomplete (down from ~12000 to ~230 lines).

Reported-by: Peter Hutterer's avatarPeter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      a68c5b7b
  9. 31 Oct, 2012 1 commit
  11. 23 Mar, 2012 1 commit
  13. 29 Dec, 2011 2 commits
    • Alan Coopersmith's avatar
      Delete redundant redeclarations of extern Atoms in symbols.c · b4fafc4f
      Alan Coopersmith authored 
      

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's avatarAlan Coopersmith <alan.coopersmith@oracle.com>
      b4fafc4f
    • Alan Coopersmith's avatar
      Make sure to leave room for trailing nil byte in yyGetNumber · 3835cae3
      Alan Coopersmith authored 
      ...though really, by the time you've added 1023 digits to the number
you want to parse, you've got much bigger problems than an off-by-one
error in your buffer count.

Fixes parfait warnings:
   Buffer overflow (CWE 120): In array dereference of (*buf)[nInBuf] with index 'nInBuf'
      Array size is 1024 bytes, nInBuf >= 1 and nInBuf <= 1024
        at line 625 of xkbscan.c in function 'yyGetNumber'.
   Buffer overflow (CWE 120): In array dereference of (*buf)[nInBuf] with index 'nInBuf'
      Array size is 1024 bytes, nInBuf <= 1025
        at line 632 of xkbscan.c in function 'yyGetNumber'.

[ This bug was found by the Parfait 0.4.2 bug checking tool.
  For more information see http://labs.oracle.com/projects/parfait/

 ]

Signed-off-by: Alan Coopersmith's avatarAlan Coopersmith <alan.coopersmith@oracle.com>
      3835cae3
  15. 06 Dec, 2011 1 commit
  17. 11 Nov, 2011 6 commits
  19. 07 Jul, 2011 1 commit
  21. 22 Jun, 2011 1 commit
  23. 21 Jun, 2011 6 commits
    • Daniel Stone's avatar
      Bump to 1.2.3 · cef4ba4b
      Daniel Stone authored 
      

Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      cef4ba4b
    • Daniel Stone's avatar
      Interp: Allow explicit Any/NoSymbol mappings · 3caab5aa
      Daniel Stone authored 
      

Brown paper bag in full effect.

The previous fix, while crushing the previous problem where an unknown
keysym for an interp def would lead to every key matching it, also
ignored explicit Any+AnyOfOrNone(All) mappings.

Such as the one xkeyboard-config relied on for Control to actually
update the modifier state.

Fix this by allowing mappings explicitly declared as Any/NoSymbol, while
ignoring only those with failed keysym lookups.  Unfortunately, due to
the structure of the parser, it's a deeply inelegant fix.

Verified with a quick check of all layouts (albeit using default
variants only) in xkeyboard-config that this results in no changes to
the output at all, compared to xkbcomp 1.1.1.

Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      3caab5aa
    • Daniel Stone's avatar
      Constify LookupKeysym input argument · b34af8b0
      Daniel Stone authored 
      

Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      b34af8b0
    • Daniel Stone's avatar
      Bump to 1.2.2 · e68a8ef1
      Daniel Stone authored 
      

Emergency bugfix for the syminterp declarations.  They've been broken
since the beginning of time, but I guess we just haven't added new
interpreted symbols since around then.

Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      e68a8ef1
    • Daniel Stone's avatar
      Interp: Ignore NoSymbol definitions · 2a473b90
      Daniel Stone authored 
      

A NoSymbol syminterp leads to a Any+AnyOfOrAll(None) definition, which
is never what anyone would actually ever want in a sym interp.  The
failure mode here was:
    interpret XF86SomeKeysymYouDontHaveYet {
        action=Something()
    }
leading to:
    interpret Any+AnyOfOrAll(None) {
        action=Something()
    }
leading to every single non-action key triggering your shiny new action
that was only supposed to come from a new keysym.

At least you could VT switch or zap, I guess.

Later, more invasive, revisions will bring a more useful error message.

Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      2a473b90
    • Daniel Stone's avatar
      Interp: Don't make modifier lookup failure fatal · a2816345
      Daniel Stone authored 
      

If we can't look up a modifier in a SymInterp declaration, don't make
that fatal enough to kill the entire file, just ignore it and move on.

Signed-off-by: Daniel Stone's avatarDaniel Stone <daniel@fooishbar.org>
      a2816345
  25. 16 Jun, 2011 2 commits
  27. 09 Jun, 2011 1 commit
  29. 03 Apr, 2011 1 commit
  31. 13 Feb, 2011 1 commit
  33. 10 Feb, 2011 1 commit
  35. 19 Jan, 2011 1 commit
  37. 13 Jan, 2011 1 commit
  39. 12 Jan, 2011 2 commits
  41. 04 Jan, 2011 1 commit
  43. 01 Nov, 2010 3 commits
  45. 30 Oct, 2010 1 commit
  47. 24 Sep, 2010 2 commits