Hi, [EMAIL PROTECTED] (2007-11-20 at 2305.37 +0200): > "Alex Smirnov" <[EMAIL PROTECTED]> writes: > > shift Shift_L (0x32), Shift_R (0x3e) > > lock Caps_Lock (0x25) > > control Control_L (0x42), Control_R (0x6d) > > mod1 Alt_L (0x40), Alt_L (0x7d), Meta_L (0x9c) > > mod2 Num_Lock (0x4d) > > mod3 > > mod4 Super_L (0x7f), Hyper_L (0x80) > > mod5 Mode_switch (0x5d), ISO_Level3_Shift (0x7c) > > And that layout does not work, producing state 0x2000 when you press > Alt_L? Very strange, because Alt_L is bound to mod1 and Mod1Mask is > 8, like you got with the other layout. 0x2000 does not correspond to > any defined modifier or mouse button mask (mouse button masks are only > defined up to Button5Mask == 0x1000).
If I understand what you are talking about, src/keys.h has up to Button9Mask, for example Button6Mask (1<<13) so 0x2000 (there is a patch to add buttons 10-13). Looking in X11/X.h I see Sawfish's Button8Mask has the same value than X.h's AnyModifier, and biggest official one is 5. I also see other WM also had issues with 5+. http://www.mail-archive.com/[EMAIL PROTECTED]/msg04995.html So in general, this needs a really deep inspection, maybe looking how toolkits are handling such big number of buttons when X11 core only prodives 5. GSR
