2009/2/22 Dominik Vogt <[email protected]>: > On Sun, Feb 22, 2009 at 12:24:14AM +0000, Thomas Adam wrote: >> 2009/2/22 Thomas Adam <[email protected]>: >> > 2009/2/21 Thomas Adam <[email protected]>: >> >> Hi -- >> >> >> >> 2009/2/21 Dominik Vogt <[email protected]>: >> >>> I've applied the patch after some modifications so that it's not >> >>> necessary to copy window names etc. But the code in charmap.c >> >>> needs some proof reading. I'm really not sur the memory >> >>> management is correct. >> >> >> >> I was *just* this moment cleaning that up -- I will ditch my attempts >> >> and take a look. >> > >> > Thanks for those improvements -- looks a little more sane now. I've >> > gone and removed some of the calloc() stuff in print_bindings() which >> > ought to make more sense now and commited those. >> > >> > Any problems, just shout -- I'm still a little cautious adding little >> > things like this to FVWM for fear of completely buggering it up, >> > although I suppose until someone shouts, I don't have to duck too low. >> > :P >> >> Oh -- and umm, without wanting to sound too pushy, can we have a >> 2.5.27 unstable release soon? I've read through the instructions in >> CVS on how to do this (assuming they're up to date) and whilst I could >> probably do it, I feel neither that confident, > > It's no big deal. If you follow the instructions precisely, you > shouldn't encounter any problems. > >> nor that I have the >> right to enforce such a decision, but I do think it's something which >> is due, especially as more and more stuff is added in CVS now. :) > > We should definitly try to get the print_bindings() stuff stable > first.
Oh sure. I ran it over my own bindings (as well as making a load up -- see file attached). Looks good, doesn't seem to crash anything. Is this the sort of thing I ought to be adding in a test case for into the build? (It seems that side of things is rarely used these days for new features.) -- Thomas Adam
keyTest
Description: Binary data
