Message 1 of 4
still performs erratically. It's not possible to determine any
pattern, but sometimes it works, sometimes not. If an entity is
rotated, undo will sometimes unrotate it, sometimes not.
Occasionally it will jump back 3 or 4 steps when 1 UNDO step is
specified. Sometimes a regen shows that UNDO has been successful,
sometime a regen deletes a copied entity. Same applies to DELETE,
MOVE,other commands too. It makes no difference how UNDO is invoked:
either by CtrlZ, typing UNDO or a lisp keyboard shortcut. A most
frustrating state of affairs which causes the loss of of work - anybody
with a solution please ? This has been the case since the first
installation of ZWCad and all subsequent patches failed to remedy the
Message 2 of 4
This might be relative to hardware and software configurations. Since it happens erratically in some PCs. I suppose you send the hardware and software configurations to firstname.lastname@example.org for professional help.
They should find out the common causes of these issues and solve them.Sirius Zhu2009-06-18 11:45:12
Message 3 of 4
Thanks- but the mystery was solved with the help of a contribution on the ZWCad mailing list at googlegroups: - my setup is to blame. My zwcad.lsp file is not standard, it's modified to load keyboard macros, all of them short lisp routines. This enables me to operate the program with both hands working fulltime : mouse right, keyboard shortcuts left. It seems that ZWCad feels uncomfortable with so many lisp routines lurking in the background. When the macros are disabled, no problem, UNDO works perfectly.
I had a few lisp macros with names identical to the ZW default aliases. This apparently confused ZW, causing the erratic behaviuor. I therefore carefully redefined the ZW aliases, and disabled any lisp macros which could cause a conflict. I may in future revert back to the lisp macros (more functional than the rather generic aliases) , but for the time being I shall sacrifice functionality - the program now performs like it should.
ps- Autocad is quite happy with lisp routines and aliases having the same names - it gives preference to lisp and does not get confused. just a word of warning to those who think migration is plug-and-play
Message 4 of 4