Re: [Aegis-developers] more problems building aegis with cook
Brought to you by:
pmiller
From: Ralph A. S. <ral...@ve...> - 2008-01-31 12:56:14
|
Aryeh M. Friedman wrote: > Ralph A. Smith wrote: > > > >> > > I don't see any sign that cook looks for the move_list.cc file. You > > might check the output of > > > aegis -list pf | grep move_list > > aegis -list cf | grep move_list > > aryeh@flosoft:/home/aryeh/aegis.4.23.C010% aegis -list pf | grep move_list > source 1 common/str_list/remove_list.cc > source 1 libaegis/move_list.cc > source 1 libaegis/move_list.h > aryeh@flosoft:/home/aryeh/aegis.4.23.C010% aegis -list cf | grep move_list > source modify 1 common/str_list/remove_list.cc > source remove 1 libaegis/move_list.cc > > > to make sure the metadata is consistent. I've got to quit for the day, > > but perhaps someone else has an idea. > > > Ok. That is a clue - the metadata is not consistent. The "remove" action for move_list.cc is wrong, and I remember something like that happened if certain versions of aedist were used for sending and/or receiving the .ae file. Peter and Walter fixed a number of aedist bugs in the course of 4.23 development, and I am not sure if all the fixes were backported to 4.22. Since you have been working at this for a while, it is probably hard to figure out which delta of aedist was used (and which delta of other binaries were in your path at the time). I suspect the change of where the -L option was defined was mixed in with version switching, since it should not make a difference. It would be best to start your aegis repo over again, doing everything with a bleeding edge setup. Before you do, can you post a summary of the modifications you have needed (just for completeness)? If Peter or Walter puts them into aegis-4.24rc2.ae you may be able to import it without pain. |