On Tue, 2008-09-09 at 20:54 +0200, David Kuehling wrote: > >>>>> "Peter" == Peter Clifton <[EMAIL PROTECTED]> writes: > > > On Mon, 2008-09-08 at 17:53 +0200, David Kuehling wrote: > >> I can't find the reason for that error. Is there any way I can make > >> gsch2pcb give some more output about _why_ it chose to delete an > >> element? Is anything known about that anomaly? > > > From the output, it doesn't appear to be trying to re-add the > > components it deleted either. I wonder if it plain can't find them on > > disk anymore, or if there is some other problem with the design not > > matching the PCB board. > > Yes, PCB seems to think that: after some more experimenting, adding > option --preserve tells me for all elements previously deleted, that > they are not contained in my schematic. > > > The output might suggest that the netlist / parts list produced by > > your gsch2pcb.project isn't actually matching your board. Do you have > > multiple schematics or hierarchy in the project? > > I have multiple schematics, and all are listed in the gsch2pcb.project > file. Not hierarchy. > > The strange thing is: if I remove my board_new.pcb file, and run > gsch2pcb, everything works fine, and *all* components are successfully > added to a newly created PCB design: > > rm board_new.pcb > gsch2pcb -v -v -v "gsch2pcb.project" -o "board_new" > [..] > ---------------------------------- > Done processing. Work performed: > 0 file elements and 107 m4 elements added to board_new.pcb. > > So somehow an already existing pcb file, for which merging is attempted, > has an influence on generation or parsing of the pcb-file generated by > the spawned gnetlist process. > > > Check the resulting netlist, are all components appearing to show up, > > or are some missing? > > Yes, the netlist is fine, even if board_new.pcb previously existed, and > most of my elements got delted. > > > Do you get any errors in a gnetlist.log? > > Well yes, many minor things ("symbol version mismatch", "Found an > improper attribute"...), but nothing consistent with the selection of > elements that get removed. > > Looks like I have to get my hands dirty and run gsch2pcb from within > gdb... > > BTW I'm running gsch2pcb 1.6, included with Ubuntu 7.10's geda-utils > package version 1.0.1.20070626.
Its a little old, but I can't remember any particular bug-fixes to gsch2pcb since then. A few command line options have been added, but nothing jumps out. Try running gnetlist manually: (E.g. for one of my designs..) gnetlist -g gsch2pcb winch-board.sch winch-channel1.sch winch-channel2.sch winch-channel3.sch winch-channel4.sch winch-channel5.sch winch-channel6.sch -o test.output I get a file test.output, which lists all components: > # release: pcb 1.6.3 > PCB("" 6000 5000) > Grid(10 0 0) > Cursor(0 0 3) > Flags(0x000000d0) > Groups("1,2,3,s:4,5,6,c:7:8:") > Styles("Signal,10,40,20:Power,25,60,35:Fat,40,60,35:Skinny,8,36,20") > PKG_0805(0805,C53,0u1) > PKG_ACY400(ACY400,R36,10K) > PKG_FUSEHOLDER-2250P-2200L-900W(FUSEHOLDER-2250P-2200L-900W,F1,4A) > PKG_0805(0805,C52,1uF) > PKG_ACY400(ACY400,R35,10K) > > [etc...] > Do your missing components appear in this list? (I'd guess they probably do, as gsch2pcb with a blank board works). If you're able to and you're stuck with debugging, you could send me the relevant files and I can have a poke around at what is causing the problem. Sometimes PCB would deem to replace a component on the board if the description in PCB doesn't match gsch2pcb's idea of the right footprint, but since it isn't re-adding your elements, it would appear something more sinister is going on. One last common gotcha.. do any of your compoennt footprints have "-" in them? If so, (and you're not using M4 footprints), try running gsch2pcb with --skip-m4 In any case, I'd really like to get to the bottom of this problem. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user