On Tue, 2009-12-22 at 08:14 +1100, Hamish Moffatt wrote: > Thanks! I see now that it's trying to write logs into ~/.gEDA which won't > work with an unwritable/non-existent $HOME.
Perhaps we will be able to fix this upstream (for 1.6.1?). Hamish.. is there any way we can work around the issue for the 1.6.0 build? Can we export a temporary $HOME somewhere in the build tree whilst the tests are run? In the long term, we should make gEDA apps fail more gracefully if they can't open a log file. Dumping output to stderr "might" work.. although it would change what is output in the case of $HOME being unwriteable - which could in its-self break a more rigorous test-suite that checked expected stderr output. Fortunately for gsymcheck, the output we are testing against is stdout, so that should be ok. I can't think of any tests which rely on known stderr output, so perhaps we are OK here (for now). Personally, I see this as one more argument in favour of logging in $PWD, like we used to. (or was it $(dirname $schematic)... I can't recall). It would still be nice if that worked on a read-only dir. Of course.. there is other stuff gEDA looks to write / read on ~/.gEDA, so we'll have to test those cases as well. Best wishes, Peter C. -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org