[12:05] <Laney> cjwatson: I got around to writing that script so it's easy to add new transitions now. We can add pages for the other transitions if you want.
[12:07] <cjwatson> Laney: oh, yes, please do.  I'm using http://paste.ubuntu.com/609451/ and http://paste.ubuntu.com/609452/
[13:26] <Laney> up now
[13:53] <doko> Laney: could you setup a page for libffi5/libffi6 too?
[13:53] <Laney> doko: yes, can you provide the is_affected/good/bad lines?
[13:53] <doko> and then there's libobjc2/libobjc3, but that can wait
[13:53] <doko> Laney: is there any docs for these?
[13:54] <Laney> not really i'm afraid
[13:54] <Laney> should be easy to figure out from the existing transitions
[13:55] <Laney> .field ~ /regexp/ | ... basically
[13:57] <doko> Laney: what does the .package mean?
[13:57] <Laney> doko: it means that there's a binary package matching regexp
[13:57] <Laney> usually you'll just want .build-depends and .depends
[13:57] <Laney> openssl is probably a good one to look at
[14:01] <cjwatson> Laney: yay, thanks
[14:01] <Laney> :-)
[14:02] <Laney> we'll probably want a separate branch for the .ben files i suppose
[14:06] <doko> Laney: http://paste.ubuntu.com/609501/
[14:07] <Laney> ty
[14:08] <Laney> added, should show up at the next cron run
[16:14] <Laney> doko: http://orangesquash.org.uk/~laney/transitions/libffi.html does that look right?
[16:15] <doko> $ apt-cache rdepends libffi5|wc -l
[16:15] <doko> 205
[16:15] <doko> strange ...
[16:18] <Laney> look at grep-dctrl -FBuild-Depends -sPackage libffi-dev ...
[16:19] <doko> interesting ... :-/
[16:20] <Laney> haskell stuff seems to get a binary dep without BDing
[16:21] <doko> wondering if we can do with an empty b-d field
[16:28] <cjwatson> you could do  is_affected = .depends ~ /libffi/;  maybe
[16:28] <cjwatson> I assume is_affected is basically just a performance optimisation
[16:28] <cjwatson> or .build-depends ~ /libffi/ | .depends ~ /libffi/
[16:30] <doko> interesting syntax
[16:32] <Laney> it's so you don't see a list of every source package
[16:33] <Laney> not sure if is_affected looks at binaries though
[16:33] <Laney> might as well try
[16:36] <cjwatson> you know your implementation language is too obscure when it's easier for *somebody who works on a functional language* to just try it and see what happens
[16:38] <Laney> the problem is more that I'm trying to do real work concurrently ;)
[16:38] <ScottK> What is this 'real work' you speak of?
[16:38] <ScottK> Oh.  Right.  Better get some of that done myself.
[17:08] <Laney> that seemed to help :-)
[17:08] <Laney> although you can ignore the haskell ones