The group is specified using a singleton list, so the loaOf merging is
done by iterating through the list items with imap, so it enumerates
every element and sets that as the default "name" attribute.
From lib/types:143:
name = elem.name or "unnamed-${toString defIdx}.${toString elemIdx}";
So, people get groups like "unnamed-X.Y" instead of "mpd".
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Reported-by: devhell <"^"@regexmail.net>
Tested-by: devhell <"^"@regexmail.net>
It now strictly evaluates all remaining attributes, preventing
unevaluated thunks that cannot be garbage-collected. It's also applied
to all jobs in Nixpkgs' release.nix.
This reduces hydra-eval-jobs' memory consumption on the 14.12
release-combined jobset from 5.1 GB to 2.0 GB.
I'm not adding this to pkgs/development/libraries because it somewhat is
strongly tied to Haxe itself, because otherwise you can't compile to C++
and in the event that someone is going to create something like
"haxePackages" someday it is easier to notice when it's residing in the
Haxe folder.
In theory it would also work by using imperative haxelib, but you'll get
precompiled libraries which need to be patched on NixOS systems. That's
the main reason I was packaging this, among from the fact that even when
patching the libraries, it still leads to occasional library hell and
instabilities.
The package has two outputs: One with the library itself, needed for
compile time ($out) and another one ($lib) which is needed at runtime,
so after compiling, the $out path can be safely garbage collected.
Right now, I've set meta.platforms to Linux only, because that's where
I've tested it. In order to get it running on other platforms the
targetArch attribute has to be set accordingly.
We also build everything completely from scratch, even though there are
binaries within the source ZIP file. The main reason is to make smaller
library dependencies by avoiding bundled libraries and using the ones we
already ship with nixpkgs.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
