As requested by some users, we finally have support for cloud sync,
spelling, geolocation and a lot more of the services that require API
keys from Google. Details about which services are involved can be found
at: http://www.chromium.org/developers/how-tos/api-keys
Thanks to Paweł Hajdan <phajdan@google.com> for giving us permission to
distribute the API keys with our build of Chromium:
> Note that the public Terms of Service do not allow distribution of the
> API keys in any form. To make this work for you, on behalf of Google
> Chrome Team I am providing you with:
> Official permission to include Google API keys in your packages and to
> distribute these packages. The remainder of the Terms of Service for
> each API applies, but at this time you are not bound by the
> requirement to only access the APIs for personal and development use,
> and Additional quota for each API in an effort to adequately support
> your users.
As noted in the source: Those keys are for use in NixOS/nixpkgs ONLY!
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This also fixes the annoying issue that minicom doesn't work out of the
box:
$ minicom
minicom: there is no global configuration file /etc/minirc.dfl
Ask your sysadmin to create one (with minicom -s).
$ sudo minicom -s
minicom: there is no global configuration file /etc/minirc.dfl
Ask your sysadmin to create one (with minicom -s).
minicom 2.4 basically refuses to enter setup unless /etc/minirc.dfl
already exists. sudo touch /etc/minirc.dfl is enough to fix that though,
but with this commit "sudo minicom -s" will work out of the box.
Since "src" is a fetchsvn directory, the source is copied with "cp
--no-preserve=timestamps" (see commit
6d928ab684). So some source files might
get a slightly different timestamp. Here, if lib/standard.ppmdfont
gets a newer timestamp than the generated file lib/standardppmdfont.c,
Make will try to rebuild the latter. But that fails because the
ppmdcfont program doesn't exist (yet).
Probably stdenv should ensure that every file has the same timestamp.
For Linux, the phases was run in the wrong order. I've
fixed that, but the package is still a complete mess that
needs to be cleaned up. The linux and darwin builds should
probably be separated into two different Nix expressions
to avoid the current conditional mess.
