[racket-dev] current packages' docs, errors, and conflicts

At Tue, 8 Jul 2014 10:15:10 -0400, Sam Tobin-Hochstadt wrote:
>  - I wonder if using Docker instead of VirtualBox could make
> incrementality easier, since that's one of things that they focus on.

I don't think it would be easier, but it might be more efficient and
even easier to set up, so it seems worth trying.

If you're interested in trying that, the information below should help
you get oriented.


>  - I wanted to be able to see which of my packages had problems, so I
> wrote this PR: https://github.com/plt/racket/pull/721 but I'm not sure
> how to test it.

I've merged and tweaked that change, but maybe you want to make more
changes... I don't have a good way to try it other than to run a full
build.


I've enclosed a script that will build based on the latest snapshot at
Northwestern (but running the script with the latest from the Racket
repo). The current directory is used as working space. After a first
run, changing `just-summary?` to #t can help you run and fine-tune the
summary part even faster.


To run the script as-is, you'll need a Virtual Box VM named "Ubuntu
Server 14.04" that is accesible from its host at 192.168.56.107 and
that has a snapshot named "build" that is the starting point.

You could use this image:

 build.ova

Boot that image, save its state, and take a snapshot named "build". You
can make things faster by cloning the VM, adjusting each to have its
own static IP, and adding the new VMs in the `#:vms` argument --- and I
bet Docker is better at that kind of thing.


>  - I think we need to support planet packages -- there are some people
> still releasing new ones, and there are old ones would take
> non-trivial work to port.

Supporting Planet packages is a lot of work. Overcoming constrained
network access in the sandbox is the most obvious problem and probably
easy to solve. A more subtle and important piece of the puzzle is the
notion of "built" packages, which can be quickly installed for
dependent packages or for assembling documentation at the end. Planet
packages don't have a "built" concept, and a package that depends on a
Planet package won't have the right "built" properties: it will
install, but not quickly.

I think we're much better off moving Planet packages to supported
packages in the new package system --- at least, for use by packages in
the new package system.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nwu-docs.rkt
Type: application/octet-stream
Size: 447 bytes
Desc: not available
URL: <http://lists.racket-lang.org/dev/archive/attachments/20140708/6be8aec0/attachment.obj>