[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Proposal: Fixing non-free content issues via installer packages



On 27/07/06, Goneri Le Bouder <goneri@rulezlan.org> wrote:
>> ...and install useless dependency.
> Do you mean fakeroot and dpkg-buildpackage etc. in Depends:
> for users?
Yes, for standard users this is useless.

I don't think is too hard to make meta packages without using those,
*but* I think it would be hard to catch every file in a tarball *and*
place them in the proper places if splitting is needed.

>> A prefere a script that find a valid mirror with netselect
>> and then download the tarball, apply rules on it and
>> extract its content in /usr/share/games/pkgname/
> That's definitely a good idea where possible but not an
> option for commercial data e.g. the doom IWAD file, Quake
> pak files. In this case pulling it from local
> CD-ROM/floppies/filesystem is necessary.

Maybe we should implement some abstract retriever which would take the
files from the local file system, a CD or a site ... you name it. The
user would choose where from to get the file (via a debconf question,
maybe) and then the metapackager would create the assiciated
metapackage which, when uninstalled would remove the files added.

Yep why not. With some md5/sha1 checks to prevent invalid installation.

This is risky, we would have to whitelist *every* data file set which
ever existed, including mods for new things, new transaltions,
expasions, whatever. If the test results only in a warning that the
md5sum did not check, that would be fine. Also, note that for a data
set like quake's we would have to get out hands on such a data set in
order to validate it. Would you trust a user to tell you the correct
data set for a mod which you don't have access to? You might be ending
up in backing up a corrupted, or maliciously crafted data set just by
saying "we expect the data set to have checksum X".

--
Regards,
EddyP
=============================================
"Imagination is more important than knowledge" A.Einstein



Reply to: