FS#64028 - [base] Remove compressors
Attached to Project:
Arch Linux
Opened by Sébastien Luttringer (seblu) - Sunday, 06 October 2019, 10:14 GMT
Last edited by Buggy McBugFace (bugbot) - Saturday, 25 November 2023, 20:13 GMT
Opened by Sébastien Luttringer (seblu) - Sunday, 06 October 2019, 10:14 GMT
Last edited by Buggy McBugFace (bugbot) - Saturday, 25 November 2023, 20:13 GMT
|
Details
I suggest to remove compressors, they have to be pulled by
deps by packages which require them.
|
This task depends upon
Closed by Buggy McBugFace (bugbot)
Saturday, 25 November 2023, 20:13 GMT
Reason for closing: Moved
Additional comments about closing: https://gitlab.archlinux.org/archlinux/p ackaging/packages/base/issues/1
Saturday, 25 November 2023, 20:13 GMT
Reason for closing: Moved
Additional comments about closing: https://gitlab.archlinux.org/archlinux/p ackaging/packages/base/issues/1
Anyway, it's not like a valid archlinux system can do without them as they are required by libarchive and therefore pacman.
To be on that `base` list, they should be required by themselves.
xz is quite standard in a linux-specific way, but I'd argue even bzip2 is sufficiently useful that you don't want to be rendered unable to even read some fairly common file distribution formats out of a desire to save 150 kilobytes. Hence "required for itself".
The fact that pacman requires them simply means that any desire to remove them from base must be based on ideological purity rather than an actual desire to remove it from minimal systems in practice.
Would be easier/simpler to rely on that standard, but the cost would be to pull command like lp (from cups), mailx (from s-nail) which are not really necessary nowadays and with a lot of dependencies.
[1] http://refspecs.linuxfoundation.org/LSB_5.0.0/LSB-Common/LSB-Common/rcommands.html#TBL-CORECMD
xz and gzip is what people in practice will overwhelmingly use when they want to compress files and move them around, depending on whether they care more about portability and CPU time, or bandwidth.
bzip is not overwhelmingly used, though -- it is merely technology that has been around since 1996 and been stable since 2000, available on most/all Linux systems, and therefore used by lots of people if not most people. (And also before xz came on the scene, it was *the* popular go-to for "better compression than gzip at the cost of a bit more processing time".) So there will still tend to be an awful lot of places where bzip2 is used, and it's valuable to be interoperable with that. But I could potentially hear the argument that it's not quite as important these days. (I still think 150KB is a very cheap price to pay for ensuring data portability.)