[pacman-dev] bsd port?

newer
[pacman-dev] _alpm_depmiss_isin is...

Xavier

23 Oct 2007 23 Oct '07

3:28 p.m.

On Tue, Oct 23, 2007 at 04:03:34PM -0500, Aaron Griffin wrote:

The above applies here as well, BUT, I wanted to know where this was being discussed. If pacman was ported to BSD, I've never seen, nor heard of, patches for it. This would be valuable information for us. Not only would it save us work, but Dan is doing this _right now_.

I just read the above on arch-dev-public. A while ago, someone came on IRC while no one was around to report the status of getting pacman running on freebsd : "got pacman to compile on libfetch and libdownload that are in FreeBSD 6.2 1. cut out with that damn bashism and linux non-posix crap 2. check the makefiles, they're full of bugs but all this du -cb and cp -a crap... and then the "source" syntax instead of "." and there's also chown root.root, which should be root:root, or even better 0:0" I just noted it because I thought it could be interesting, but I don't know anything about this stuff. And I thought no one else cared. But if Dan is interested in doing this, I just thought he should know about the last 3 lines of this report. If that doesn't help at all, sorry, just ignore this mail :)

Show replies by date

Mateusz Jedrasik

23 Oct 23 Oct

3:35 p.m.

Tuesday 23 of October 2007 23:28:42 Xavier napisał(a):

...

On Tue, Oct 23, 2007 at 04:03:34PM -0500, Aaron Griffin wrote:

...
The above applies here as well, BUT, I wanted to know where this was being discussed. If pacman was ported to BSD, I've never seen, nor heard of, patches for it. This would be valuable information for us. Not only would it save us work, but Dan is doing this _right now_.

I just read the above on arch-dev-public.

A while ago, someone came on IRC while no one was around to report the status of getting pacman running on freebsd :

"got pacman to compile on libfetch and libdownload that are in FreeBSD 6.2 1. cut out with that damn bashism and linux non-posix crap 2. check the makefiles, they're full of bugs but all this du -cb and cp -a crap... and then the "source" syntax instead of "." and there's also chown root.root, which should be root:root, or even better 0:0"

I just noted it because I thought it could be interesting, but I don't know anything about this stuff. And I thought no one else cared. But if Dan is interested in doing this, I just thought he should know about the last 3 lines of this report. If that doesn't help at all, sorry, just ignore this mail :)

_______________________________________________ pacman-dev mailing list pacman-dev@archlinux.org http://archlinux.org/mailman/listinfo/pacman-dev

I've tried it some time ago, didn't get too far as I couldn't sort of be arsed too much ;-) But now that I see someone else's tried removing the linuxisms off of pacman's files, I might give it a go, time permits. Cheers! -- Mateusz Jędrasik tel. +48(79)022-9393, +48(51)69-444-90 http://imachine.szklo.eu.org

Aaron Griffin

3:40 p.m.

On 10/23/07, Mateusz Jedrasik wrote:

...

I've tried it some time ago, didn't get too far as I couldn't sort of be arsed too much ;-) But now that I see someone else's tried removing the linuxisms off of pacman's files, I might give it a go, time permits.

See, here's the thing. I know this is true of myself, and I don't speak for Dan. But most of "us" linux users don't really know what IS and IS NOT a "linuxism". So comments like Xavier showed us... "cut the bashism crap" and the like couldn't be more useless. The best thing that can be done is actually pointing out SPECIFIC "linuxisms" so we: a) Know what is and is not a linuxism b) Don't repeat it again c) Can fix the code.

Dan McGee

3:47 p.m.

On 10/23/07, Aaron Griffin  wrote:
> On 10/23/07, Mateusz Jedrasik  wrote:
> > I've tried it some time ago, didn't get too far as I couldn't sort of be arsed
> > too much ;-) But now that I see someone else's tried removing the linuxisms
> > off of pacman's files, I might give it a go, time permits.
>
> See, here's the thing. I know this is true of myself, and I don't
> speak for Dan. But most of "us" linux users don't really know what IS
> and IS NOT a "linuxism".
>
> So comments like Xavier showed us... "cut the bashism crap" and the
> like couldn't be more useless.
>
> The best thing that can be done is actually pointing out SPECIFIC
> "linuxisms" so we:
> a) Know what is and is not a linuxism
> b) Don't repeat it again
> c) Can fix the code.

>From Xavier too so we have context:
"got pacman to compile on libfetch and libdownload that are in FreeBSD 6.2
1. cut out with that damn bashism and linux non-posix crap
2. check the makefiles, they're full of bugs
but all this du -cb and cp -a crap...
and then the "source" syntax instead of "."
and there's also chown root.root, which should be root:root, or even better 0:0"

I am 100% with Aaron here. Why the hell should we cater to BSD when we
have not had a single use case out of their corner? I'm doing it more
out of curiosity, and because I believe it helps weed out bugs we
don't know of being only mono-platform.

I don't want to get mad, but if this asshat has a problem wit h the
code then submit a fucking patch. And how hard is a fucking "alias
source='.'"?

And now in a nicer tone of voice, I don't plan on killing anything
bash- its the shell of choice by anything pacman and makepkg do, so
that is staying. But some of the other concerns could be fixed.

-Dan

Mateusz Jedrasik

4:54 p.m.

Tuesday 23 of October 2007 23:47:28 Dan McGee napisał(a):

...

On 10/23/07, Aaron Griffin wrote:

...
On 10/23/07, Mateusz Jedrasik wrote:

...
I've tried it some time ago, didn't get too far as I couldn't sort of be arsed too much ;-) But now that I see someone else's tried removing the linuxisms off of pacman's files, I might give it a go, time permits.

See, here's the thing. I know this is true of myself, and I don't speak for Dan. But most of "us" linux users don't really know what IS and IS NOT a "linuxism".

So comments like Xavier showed us... "cut the bashism crap" and the like couldn't be more useless.

The best thing that can be done is actually pointing out SPECIFIC "linuxisms" so we: a) Know what is and is not a linuxism b) Don't repeat it again c) Can fix the code.

From Xavier too so we have context:

"got pacman to compile on libfetch and libdownload that are in FreeBSD 6.2 1. cut out with that damn bashism and linux non-posix crap 2. check the makefiles, they're full of bugs but all this du -cb and cp -a crap... and then the "source" syntax instead of "." and there's also chown root.root, which should be root:root, or even better 0:0"

I am 100% with Aaron here. Why the hell should we cater to BSD when we have not had a single use case out of their corner?

No one ever said you should. And you by all means don't have to - one particular person and some others have tried, and got somewhere, so why bother heh? I SUPPOSE a cooperation wouldn't hurt ofcourse... but then again that's entirely up to both pacman devs and particular possible future bsd pacman users... You could perhaphs be aware that just like there isn't much knowledge of bsd perhaphs among the arch users/devs, similarly there isn't much or maybe won't be much knowledge of arch among the bsd users - that, however, doesn't mean that the cooperation of both could not be fruitful. Once something works on FreeBSD it's probably reasonably easy to port to say NetBSD... or OpenBSD... then think of all the millions of architectures that NetBSD supports.. with pacman.. etc... etc etc etc ;-) Tho they probably would like a non-gnu package manager- but that's another story. Just pointing out a possibility that could be taken advantage of.

...

I'm doing it more out of curiosity, and because I believe it helps weed out bugs we don't know of being only mono-platform.

Yeah that's another thing - coding for the pure pleasure of getting better at it. Just because it works under linux, or because some compilers/etc are friendlier on the bugs than other, does not mean the code is proper or so. Hence the suggestions about 'linuxisms' by the user - I wouldn't take his comments too serious, at least those which don't bring in much particular info, and hence, I suppose, weren't meant to be serious first off - otherwise they'd probably bring some info in.

...

I don't want to get mad, but if this asshat has a problem wit h the code then submit a fucking patch. And how hard is a fucking "alias source='.'"?

Is this asshat a traceable asshat? Do we know how to reach the particular asshat? Perhaphs the asshat has gotten somewhere with his work... Perhaphs we(you) and the asshat are doing the same things twice, perhaphs someone's even doing it better. Perhaphs contacting the asshat might be fruitful; Perhaphs dropping the asshat suggestion at that might not hurt too much either. As regards the source="." afaik, if you build it say from ports, or not in a chroot or so, wouldn't that hurt the build process? I'm not a dev/programmer , but that entire alias source sounds and feels icky even to me.. It's ripped out of context here tho as I haven't looked through the code much personally, so I'll restrain myself from commenting.

...

And now in a nicer tone of voice, I don't plan on killing anything bash- its the shell of choice by anything pacman and makepkg do, so that is staying. But some of the other concerns could be fixed.

Bash is fine for some, but sh is the more preferred imho nonetheless shell for build processes/compilations. Bash is an extra on all the BSD's (afaik). having an extra dep just for building something kind of sucks, especially if it's an entire shell. I believe sh is just as good (and should be used). Using bash when the program runs, well, that's a different story and that too should not be encouraged - after all pacman is not a bash script. And users in arch use other shells than bash too.

...

-Dan

_______________________________________________ pacman-dev mailing list pacman-dev@archlinux.org http://archlinux.org/mailman/listinfo/pacman-dev

-- Mateusz Jędrasik tel. +48(79)022-9393, +48(51)69-444-90 http://imachine.szklo.eu.org

Xavier

4:14 p.m.

On Tue, Oct 23, 2007 at 04:40:56PM -0500, Aaron Griffin wrote:

...

On 10/23/07, Mateusz Jedrasik wrote:

...
I've tried it some time ago, didn't get too far as I couldn't sort of be arsed too much ;-) But now that I see someone else's tried removing the linuxisms off of pacman's files, I might give it a go, time permits.

See, here's the thing. I know this is true of myself, and I don't speak for Dan. But most of "us" linux users don't really know what IS and IS NOT a "linuxism".

So comments like Xavier showed us... "cut the bashism crap" and the like couldn't be more useless.

The best thing that can be done is actually pointing out SPECIFIC "linuxisms" so we: a) Know what is and is not a linuxism b) Don't repeat it again c) Can fix the code.

I said the LAST three lines were interesting in my opinion, not the first three :) I agree the first three are useless, but I thought the last three did exactly what you just said, pointing out specific linuxisms. I apologize for not having reformulated that report in a nicer way, I didn't know it would be taken so seriously, so I apologize for that.

Aaron Griffin

3:37 p.m.

On 10/23/07, Xavier wrote:

...

But if Dan is interested in doing this, I just thought he should know about the last 3 lines of this report. If that doesn't help at all, sorry, just ignore this mail :)

For the record: http://code.toofishes.net/gitweb.cgi?p=pacman.git;a=shortlog;h=freebsd He was working with Eliott to get this at least compiling. It saddens me to know that someone may have already done this work, yet we still need to duplicate it.

Miklos Vajna

24 Oct 24 Oct

3:59 p.m.

On Tue, Oct 23, 2007 at 04:37:37PM -0500, Aaron Griffin wrote:

...

For the record: http://code.toofishes.net/gitweb.cgi?p=pacman.git;a=shortlog;h=freebsd

He was working with Eliott to get this at least compiling. It saddens me to know that someone may have already done this work, yet we still need to duplicate it.

for the record once ([1]) pacman-g2 was ported to both maxosx and openbsd, and we merged those patches but then the guy who did it disappered so i have no idea if the current pacman-g2 compiles/runs on those platforms or not. oh and once i needed libalpm on solaris.. the story is the same. ([2]) just in case somebody is interested in these tricks, maybe it can save some time compared to starting from scratch :) 1. http://git.frugalware.org/gitweb/gitweb.cgi?p=pacman-g2.git;a=blob;f=NEWS#l2... 2. http://git.frugalware.org/gitweb/gitweb.cgi?p=pacman-g2.git;a=blob;f=NEWS#l1... - VMiklos

6034

Age (days ago)

6035

Last active (days ago)

List overview

Download

7 comments

5 participants

participants (5)

Aaron Griffin
Dan McGee
Mateusz Jedrasik
Miklos Vajna
Xavier