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

Re: Midori dependencies problem with sid



In <[🔎] 20090529013505.GA12317@emurlahn.burrows.local>, Daniel Burrows wrote:
>On Thu, May 28, 2009 at 11:21:38AM -0500, "Boyd Stephen Smith Jr." 
<bss@iguanasuicide.net> was heard to say:
>> In <[🔎] 20090528153521.GA31559@emurlahn.burrows.local>, Daniel Burrows wrote:
>> >On Thu, May 28, 2009 at 09:02:59AM -0500, "Boyd Stephen Smith Jr."
>> <bss@iguanasuicide.net> was heard to say:
>> >> Isn't that supposed to change in the future (but perhaps not before
>> >> Squeeze) so that we can have versioned dependencies on virtual
>> >> packages?
>> >  Yes.  It's been supposed to change in the future for at least ten
>> >years. :-)
>> Maybe this is a better discussion for debian-devel or the bug itself,
>> but what the the technological issues in implementing this?
>  Adding versioned Provides would affect all the software that tries
>to process Debian packages and reason about their dependency
>relationships.  That's a lot of software, and all of it would be broken
>until it learned about the new feature.

Not completely broken.  I imagine versioned Provides would be rather rare.

Also, shouldn't most of that software be using libapt?  If someone wanted to 
implement it, how could they get a list of software to fix?

>  Also, I don't know offhand of any situations where someone has really
>needed versioned Provides -- the main one that comes to mind is when
>you want to rename a package that has versioned reverse dependendencies,
>and people usually just leave a dummy package behind in that case.

Dummy packages solve just about everything. :)

But, that does raise the point that I really couldn't think of a good reason 
to use versioned provides.  The best thing I've come up with is being able 
to state that your package is (e.g.) a drop-in replacement for a package 
that (a) already has versioned dependencies on it and (b) will be available 
(and possibly installable) in parallel with your package.

(Theoretical example)
unishell-1.0.2 is a drop-in replacement for bash-3.2 and ksh-93s.
bash-fw Depends on bash (>= 3), iptables
It would be nice to have unishell-1.0.2 Provides bash-3.2.
Right now, you'd have to modify bash-3.2 to Provide bash3-ui and have 
unishell-1.0.2 Provides bash3-ui and modify ipmasq to Depend on bash (>= 3) 
| bash3-ui.  Heaven forbid you had another package that depended on bash (>= 
3.1~bpo) or somesuch.

Might be useful for when projects fork, too.
-- 
Boyd Stephen Smith Jr.           	 ,= ,-_-. =.
bss@iguanasuicide.net            	((_/)o o(\_))
ICQ: 514984 YM/AIM: DaTwinkDaddy 	 `-'(. .)`-'
http://iguanasuicide.net/        	     \_/

Attachment: signature.asc
Description: This is a digitally signed message part.


Reply to: