[Rpm-maint] [RFC/PATCH] comparisons without release

[Anssi Hannula]

James Antill kirjoitti tiistai, 10. elokuuta 2010 19:45:24:
> On Tue, 2010-08-10 at 19:23 +0300, Anssi Hannula wrote:
> > > > One solution that may seem obvious at first is making the comparison
> > > > ignore releases only if neither has one. However, it would break a
> > > > situation where A: Provides: foo = 1.2-5
> > > > B: Requires: foo = 1.2
> > > > 
> > > > So, in this "requires/conflicts/etc has RPMSENSE_EQUAL on
> > > > non-released EVR"
> > >  
> > >  You can't change the semantics just for '=' operations, or you get
> > > 
> > > insanity like:
> > > 
> > > Requires: foo = 1.2 (matches)
> > > Requires: foo >= 1.2 (doesn't match)
> > 
> > That was not the case, '>=' is AFAIK (RPMSENSE_EQUAL | RPMSENSE_GREATER)
> > and thus has RPMSENSE_EQUAL, getting the same behaviour.
> 
>  Ok, I misread your intent, but that just changes the insanity. Given
> the code:
> 
> +    /* "= 1.0" and similar have to match 1.0-1 */
> +    if((B->Flags[B->i] & (RPMSENSE_EQUAL | RPMSENSE_GREATER)) && *bR ==
> '\0') {
> +       aR = NULL;
> +       bR = NULL;
> +    }
> +
> 
> ...you get:
> 
> Provides: foo = 1.2-0.1
> Requires: foo >= 1.2 (match)

As before.

> Requires: foo >  1.2 (doesn't match)

As before.

> Requires: foo <  1.2 (match)

How is 1.2-0.1 < 1.2 true?
AFAICS this is "no match", as before.

> Requires: foo <= 1.2 (match)

As before.

-- 
Anssi Hannula