Re: [Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Alexander Hansen
On Sep 13, 2008, at 1:13 PM, Jean-François Mertens wrote: On 13 Sep 2008, at 19:09, Jean-François Mertens wrote: on the other hand, (>= %v-1) together with (<= %v-) should work, and be equivalent, no ? I.e., something like Depends: foo (>= %v-1), foo (<= %v-) JF It appears to

Re: [Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Alexander Hansen
On Sep 13, 2008, at 1:13 PM, Jean-François Mertens wrote: On 13 Sep 2008, at 19:09, Jean-François Mertens wrote: on the other hand, (>= %v-1) together with (<= %v-) should work, and be equivalent, no ? I.e., something like Depends: foo (>= %v-1), foo (<= %v-) JF I'd asked abou

Re: [Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Alexander Hansen
On Sep 13, 2008, at 12:34 PM, Benjamin Reed wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Alexander Hansen wrote: I didn't realize it, but fink supports a dependency of the form ( = ) without specifying a revision. If all octave-using packages agree to use Depends: octave ( = ) | octa

Re: [Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Jean-François Mertens
On 13 Sep 2008, at 19:09, Jean-François Mertens wrote: > on the other hand, > (>= %v-1) together with (<= %v-) > should work, and be equivalent, no ? I.e., something like Depends: foo (>= %v-1), foo (<= %v-) JF - T

Re: [Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Jean-François Mertens
On 13 Sep 2008, at 18:34, Benjamin Reed wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Alexander Hansen wrote: >> I didn't realize it, but fink supports a dependency of the form >> ( = ) without specifying a revision. >> >> If all octave-using packages agree to use Depends: octave (

Re: [Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Benjamin Reed
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Alexander Hansen wrote: > I didn't realize it, but fink supports a dependency of the form > ( = ) without specifying a revision. > > If all octave-using packages agree to use Depends: octave ( = version> ) | octave-atlas ( = ) [or an equivalent] >

[Fink-devel] I think we can go back to the old-style octave

2008-09-13 Thread Alexander Hansen
I didn't realize it, but fink supports a dependency of the form ( = ) without specifying a revision. If all octave-using packages agree to use Depends: octave ( = ) | octave-atlas ( = ) [or an equivalent] rather than ( >=-1), then that will satisfy the prior breakage issue in a simpler packa