Re: [PATCH]: gnu: Add lash.

[Mark H Weaver]

Ricardo Wurmus <address@hidden> writes:

> David Thompson writes:
>
>>> +             (uri (string-append
>>> +                   "mirror://savannah/lash/lash-"
>>> +                   (string-map (lambda (x) (if (char=? x #\-) #\~ x)) 
>>> version)
>>
>> How about splitting the string on #\- and then joining on #\~ instead?
>
> Yeah, that would probably look better.  (I just adapted this from
> another package without thinking much of it.)

Either way is fine with me.

Also, could you please add a comment explaining why we are doing this
instead of just putting '~' in the 'version' field?

>>> +                   ".tar.bz2"))
>>> +             (file-name (string-append name "-" version ".tar.gz"))
>>
>> The extension should be ".tar.bz2".
>
> Ouch!  Good catch!

Indeed, thanks David!

>>> +             (sha256
>>> +              (base32
>>> +               "12z1vx3krrzsfccpah9xjs68900xvr7bw92wx8np5871i2yv47iw"))))
>>> +    (build-system gnu-build-system)
>>> +    (inputs
>>> +     `(("bdb" ,bdb)
>>> +       ("gtk" ,gtk+-2)
>>> +       ("jack" ,jack-1)
>>> +       ("libuuid" ,util-linux)
>>> +       ("readline" ,readline)
>>> +       ("python" ,python-2)))
>>> +    (propagated-inputs
>>> +     `(("alsa-lib" ,alsa-lib)
>>> +       ("dbus" ,dbus)
>>> +       ("libxml2" ,libxml2)))
>>
>> I'm always wary of propagated inputs, but I'll take your word for it. :)
>
> I'm not sure I'm doing this right, but I noticed that in a package using
> lash pkg-config complained about not being able to confirm a working
> lash because of these three missing dependencies.  These are lash
> dependencies that are listed in PC_REQUIRES.  As far as I can tell any
> package dependent on lash would have to have at least these three
> packages as inputs.

Yes, I think this justifies adding them to 'propagated-inputs'.  Could
you add a comment?

> The complete list of PC_REQUIRES is this:
>
>   dbus, libuuid, libxml2, alsa, jack
>
> I would be happy if someone could confirm that this is how propagated
> inputs are to be used, and if lash should propagate all five of these
> inputs (instead of only the three here).

I'm not sure.  For now, I'd suggest just adding the three here, and we
can revisit later if we run into problems or if someone does a more
careful study.

Okay to push with these changes.

    Thanks!
      Mark