Re: "features" problems

[Carnë Draug]

On 21 March 2016 at 21:52, Philip Nienhuis <address@hidden> wrote:
>>>> The reason for why PKG_ADD behaves like this is that it is meant for
>>>> simpler
>>>> things, such as registering options.  If you need a complex piece of
>>>> code,
>>>> use a separate function for it and call it from PKG_ADD.
>>>
>>>
>>> That is a good suggestion.
>>> But it doesn't change the PKG_ADD location issue which was at the base of
>>> this part of the discussion - to wit, why certain octave_config_info
>>> output
>>> could be required.
>>>
>>
>> It does.  Instead of writing another long email trying explaining this, I
>> just did the coding myself.  See bug #47480 and the attached patch.
>
>
> Must be my ignorance, but in that bug report + linked reports I see nothing
> that negates what I wrote: that pkg.m and/or subsidiaries screwed up as
> regards PKG_* location. A fix for that was indeed needed.
> You made that fix, thanks for it and good that it is pushed. Too bad that it
> took such a long time for a solution for those PKG_* issues with pkg.m
> (absolutely no blame on you).
>
> PKG_ADD's location and what io's PKG_ADD is doing are separate issues. You
> fixed the former; my job now is to fix the latter.

A PKG_ADD file at the root of a package will be installed in:

  1) in the arch dependent directory if exists one;
  2) in the arch independent directory otherwise.

The io package has oct files therefore its PKG_ADD file goes into the arch
dependent directory.  If a package needs a PKG_ADD in the arch independent
directory, then it must use PKG_ADD directives in m files.

There is no screw up.  There is nothing that needed fixing in pkg to have
the io package install fine without moving PKG_* files itself during
post_install.

I have been fixing things in pkg but they are unrelated to your issues.

Carnë