[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: pkg.m planned refactoring [was: Help trying to merge my changes to c
From: |
Carlo De Falco |
Subject: |
Re: pkg.m planned refactoring [was: Help trying to merge my changes to current default] |
Date: |
Fri, 23 Mar 2018 09:32:51 +0000 |
> On 22 Mar 2018, at 13:55, Juan Pablo Carbajal <address@hidden> wrote:
>
> Hi,
>
> The plan is to use dispatching (just as was done in 2012) and let the
> subfunctions do the parsing. Subfunctinos could use inputParser, no
> problem at all.
>
> The idea is that pkg.m becomes just a bookeeping function (very
> simple!) and the work is passsed to the subfunctions.
>
> I will define a standard procedure to parse options such that the
> structure of the subfunctions is more or less always the same (ease
> extending behavior, this was also done in 2012).
>
> The only thing that is expect that wont be backwards compatible is the
> future help system, which I will copy hg system, that is
>
> pkg help
>
> is the same as
>
> help pkg
>
> , but
>
> pkg help install
>
> (which currently doesn't exist) will provide detailed info about the command.
> This is to also moves the writing of help to the subfunctions
> (although I do not think it can be texinfo unless we make them
> private). Being modular will simplify the maintenance.
>
> I would try to keep backward compatibility as much as possible.
>
> Cheers
>
JPi,
I totally support this plan, actually I already agreed with this plan in 2012
it's such a pity it wasn't implemented earlier ...
I just noted that in 2012 InputParser was not available, now that it is I think
it would worth considering whether it may help simplify things further.
c.