Re: [PATCH] *-map, *-omap: Allow passing NULL to search

[Bruno Haible]

Hi Colin,

> >   2) Allowing NULL pointers as arguments in all possible places is
> >      *not* a good coding style in general. (It may be a good practice,
> >      though, when you're being paid for a consulting project and your
> >      code will never be maintained once you have delivered it.)
> 
> I respect your opinion as the maintainer and if you don't want to
> include this that's of course fine, but please could you refrain from
> implying incompetence?  It's not nice.

I'm sorry that you understood it in that way. It was really not meant that
way. I remember your name as a long-time serious contributor to various
packages.

It must be possible to talk about coding style and good/bad practices.
Of course, what I said about these practices are _opinions_, and
  - Anyone can disagree with my opinions.
  - Opinions can change over time: You are showing that when I wrote
    the gl_list_iterator_next code, I apparently had a different
    opinion.
  - I do not even know whether you have the habit of allowing NULL
    arguments in general. Therefore I was not criticizing you; I was
    criticizing a certain aspect of the proposed patch. (Which is also
    what I constantly do before committing or submitting a patch myself:
    scrutinize it under aspects that, from experience, are known as
    good or bad practices.)

The sentence about the consulting project was not about you as a person
either; it was meant as a reflection about a condition when I would find
it justified to employ that "bad" coding style myself.

Can we be friends again?

Bruno