Re: Output quoting (was: Re: Possible solution for special characters in makefile paths)
Eli Zaretskii <[email protected]> Fri, 11 Apr 2014 16:13:39 +0300
| Newsgroups | gmane.comp.gnu.make.devel |
|---|---|
| Message-ID | <[email protected]> |
> Date: Fri, 11 Apr 2014 14:02:59 +0100 > From: Tim Murphy <[email protected]> > Cc: Frank Heckenbach <[email protected]>, [email protected] > > > Then SHELL_QUOTE, whatever its name, is not what you want. What you > > want is a way to write an extension (in C or in Guile) that would hook > > into this expansion, and let you do whatever you need. IOW, you want > > to ask for infrastructure that would allow extensions to do things > > like that (I don't think we have such an infrastructure right, but > > maybe I'm mistaken). > > > > Let's not (ab)use quoting for things it is not supposed to provide and > > support. > > You can't control what people do with a feature that you add to make. No, but I can tell them "if it hurts, don't do that". > When you show them a new thing that is possible they will use it. Why > make it a crap one-purpose feature when it can so obviously be made > generally useful and with so little work? There's nothing wrong with one-purpose features. Each feature should do its intended job well, no more, no less. You are asking for a much more broad feature. I agree that it will be very useful, I just don't see why we need to lump it together with quoting. It's not a natural extension of quoting in any sense. > "Feature creep" is just a pejorative term that one can throw at any > unexpected suggestion. Surely one can distinguish between a useful > generalisation of a very specific concept and an unnecessary feature. > Surely it's a good thing to generalise a capability so that it can be > used widely? I don't see how your suggestion is a generalization of file-name quoting.