Symlinks support status?

[Dov Feldstern]

Matt Mackall wrote:
> On Mon, Nov 05, 2007 at 10:32:22PM +0200, Dov Feldstern wrote:
>> Bryan O'Sullivan wrote:
>>> Dov Feldstern wrote:
>>>
>>>> The idea is basically to have the working directory appear --- to all 
>>>> the usual tools (compiler, editor, etc.) --- as it would on a 
>>>> symlink-supporting system, i.e., the contents of the symlinks look like 
>>>> the contents of the linked-to files.
>>> Unfortunately, that's both impossible and completely wrong.
>>>
>>> 	<b
>> Would you care to elaborate? I don't entirely agree on either count... :)
> 
> I suppose I'll chime in now..
> 
> While we could probably do something like this, it seems like WAY too
> much trouble. So I'd downgrade impossible to impractical.
> 
> And it really strikes me as an complex ugly hack, which makes it
> fairly high up in the "wrong" scale as well.
> 

Fair enough. I'm not sure that it would be have to be that complex or 
ugly, but then again I haven't looked at mercurial's internals at all, 
yet. I suppose I could always try to implement this as an extension? 
(though I don't think I'll get to this in the foreseeable future :( . 
I'd be happy to see someone else who cares about this pick it up...)

> Symlinks are nice, but by no means essential (as evidenced by the
> numerous filesystems and operating systems that get by just fine
> without them). Mercurial supports them (grudgingly) but there's no
> reason a build environment can't get by without them either. Going to
> Herculean lengths to emulate them (poorly) on systems where they're
> not supported is not likely to happen.
> 

Err, I beg to differ about the "get by just fine without them" part. I 
hate every minute that I have to, unfortunately, work on Windows --- 
although arguably, symlinks is only a minor --- though not negligible 
--- factor in that... ;)

Seriously, though, I don't think that you need to take such a grudging 
view of the symlinks support you've added. It's a nice feature to have, 
and I, at least, thank you for it!

> To the person who proposed using Windows hardlinks: the semantic
> differences are still too large. The same applies for shortcuts.
> 

Thanks for the responses!
Dov