[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnu-arch-users] version aliases

From: Aaron Bentley
Subject: Re: [Gnu-arch-users] version aliases
Date: Mon, 05 Apr 2004 18:24:32 -0400
User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.6b) Gecko/20031205 Thunderbird/0.4

Miles Bader wrote:

On Mon, Apr 05, 2004 at 09:04:40AM -0700, Tom Lord wrote:

Not really.   Aliases would need a distinctive syntax -- perhaps a
prefix character.

        tla get :latest-emacs

How come?  Existing arch names already have a distinctive syntax, so an alias
could be anything else (with the rule that they not intrude on the real
namespace).  It might possibly impact error handling -- a typo could
syntactically look like an alias -- but even this seems very minor, as only
valid aliases would be accepted, and it's very unlikely that anybody would
keep lots of aliases that are a small edit-distance from real names.

The problem is, what if I make devo="address@hidden/tla--devo--1.3" and then try to get "address@hidden/tlacontrib--devo--1.3"? Some kinds of alias expansion would turn that into address@hidden/address@hidden/tla--devo--1.3--1.3

But we don't want to require whitespace around aliases, because that prevents us from doing "tla changes devo--patch-1".

The existing patch-NN syntax would look like an alias in such a scheme, but
that's an unproblematic special case I think, and indeed, that feature might
even be able to use the alias mechanism if it were flexible enough (e.g.,
allowed `computational aliases', if only for internal tla use).

I'd rather not let aliases conflict with patch-N syntax. I think we should be able to specify "patch-N of the current tree-version" by using it.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]