On Fri, Oct 27, 2000 at 10:50:12AM +0200, Daniel Stenberg wrote:
>
> Hm, I've never really concidered the first case being a separate function.
> I've more seen it as "info we can just as well give you when we do the update
> operation".
Sure. I just want a way to get that information without modifying my
working copy.
> Well, if you take the mentioned 'cvs status' command, it already today tells
> me the status of my working copy. It shows what files I've modified and which
> files that have been modified in the repository since my last update.
>
> In what way would this suggested other command differ?
Purely in the output format. I almost never use `cvs status' for any
reason other than to inspect sticky tags, because it is impossibly verbose.
I may be alone in this; but the few observations I've been able to make of
other cvs users suggest I'm not.
So it's fair to say that what I want would be achived simply by making
the default output format of `svn status' similar to what `cvs update'
gives you. I attach a great deal of importance to that output format. A
lot of people seem to like it, which is why I'm so gung-ho on making it
possible to get that format without side-effects. My suggestion loses
most of its weight if most CVS users use `cvs status' in the way that you
do, but I'm not sure that they do.
Machine-readability is the other reason the `cvs update' output format is
important. pcl-cvs, for example, has two commands that are nearly
identical in what they present to the user -- cvs-status and cvs-update
-- but they have to parse entirely different output formats. It would
be a convenience to the authors of such programs to have a standard output
format that would be available with or without side-effects.
> Personally, I usually have a little alias or script to do 'cvs status | grep
> Status: | grep Locally Modified' or similar to get a brief list of files I've
> modified...
>
> Or am I missing something?
Only, perhaps, the fact that the existence of such a shell script indicates a
deficiency in CVS's user interface. :-)
Received on Sat Oct 21 14:36:12 2006