Masatake YAMATO <[EMAIL PROTECTED]> writes:

> Good. BTW, Is there anyway to generate conflicts repeatedly?
> I'd like to write bridge xtla and smerge-mode. 

Well, if  you know what you  expect, you can also  have your "private"
test  project. Get  two working  copy,  modify the  first and  commit.
Modify the second and update. 

There's not much to do for  a "bridge" between smerge and xtla. Mostly
a good keybinding and a menu item I think. 

The code I had written to  solve conflicts should be dropped for that.
smerge-mode is better. 

> - The command is rather directory oriented. It is not bad to be called from 
> dired.
>   (I like dired:-) The bad news is that ?T is bound to dired-do-touch.

I don't like dired ;-)

Well, nothing against it, but  I do file manipulation from the command
line, so,  I prefer not  touching dired-related code, because  I don't
know what people would expect from it.

> - A bit crazy idea: Bind the funciton to ?< in *tla-revision* buffer.
>   Analogy of ?> (get).

You usually  visit *tla-revision* buffer for an  existing version, and
here, we're going to import a  new one. This would only be meaningfull
from a newly created version, with '+ v' in the versions buffer. 

It doesn't seem to me to be the best approach, but that may be usefull
to have it in addition to the directory-oriented feature, because some
people (mainly  beginners) may want to  create a new  project from the
archive browser.

> One of the primary target of xtla is pcl-cvs users. pcl-cvs users may expect 
> xtla.el
> to have functions found in pcl-cvs.

I'm an ex pcl-cvs user.  I never used CVS for cooperative development,
but  only for multi-sites  development and  history management,  so, I
almost never had to manage conflicts, and updates were trivial. So, my
main use  of PCL-CVS  was just  to see the  list of  "modified" files,
select them, and commit.

This is  not what I find in  the inventory buffer, but  in the changes
buffer. 

That's why I  almost never use the inventory buffer, but  I'm a fan of
the M-x tla-changes RET feature. (Which I'm very happy to see bound to
C-x T c ;-)

> How do you think add "priority" to the TODO list item?
>
> e.g.
>
> Toward version 1.0:
>        - tla-invetory
>          ...
>          ... 

         - tla-changes improved. 

Anyway,  my vision  of the  priority is  : When  I need  a  feature, I
implement it. 


BTW,  anyone has an  idea of  when the  next Emacs  release will  be ?
Should we target a 1.0 release before that ? 

-- 
Matthieu

Reply via email to