Re: [ANN] mc^2

2015-05-09 Thread Yury V. Zaytsev
Hi,

I've briefly had a look at mc^2, and I have to say that I was really
blown away by mooffie's work! I would be very happy if this can be
integrated into the master branch (mooffie can fix my pet peeve, the
gnome-terminal open directory thing in recent distros with just a few
lines of code!).

I feel that scripting is a vastly superior solution to the extensibility
problem in the context of mc as compared to plugins. Of course, plugins
have their uses, but they will require so much more effort to build and
maintain, whereas scenarios for a scripting engine that exposes mc
internals through a very thin layer are quick to write, easy to debug
and need no compilation.

Slava, what do you think? I know that you've been a proponent of the
plugin system, but aren't you impressed with this demonstration?

From the maintenance point of view, it would be great if most of the
tickets asking for niceties could be implemented in a high level
language outside of the core of mc, or maybe even in a longer term more
non-essential features that are currently in the core could be offloaded
to scripts, freeing up resources for important infrastructural projects
like broken regexes.

We could ship a default library of scripts with some active out of the
box, and others requiring opt-in (symlink). Think of no more arguments
about changing the gold default behavior, or adding more configuration
options...

In addition to that, we could make an official script repository with
rather loose inclusion criteria (basically, script is maintained 
compatible with the latest version of mc, and maintainer demonstrates
minimal amounts of sanity), which distro packagers could ship as a
separate addon, like they do e.g. with bundles of vim plugins.

Anyways, how do we move from there on? I'm afraid that personally I lack
competence and time to do a proper code review to help the process :-(

-- 
Sincerely yours,
Yury V. Zaytsev


___
mc-devel mailing list
https://mail.gnome.org/mailman/listinfo/mc-devel


Re: [ANN] mc^2

2015-05-09 Thread Yury V. Zaytsev
On Fri, 2015-05-08 at 19:33 +0300, Andrew Borodin wrote:
 
 It would be great if you split first huge commit of branch in several
 small commits. It will make the review easier. 

I couldn't resist and started looking at the code anyways, but quickly
discovered that a large part of the diff is simply changes to the
PO-files and such.

It would be really of huge help, if the first commit would be split in a
bunch of smaller commits with logically related changes, and the PO
stuff definitively has to go into its own separate commit.

-- 
Sincerely yours,
Yury V. Zaytsev


___
mc-devel mailing list
https://mail.gnome.org/mailman/listinfo/mc-devel


Re: [ANN] mc^2

2015-05-09 Thread Mooffie
On 5/8/15, Andrew Borodin aboro...@vmail.ru wrote:

 It would be great if you split first huge commit of branch in several small
 commits. It will make the review easier.

I'll take notice of that.
___
mc-devel mailing list
https://mail.gnome.org/mailman/listinfo/mc-devel


Re: [ANN] mc^2

2015-05-09 Thread Mooffie
On 5/9/15, Yury V. Zaytsev wrote:

 Anyways, how do we move from there on?

I'll be doing two things:

- I'll prepare a document explaining to reviewers the overall
structure of the code and the modifications to existing files.

- I'll look into porting the code to 4.8.14.

The thing I'll appreciate from reviewers at this early stage is
not to criticize the small things (e.g., coding style) but to look
at the big picture.

(I already know some places where I'll get criticism. E.g., in
places where I didn't wan't to refactor things in the old code (e.g., in
src/filemanager/panel.c). Why didn't I? because I didn't think
it was wise to refactor/modify old code before I first get an ok on
the big picture.)
___
mc-devel mailing list
https://mail.gnome.org/mailman/listinfo/mc-devel