help-smalltalk
[Top][All Lists]
Advanced

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

Re: [Help-smalltalk] VisualGST


From: Paolo Bonzini
Subject: Re: [Help-smalltalk] VisualGST
Date: Thu, 31 Dec 2009 18:52:17 +0100
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.5) Gecko/20091209 Fedora/3.0-4.fc12 Lightning/1.0pre Thunderbird/3.0

On 12/31/2009 01:29 PM, Gwenael Casaccio wrote:
   - one window ide

This should come first if only for the refactoring opportunities it requires (if you have a goal for refactoring, it's more likely that you do it...).

Introducing abstract models without removing duplicate code in the concrete widgets and models, means doing useless work (introducing abstract models in 5 places instead of 2, say).

   - package management !
   - git integration !

See below.

   - improve the debugger

Fix-and-continue would definitely be needed.

   - improve the sunit tool

Like?

   - a dialog box for the class creation

Yes, nice thing to have.

   - add side bar
         * on the left : a file browser

Why a side bar and not a separate tab?

>         * on the right : gtk documentation browser
>                          code snippet
>                          inspector

I would see the sunit tool integrated into every browser as a side bar. I don't understand exactly how you plan to introduce inspectors: as a sidebar to a workspace, moving workspaces to the top half of the visualgst window? Or for browsers? Maybe an inspector on the class pool, i.e. for class variables?

Again, however: first refactor, then introduce sidebars in the abstract tool classes or into the VisualGST class, then add more concrete sidebars.

It's a pity that GTK doesn't provide dockable windows, tools such as the ones you mention or workspaces would work well there.

>                           a package manager

I would see this as a different view replacing the category/namespace view (using tabs), not as a sidebar.

I would also add refactoring to this list, and icons for superclass-defining/subclass-overridden methods.

Paolo




reply via email to

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