Re: [anjuta-devel] GNOME Outreach Program for Women

[Nancí de Brito Bonfim <nancibonfim gmail com>]

Hi Sébastien,

   * New build plugin not using make file.

I think it's a good idea, it sounds very useful. Could be a nice entry
point.

I think the difficulty is to define how to implement it. I have described a way to do it but perhaps it can be done in a better way. The goal is to fill the need of someone wanting to learn C programming. I imagine that he doesn't know make and the compile and link steps should looks strange.

 
I'll research and think of a nice way for implement this.

   * Improve indentation

Sounds good too. I think the second one isn't difficult, but the first
one sounds complex. I can leave this task as one of the lasts.

Ok.

   * Improving debugger

Although I'm not very used to GDB, I think this set of tasks in very
interesting, because I could focus for a long period in the same part of
code.
Also it looks like another part that is requiring a lot of enhancements.
Also it could help me and the other users in the debugging of programs
using anjuta.

Ok, I have written most of this part but I spend most of the time on the low level functions and take less attention to the user interface.

I was talking mainly about the user interface. It's not that easy to use, principally for new programmers and I think that is a point that we can improve. The interface can simplify the use for the programmers that don't know GDB.
 

   * Port missing plugin to GSettings

I prefer something related to the user interface or at least to the user
directly. I feel more comfortable doing that.
if that isn't too hard I could do this too.

I have proposed quite a lots of topics to give you more choices. But I don't expect that you do everything and you can choose to do other things too, I even encourage you to do so.

Ok.

   * Improve version control

I'm interested on this. I could work on git and SVN if the git-shell
won't fix this for git.

Ok. I think James is not updating the SVN plugin anymore, for git it's better to see this with him.

 
James: Will git-shell fix this? Do you plan to work on this?

   * Drag and drop in the project manager view to open a new project.
I didn't quite understand this. What would I drag'n'drop?

The idea is to drag a directory from nautilus to drop it in Anjuta to open it as a new project. I think it could be a nice way to open a project but I'm not sure people will think that we have taken care of such detail.

This is eye-candy, but I don't think this is so needed to put this as a priority.
 

   * Convert the makefile project to the new API.

That is very interesting, because I will see the new project manager.
I've worked on the old project manager, at least I've read a lot of the
code and worked with it.
Sounds good for me to start working with this.

This API is new so it could change a bit. It's really different from the old project manager. The makefile plugin is completely in C (no more perl script) and uses flex and bison. Both are interesting tools but you don't need to understand them to do this job.

Ok, looks fine.
 

 

I liked a lot of your ideas. Maybe it's enough for three months, but I
can't be precise of whether I would fill three months with it or not.

Sure, it should be more than enough for three months. I have proposed quite a lots of idea to give you more choices. I don't expect that you complete everything. Anyway, if you do everything, I think we could find other things to keep you occupied :). At this point, you could probably find new things yourself.

Great. Thanks for your support :)

Massimo has proposed two additional ideas.

* An editor for highlight settings in GtkSourceView
GtkSourceView is the main editor, you can select different ways to highlight keyword but you cannot customized this. By example define that keywords must be in red and comments in gray. It's already existing for Scintilla in Edit->Preferences->Scintilla->Editor->Color and Fonts.

I think this task would take too long to finish. I would guess more than a month. This would be great, but I prefer the smaller tasks.
 

* Improve _expression_ parsing in symbol-db
I think the goal is to implement member completion. He wants to have test case for every improvement. I don't know if this work has already started.

I will think of this task. I thought of doing member completion, but I had planned some time to study. Johannes told me this was a hard task and would be OK if it was unfinished at the end of the program. I submitted it because there were no better options for me at the moment of the submission.
The task itself is a great one, but the main objective is to work with the community. In my humble opinion, since this task would require one or two weeks of studying, I liked the idea of doing smaller tasks.

I would have choose working with GDB instead of member completion if I had thought of it or if someone had suggested it for me.

If you think that it's better to work on this than in some of the other tasks, I could do this. What do you think?

Well, I didn't know I could make a lot of small tasks. I first though of
only one task, then it would be two.

It's a difference comparing to Google Summer of Code. The emphasis is less on writing a valuable piece of code and more on working with other people. You can do something else than code too, by example writing documentation or drawing icons.

Nice, I got it.

For small tasks, I think just fixing a deadline is enough. By example, I plan to fix bug #631139 and #631156 within the next 4 days. My goal is just to check that you don't have more troubles than I expect. Anyway before the beginning of the program, it's nice if you already start working on Anjuta but it's not require and I expect that you spend much less time.

I meant that I don't know yet if I will take three days or a week to make must of the tasks. I will take a look at the code on the next days and I'll try to fix deadlines for them.

I liked the idea of working on
several small tasks. Maybe I can choose something a little bigger to
work on the last  month, for instance - just like the GDB part, because
there are some enhancements I can think of.

Perfect.

Until the beginning of December I'll be in classes and it's hard for me
to be online at 21:00 (Paris time), but that is the end of evening for
me, so I'll be able to be on IRC almost every day.

I'll try to answer the other emails faster.

I spend quite some time on Anjuta. I'm not alone, Johannes is doing a lots of work too but I don't expect that everybody spent as much time. I think it's more important to be regular than fast.

 
I got it ;)

 

Regards and good luck for your exams

Thanks a lot,

Regards,

-- 
Nanci de Brito Bonfim