[MonoDevelop] [GSoC] Python debugger

Christian Hergert christian.hergert at gmail.com
Thu Mar 25 14:46:02 EDT 2010


On Thu, Mar 25, 2010 at 2:26 AM, Martin Vejnár <avakar at ratatanek.cz> wrote:
> I should probably say that I am very new to MD (I first installed it three
> weeks ago), my knowledge of its codebase is very limited -- mostly to the
> debugger -- and I've only skimmed the PyBinding sources. As such, it is
> difficult for me to gauge the complexity of the tasks below.

That's why you get a mentor :-)

> Furthermore I should mentioned that I won't be able to invest more than 120
> hours on the project. Perhaps that time budget is not sufficient for a GSoC
> project, but I do think I'd manage to write the debugger support and
> complete some of the tasks I've picked out below.
>
> I will have to rely on you to help me choose the appropriate tasks. Thank
> you all in advance for your comments.

Good to know.

> On 3/24/2010 11:09 PM, Christian Hergert wrote:
>>
>>  * Enhance code completion to support multiple-inheritance
>
> I assume you are referring to a proper MRO support?

Right.

>>  * Smart formatting for the python syntax following PEP8
>>  * Basic refactoring such as Extract/Rename method, rename class
>>  * Unit testing (twisted-trial, pyunit, etc)
>
> Unit testing is interesting, but at the moment I have no idea how it
> integrates with MD. I'll try to get up to speed before I write my proposal.

Thankfully, there is already a NUnit addin which could hopefully share
a lot of code for the visualization aspects.

> On Wed, Mar 24, 2010 at 2:58 PM, Michael Hutchinson
> <m.j.hutchinson at gmail.com>  wrote:
>>
>> Martin: if you're interested in Python + MD, you might also like to
>> consider the IronPython addin idea - note that you can submit multiple
>> proposals.
>
> I personally have no experience with IronPython. As I am primarily a CPython
> user, I'd much rather work on supporting debugging than on .NET integration.

I don't mean supporting IronPython in any way.  Simply that the
language parser I wrote is *such* a hack that it would be much
faster/efficient to use the IronPython language parser in-process.
(It just generates an AST which we can analyze).

Anyway, just trying to provide ideas.  Ultimately, it's up to you to
propose what you'd like to do.

Good luck!

-- Christian


More information about the Monodevelop-list mailing list