octave-maintainers
[Top][All Lists]
Advanced

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

Re: Progress Report: Symbolic Project


From: Colin Macdonald
Subject: Re: Progress Report: Symbolic Project
Date: Tue, 31 May 2016 12:53:33 -0700
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.1.0

[oops for got reply-to-list earlier!]

On 31/05/16 12:13, Colin Macdonald wrote:
> On 31/05/16 11:34, Abhinav Tripathi wrote:
>> |30th May Work on adding the sym conversion to PyTave and cleaning up
>> the conversion mechanism in Symbolic.|
> 
> I think this project has gone well so far!
> 
>> Looking on the midterm goals (posted on octave wiki as well as on the
>> blog), I feel we have acheived most part of 1 a) through d) and also 2
>> a) [I hope so]. Is something left?
> 
> Well, its all subject to change but yes I think we have an initial
> implementation that satisfies those.
> 
>> As 2b) and 2c) state, to increase test coverage for pytave. I saw that
>> almost all basic functions have a test for them (in python). Do we have
>> to get more specific test cases? Some clarification would be good.
> 
> I do think we do need more tests.  As I've commented elsewhere, I don't
> see any tests in Octave of using PyTave.  All of the tests seem to be
> about calling Octave from Python.
> 
> I think pyexec, pyeval and pycall should have BIST (`%!` tests).  A
> couple illustrative doctests are useful too.  I did add some of those.
> 
> address@hidden will need to have BIST as well.
> 
>> Also the next phase of timeline states: |15th June Improve tests and
>> doctests. Work on building PyTave and testing on Windows. No more crufty
>> XML!|
>> I guess we have already gotten rid of the XML part (haven't we?)
> 
> Yes, and good riddance ;-)
> 
>> So, should we now focus on tests? (for pytave and also symbolic). Or do
>> we have to get some other conversion stuff cleared up before that?
> 
> My thoughts is to work on both:
> 
> (a) tests
> (b) there is still "crufty" importing from Symbolic to Python---even
> though its not XML.
> 
> Thanks for the update and keep up the great work!
> 
> Colin
> 




reply via email to

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