Reviewing PRs and old SymPy issues

[Oscar Benjamin]

Hi all,

There are currently a large number of new contributors opening various
pull requests and commenting on issues in the SymPy repo. This is
great to see, and I want to thank everyone who has been contributing.

I would like to highlight though that opening pull requests is not the
only way to contribute to SymPy. There are many other ways to
contribute. In fact more important than opening pull requests is
*reviewing* pull requests and issues. For example if you can see that
a pull request likely needs some changes, then it is helpful to
comment on the pull request and suggest what changes are needed until
it looks ready to merge. I am sure that some of the new contributors
could tell quite quickly what changes would be needed in many of the
recently opened pull requests.

In fact one of the most useful things that a relatively new
contributor can do is to review old issues. The SymPy repo currently
has over 4000 open issues. Many of these are old and are no longer
relevant or already fixed. You can often review these issues just by
testing the code in them. If the issue seems to be fixed then a pull
request with a test could be opened to close the issue. If you do want
to review old issues then I suggest using the issue labels to focus on
a particular area of the code base like "integrals" or "solvers" etc.

This sort of work is actually much more helpful to the project than
opening pull requests for "easy to fix" issues.

--
Oscar

[Aaron Meurer]

I'll also add that if you want to, we'll give triage permissions to
just about anyone so they can go through and label old issues (triage
permissions give you permissions to label issues). Just ask me or
Oscar.

Aaron Meurer

> --
> You received this message because you are subscribed to the Google Groups "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAHVvXxQFLH7L%3DjO8EosasVETJn1fCmWaCvbYp%3DMbxvQmg1pB9Q%40mail.gmail.com.

[Shishir Kushwaha]

Sir , I'd like to review and test pull request,  kindly give me the triage access for labelling . Github ID - shishir-11

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAKgW%3D6%2B8WUQJa_d%2BATjAekCbRgi4D6XCevJqiCmf0EsieSf8gQ%40mail.gmail.com.

[HARSH KASAT]

Sir, I would like to request triage access to review the pull request, I have been working on in the sympy core branch. Could you please give me access to label triage my GitHub ID 'harshkasat'?

[Mohit Kumar]

Hello Aaron, I would like to have triage access, my github ID is MohitKumar020291. Thanks.

[Jason Moore]

Note that we get lots of PRs at this time because we require GSoC applicants to have at least one merged PR. We could encourage the other activities if we allow those to fulfill the requirement for the GSoC application.

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAKgW%3D6%2B8WUQJa_d%2BATjAekCbRgi4D6XCevJqiCmf0EsieSf8gQ%40mail.gmail.com.

[Oscar Benjamin]

Perhaps other things should be mentioned in the GSOC application
instructions. I do think that the PR requirement is reasonable as a
baseline for GSOC though.

To be clear opening pull requests is a good thing. Reviewing pull
requests is a lot of work for maintainers though. If lots of people
start opening PRs but not doing anything else to help by e.g.
reviewing those PRs then we will just have lots of unreviewed PRs
which is not so useful. Especially many of these PRs have very basic
problems like "a test should be added" or "the test should be for what
is actually fixed". These could easily be reviewed by many of the new
contributors.

This PR is a good example of the work problem that is created by
opening PRs rather than reviewing them:
https://github.com/sympy/sympy/pull/26117

What is needed is for someone to verify that the definition of
assoc_legendre as used in SymPy matches with the definition that is
used in mpmath. This is partly about reading the docs and partly just
a case of doing some numerical checks but it needs to be done
systematically paying close attention to any special values and to
what SymPy's symbolic evaluation handles. Writing the code and opening
the PR without doing that checking leaves all of the work to the
reviewer. Anyone else could review it though by doing that checking.

--
Oscar

> To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAP7f1AhzY_A2Rt9aMQy1M_6MzZ09L1uP0p_0RLcni1TpX8rwpw%40mail.gmail.com.

[Shishir Kushwaha]

Do you have some guidelines you follow or just some general advice for people who are willing to do that work . It will also help new contributors like me understand what things need to be checked apart from the ones mentioned in the developer documentation before I make a pull request .  

Shishir

[ABHISHEK KUMAR]

Sir I'd also like to review and test pull request , I've started contributing recently , My Github Id - abhiphile

abhiphile (ABHISHEK KUMAR) (github.com)

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAKgW%3D6%2B8WUQJa_d%2BATjAekCbRgi4D6XCevJqiCmf0EsieSf8gQ%40mail.gmail.com.

[Mohit Kumar]

I have found some old opened issues that should be closed, so, is it a good idea to open a new PR (which has title like "CLOSED [issue]..." with a test, if similar tests do not exists) or creating a new conversation here, IMO PR is a better option. PRs with title like CLOSED [issue].. could be prioritized?

[Shishir Kushwaha 5-Yr IDD: Mathematical Sci.s, IIT(BHU)]

Wouldn't creating another PR add to the problem of checking more PRs? 

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/a8251f51-7852-4128-ae68-5e85e589b44en%40googlegroups.com.

[Mohit Kumar]

IMO giving triage access to anyone will lead to future problems for SYMPY because wrong labels may lead to misunderstanding and complexities. Also, a better idea could be opening a root issue or root PR for this section (closing old issues).

To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAOCT_jRqYQ53sq%3DvC4uZsd%2B%3Dkr%2BMUff7-vUhG93a2-RbwZix4g%40mail.gmail.com.