[Biopython-dev] 'testseq' update (6/10/17)

[Adil Iqbal]

*TL:DR - Game plan: (1) Resolve unittest importing issue. (2) Revert
testseq seeding. (3) Revisit testseq doctests. (4) Update unittest for
reverted seeding. (2) Thank you for your patience.*

Hello all,

I'd like to let everyone know where 'testseq' is in development. You can
find the code for it here:
https://github.com/Adil-Iqbal/Personal-Projects/tree/master/Test%20Sequence

The biggest issue (unexpectedly) was importing the 'testseq' module into
the 'test_testseq' unit test. Absolute and relative importing did not work
since the 'Scripts' folder is not a part of the 'Bio' package. After trying
out several solutions, the one that looks the most promising is using the
"importlib" module. The issue is not yet resolved but I am confident that
it is just a matter of working out the details.

I've gotten the chance to tackle part of another big issue, which is python
version and operating system compatibility. Since the 'testseq' function
uses randomization, its output can be different across versions and
platforms. This poses a problem for code testing. Luckily there are a few
work-arounds that I've implemented in the unit test which should work fine.
I will also be revisiting the 'testseq' doctests to ensure they are also
compatible, though I have not yet run into any issues with that on Travis
CI. (Peter warned me about this. ^_^)

Today, an argument was presented for reverting the seeding implementation,
which was that user intuition should preferably be developed with respect
to best practices. Unless anyone has a reasonable rebuttal, I will probably
revert the seeding to an earlier state. Though, that will have to wait
until after I've resolved the importing issue.

Also, I'd like to thank you for being patient with me. I would've completed
this sooner, but I have other obligations as well.

Warmest regards,
Adil
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.open-bio.org/pipermail/biopython-dev/attachments/20170610/65ca6102/attachment.html>