Retry(): Support negative retries value

[kristjanvalur]

Pull Request check-list

• Does $ tox pass with this change (including linting)?
• Do the CI tests pass with this change (enable it first in your forked repo and wait for the github action build to finish)?
• Is the new or changed code fully tested?
• Is a documentation update included (if this change modifies existing APIs, or introduces new ones)?
• Is there an example added to the examples folder (if applicable)?
• Was the change added to CHANGES file?

Description of change

The Retry class now accepts a negative value for retries indicating no upper limit on the number of retries; the action
will be retried until it succeeds.

[codecov-commenter]

Codecov Report

Merging #2110 (42d8069) into master (805b184) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@ Coverage Diff @@ ## master #2110 +/- ## ========================================== + Coverage 92.43% 92.45% +0.01%  ========================================== Files 104 104 Lines 24363 24387 +24 ========================================== + Hits 22520 22547 +27  + Misses 1843 1840 -3 

Impacted Files	Coverage Δ
redis/asyncio/retry.py	96.15% <100.00%> (ø)
redis/retry.py	100.00% <100.00%> (ø)
tests/test_asyncio/test_retry.py	100.00% <100.00%> (ø)
tests/test_retry.py	100.00% <100.00%> (ø)
tests/test_graph.py	90.62% <0.00%> (+0.03%)	⬆️
redis/asyncio/connection.py	84.83% <0.00%> (+0.34%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 805b184...42d8069. Read the comment docs.

[WisdomPill]

It seems to me that sync version tests are missing

[WisdomPill]

did you break pypy-3.7 standalone-hiredis tests?

[WisdomPill]

can you rebase? maybe that's the reason why the CI is broken

[WisdomPill]

LGTM!

maybe good old @chayim should have another look here since I am quite new into reviewing PRs in this repo

[chayim]

This looks like a red herring. I expect the python 3.11a builds to fail (hence, they don't break CI). But - the change looks innocuous enough. I suspect it triggered something else. Re-running the action.

[WisdomPill]

@chayim nice catch! :)

[kristjanvalur]

Yes, pretty sure nothing in the change was causing these builder issues.