459.bugfix

[summatix]

What do these changes do?

When running in a pool, upon close, fixes the error:

An open stream object is being garbage collected; call "stream.close()" explicitly.

[prryplatypus]

Just bumping this since I seem to still have this issue 6 months after the PR was opened. Any progress done at all?

[summatix]

This works. The PR was just never merged.

[prryplatypus]

Gonna be the annoying guy that tags @webknjaz to ask if he can have a look then, pls :)

[webknjaz]

I'm not the best person to review this. So I cannot accept this since I don't know if this design decision (which is effectively extending the public API) is desired. But what I can tell you is that I'd not merge this w/o any tests proving that this works.

[codecov]

Codecov Report

Merging #459 (1dad293) into master (196e325) will decrease coverage by 1.76%.
The diff coverage is 85.71%.

@@            Coverage Diff             @@
##           master     #459      +/-   ##
==========================================
- Coverage   92.66%   90.90%   -1.77%     
==========================================
  Files          35       35              
  Lines        5413     5419       +6     
  Branches      584      585       +1     
==========================================
- Hits         5016     4926      -90     
- Misses        309      399      +90     
- Partials       88       94       +6     

Flag	Coverage Δ
ubuntu-latest_3.10_mariadb-10.2	89.44% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mariadb-10.3	89.44% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mariadb-10.4	89.44% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mariadb-10.5	89.42% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mariadb-10.6	89.42% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mariadb-10.7	89.42% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mysql-5.7	90.14% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.10_mysql-8.0	?
ubuntu-latest_3.7_mariadb-10.2	88.85% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mariadb-10.3	88.85% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mariadb-10.4	88.85% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mariadb-10.5	88.83% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mariadb-10.6	88.83% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mariadb-10.7	88.83% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mysql-5.7	89.60% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.7_mysql-8.0	?
ubuntu-latest_3.8_mariadb-10.2	89.48% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mariadb-10.3	89.48% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mariadb-10.4	89.48% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mariadb-10.5	89.46% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mariadb-10.6	89.46% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mariadb-10.7	89.46% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mysql-5.7	90.18% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.8_mysql-8.0	?
ubuntu-latest_3.9_mariadb-10.2	89.35% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mariadb-10.3	89.35% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mariadb-10.4	89.35% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mariadb-10.5	89.33% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mariadb-10.6	89.33% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mariadb-10.7	89.33% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mysql-5.7	90.05% <85.71%> (-0.01%)	⬇️
ubuntu-latest_3.9_mysql-8.0	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
aiomysql/connection.py	78.59% <83.33%> (-5.96%)	⬇️
aiomysql/pool.py	95.18% <100.00%> (-1.81%)	⬇️
tests/test_sha_connection.py	38.63% <0.00%> (-61.37%)	⬇️
tests/conftest.py	81.90% <0.00%> (-5.03%)	⬇️
tests/test_connection.py	97.34% <0.00%> (-2.66%)	⬇️

---

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 196e325...1dad293. Read the comment docs.

[webknjaz]

FWIW I've restarted the CI and it fails.

[summatix]

This is quite clearly a bug, not a design choice.

I no longer use this library and so I no longer have interest in pushing this PR forward. My fork is available to those who encounter this bug.

[RazzerDE]

This is quite clearly a bug, not a design choice.

I no longer use this library and so I no longer have interest in pushing this PR forward. My fork is available to those who encounter this bug.

Did you found a better, working library for async MySQL? I really need one and this library is just broken.