Hi @wujianjack - Please read the contribution guide for details on how to resolve the black/typos errors: https://pyomo.readthedocs.io/en/stable/contribution_guide.html

Hi @wujianjack - Please read the contribution guide for details on how to resolve the black/typos errors: https://pyomo.readthedocs.io/en/stable/contribution_guide.html
@mrmundt Thanks, I have resolved the typos. Sorry for the late reply!

Hi @wujianjack - It looks like you are missing an import statement which is causing these failures. Please let me know if you need any assistance.

@wujianjack feel free to reopen this if you have time to come back to it.

@wujianjack feel free to reopen this if you have time to come back to it.

Thanks, I updated on my fork, but still failed on those checks, I will try to figure out what's going on.

@wujianjack feel free to reopen this if you have time to come back to it.

BTW, I have no access to reopen this PR, can you please reopen it?

Hi @wujianjack - It looks like you are missing an import statement which is causing these failures. Please let me know if you need any assistance.

Sorry for the delay. Now it passes the style and typos check, but still failed on other checks, like bare-env, where the CI throw exception RuntimeError: Attempting to use an unavailable solver., and I have no idea how to fix it. Thanks.

@wujianjack - I am seeing a lot of Segmentation faults; does that happen when you run the tests locally?

@wujianjack - I am seeing a lot of Segmentation faults; does that happen when you run the tests locally?

Sorry for the delay, the latest update fixes the segfault, now there are some mismatch regarding to the slack values. (In COPT, slack means activity of constraints in fact). I will think about how to resolve it.

Thanks!

@wujianjack - I am seeing a lot of Segmentation faults; does that happen when you run the tests locally?

Hi @mrmundt

I tried to fix most of the failures and there are 5 failures remains with latest change.

One failure is related to IIS, we writes out as c2: y + 100 x <= 0 instead of c2: 100 x + y <= 0 required by the testing script;
The other four failures are related to accuracy issue of variable solution (up to 4 digits).
See below for details.

FAILED pyomo/contrib/appsi/solvers/tests/test_persistent_solvers.py::TestSolvers::test_mutable_quadratic_coefficient_6_copt_only_child_vars_True - AssertionError: 0.41003838775982643 != 0.41024548525899274 within 4 places (0.00020709749916630882 difference)
FAILED pyomo/contrib/appsi/solvers/tests/test_persistent_solvers.py::TestSolvers::test_mutable_quadratic_coefficient_7_copt_only_child_vars_False - AssertionError: 0.41003838775982643 != 0.41024548525899274 within 4 places (0.00020709749916630882 difference)
FAILED pyomo/contrib/appsi/solvers/tests/test_persistent_solvers.py::TestSolvers::test_mutable_quadratic_objective_6_copt_only_child_vars_True - AssertionError: 0.696075712924909 != 0.6962249634573562 within 4 places (0.0001492505324471871 difference)
FAILED pyomo/contrib/appsi/solvers/tests/test_persistent_solvers.py::TestSolvers::test_mutable_quadratic_objective_7_copt_only_child_vars_False - AssertionError: 0.696075712924909 != 0.6962249634573562 within 4 places (0.0001492505324471871 difference)
FAILED pyomo/contrib/iis/tests/test_iis.py::TestIIS::test_write_iis_copt - Exception: The file /tmp/tmpl9qyjxo7/copt_iis.ilp is not as expected. Missing constraints:
c2: 100 x + y <= 0

Any suggestion on how to fix these failures? Thanks!

Are there still any plans to integrate COPT to Pyomo?

@l-kotzur Good question. At the moment, I suspect the answer is "not immediately": this PR has been dormant for quite a while (it wasn't passing tests, and now has conflicts with main). Additionally, the dev team is in the middle of finalizing a new solver interface standard (in pyomo.contrib.solver) and this PR is adding COPT to the legacy interface / APPSI. I suspect that COPT will (eventually) be incorporated into the new interfaces as we work to port over all the solvers.

If the original author (or a new interested developer) wants to revive it, it would be welcome (and that would probably accelerate things significantly).

I added COPT to the list of solvers we would like to port to the new solver interfaces at some point.