Factoring out a common interface #73
Conversation
Signed-off-by: Stefan Krastanov <[email protected]>
Signed-off-by: Stefan Krastanov <[email protected]>
Signed-off-by: Stefan Krastanov <[email protected]>
Codecov Report
@@ Coverage Diff @@
## master #73 +/- ##
==========================================
- Coverage 92.66% 92.52% -0.15%
==========================================
Files 24 23 -1
Lines 3164 3064 -100
==========================================
- Hits 2932 2835 -97
+ Misses 232 229 -3
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
|
I think this is ready for a first-pass review. I will comment on a few points in the diffs |
Krastanov
changed the title
|
Of note, the compat for QuantumInterface is set to the current LTS, julia 1.6, while the compat for QuantumOpticsBase is still at 1.3. I personally find it reasonable to push the compat to at least LTS. |
|
friendly weekly bump on the QuantumInterface PR having this in a published version would make it easier for me to work on QuantumSavory |
| @@ -16,6 +17,7 @@ UnsafeArrays = "c4a57d5a-5b31-53a6-b365-19f8c011fbd6" | |||
| Adapt = "1, 2, 3.3" | |||
| FFTW = "1.2" | |||
| LRUCache = "1" | |||
| QuantumInterface = "0.1.0" | |||
There was a problem hiding this comment.
I think it means the same. 0.1.x implies [0.1.x, 0.2) and 0.1 implies [0.1.0, 0.2) which are the same for x=0. But maybe I misunderstood https://pkgdocs.julialang.org/v1/compatibility/
There was a problem hiding this comment.
Oh, my mistake. Thanks for clarifying.
|
@Krastanov changes here look good to me (left one small comment regarding the compat). I'd now:
Thanks for the work! |
|
@Krastanov CI with JET appears to be failing after the merge. Should I hold off releasing a new version? |
|
@david-pl , I think this is fine. This CI would be failing relatively often because it is running on nightly and JET and julia nightly frequently get temporarily out of sync (JET uses some deep julia internals). If the error is "outside of tests" it is probably because of a nightly issue. If it is actually inside the tests, then there is something interesting to look for. If this is too inconsistent, I am happy to remove JET from CI until I find a better way to keep them in sync? |
|
Also, the reason this is a whole separate CI is so that it is not executed when Julia devs runs PkgEval. That way intermittent noise in the CI does not pollute the ecosystem CI. |
|
@Krastanov thanks, just wanted to clarify this again to be sure. As far as I can tell it's a test failure, but due to a MethodError in I created a new release so you should be good to go once the PR is merged. |
|
Indeed, JET is calling various internal undocumented Julia compiler internals in order to do abstract interpretation. The author of JET is probably the main sources of changes to this part of julia, but every so often JET takes a bit to get updated to use the newly changed internals. Maybe it would be reasonable to simply not run it on nightly, however JET improves so fast right now, that I am happy to deal with the intermittent CI failures among my projects. Happy to adjust the CI to the QuantumOptics preferences though! |
related to qojulia/QuantumInterface.jl#1
Best to keep overall discussion in qojulia/QuantumInterface.jl#1 and postpone discussing this PR until a decision on the above issue is made.