Conversation
|
We need to change CI to use coq-8.20 and mathcomp2.3.0, if possible. |
ce27ee7 to
46e028f
Compare
|
Let me try to add them. |
0f7590b to
b62f840
Compare
between the full sequence of events and the concatenation of the processed and future events
|
The new error message states that the build of infotheo is breaking. @affeldt-aist I need your help for this, but there is no hurry. |
|
We have released recent versions of infotheo and I PRed a fix for an older version. This should fix the problem but we need to wait a bit for opam to be updated. |
|
The second CI error (with Coq 8.20) is: and indeed I have the same error on my laptop |
|
I know I corrected that file, that is why I believed I had pushed a correct version, but now I can't find my corrected version anymore. This is pain because I had a difficult modification of an inductive proof to perform, and I can't remember the exact trick. I will have to think again. Yves |
…duction proof in complete_disjoint_general_position has been fine-tuned
|
This PR is now ready for squashing and merging. The problem with coq-8.19 has been studied and a solution is proposed in https://github.com/ybertot/trajectories/tree/lra_patch_for_coq_8.19 , but cannot be merged here because it adds workarounds that should not be kept for the long term. |
This PR:
Z from the stdlib with the MathComp structures -> use
ssrZ.vinstead)make -j3compiles up-togeneral_position.vthat I am not sure is a file that actually ought to compile