Florian Angeletti
Florian Angeletti
Did I accidentally stall this PR? Or is it waiting on another issue?
The correct syntax is ```ocaml module Top1: sig ... end[@@deprecated "A"] ``` or ```ocaml module[@deprecated "A"] Top1: sig ... end ``` The improved warning is reliable in the sense that...
The example in question is indeed a case where the attribute was ignored.
This is the case where a floating attribute in the "header" of the file (before non-attribute items) works: ```ocaml [@@@deprecated "ok"] val x: int ``` ```ocaml val x:int [@@@deprecated "not_in_the_header"]...
@joakin , did you have the time to try to reinstall `zstd` ?
`_25` is not `'a`, if we call it `𝛽`, we have `'a = 𝛽 t t` (eliding the `Seq.` prefix), and the graph is the final sub-unification problem from trying...
There are visibly too many `'a` in this PR description, sorry. And the node `β=25` is none of those `'a`. This node is introduced when inferring the type of `Seq.(cons...
What happens is that the line ```ocaml select_best_proposition [del;insert;diag] ``` enforces an implicit reverse lexicographic order on the minimal weight paths. In more details, consider the case where the three...
1. For over-applied functors, it doesn't feel accurate to consider that arguments on the right are more likely to be erroneous. For deletions, it is the asymmetry of partial applications...
I have added the explanation for the path preference in the diffing matrix in the code. I have also tweaked a bit further the weights for the definition diffing, with...