Hello, Le vendredi 11 avril 2025 à 09:49 -0400, Olivier Dion a écrit : > But some fields seems to > not be taken into account and you will end up with fetching > substitute > in the store. I want to avoid this element of surprise in my case.
>From what I understand, here (problem A), you want to avoid a hash collision (two different codes hashing to the same value)… > Like mentioned in the comment (look for the `(procedure? value)' > clause), this does _not_ yield the same hash for a procedure if the > module where this procedure came from is compiled. And here (problem B), you want to avoid having equivalent code (source vs bytecode) hashing to different values. Did I summarize it correctly? Solving A is easy: generate a new unique identifier each time you hash any functional value (or refuse to cache the result). I don’t have an answer for B, but others may. However, considering problem B, it may be more complex than just source and bytecode. What if the GOOPS object creator uses different dependencies, but they turn out to have the same behavior? I’m thinking about a non-free/free replacement situation. Since the object creator has a clearer vision of the functions and dependencies they use, maybe you could ask them to provide a version number of their relevant functional code, and hash this version number instead? Best regards, Vivien
