| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
These are also convenient from `vernac` [to be used in future PRs].
|
|
We fix quite a few types, and perform some cleanup wrt to the
evar_map, in particular we prefer to thread it now as otherwise
it may become trickier to check when we are using the correct one.
Thanks to @SkySkimmer for lots of comments and bug-finding.
|
|
In the test we do [let X : Type@{i} := Set in ...] with Set
abstracted. The constraint [Set < i] was lost in the abstract.
Universes of a monomorphic reference [c] are considered to appear in
the term [c].
|
|
- Regularly declared for for polymorphic constants
- Declared globally for monomorphic constants.
E.g mono@{i} := Type@{i} is printed as
mono@{mono.i} := Type@{mono.i}.
There can be a name clash if there's a module and a constant of the
same name. It is detected and is an error if the constant is first
but is not detected and the name for the constant not
registered (??) if the constant comes second.
Accept VarRef when registering universe binders
Fix two problems found by Gaëtan where binders were not registered properly
Simplify API substantially by not passing around a substructure of an
already carrier-around structure in interpretation/declaration code of
constants and proofs
Fix an issue of the stronger restrict universe context + no evd leak
This is uncovered by not having an evd leak in interp_definition, and
the stronger restrict_universe_context. This patch could be backported
to 8.7, it could also be triggered by the previous restrict_context I
think.
|
|
We mirror the structure of EConstr and move the destructors from `Term`
to `Constr`.
This is a step towards having a single module for `Constr`.
|
|
We do up to `Term` which is the main bulk of the changes.
|
|
This code was probably slightly wrong w.r.t. to a semantics defined as
equivalent to first full-blown normalization followed by kernel term equality.
Or at least, it was adding redundant constraints.
|
|
The code from Universes what essentially a duplicate of the one from EConstr,
but they were copied for historical reasons. Now, this is not useful anymore,
so that we remove the implementation from Universes and rely on the one from
EConstr.
|
|
|
|
Reminder of (some of) the reasons for removal:
- Despite the claim in sigma.mli, it does *not* prevent evar
leaks, something like:
fun env evd ->
let (evd',ev) = new_evar env evd in
(evd,ev)
will typecheck even with Sigma-like type annotations (with a proof of
reflexivity)
- The API stayed embryonic. Even typing functions were not ported to
Sigma.
- Some unsafe combinators (Unsafe.tclEVARS) were replaced with slightly
less unsafe ones (e.g. s_enter), but those ones were not marked unsafe
at all (despite still being so).
- There was no good story for higher order functions manipulating evar
maps. Without higher order, one can most of the time get away with
reusing the same name for the updated evar map.
- Most of the code doing complex things with evar maps was using unsafe
casts to sigma. This code should be fixed, but this is an orthogonal
issue.
Of course, this was showing a nice and elegant use of GADTs, but the
cost/benefit ratio in practice did not seem good.
|
|
short econstr-cleaning of record.ml
|
|
|
|
|
|
This is the continuation of #244, we now deprecate `CErrors.error`,
the single entry point in Coq is `user_err`.
The rationale is to allow for easier grepping, and to ease a future
cleanup of error messages. In particular, we would like to
systematically classify all error messages raised by Coq and be sure
they are properly documented.
We restore the two functions removed in #244 to improve compatibility,
but mark them deprecated.
|
|
|
|
This allows to use a cast in subst_of_rel_context_instance.
Also added more cast functions for further use.
|
|
|
|
|
|
|
|
The transition has been done a bit brutally. I think we can still save a
lot of useless normalizations here and there by providing the right API
in EConstr. Nonetheless, this is a first step.
|
|
For now we only normalize sorts, and we leave instances for the next
commit.
|
|
|
|
The kernel does fishy things with casts, such that ensuring there are no
two consecutive VMcast or NATIVEcast in terms. We enforce this in EConstr
view as well.
|
|
This reverts commit b5f07be9fdcd41fdaf73503e5214e766bf6a303b. The performance
difference was not conclusive enough to pay for the code ugliness.
|
|
We do one step of loop unrolling, limit the number of allocations and
mark the function as inline.
|
|
|
|
|
|
This removes code duplication between Evarutil and EConstr.
|
|
Incidentally, this fixes a printing bug in output/inference.v where the
displayed name of an evar was the wrong one because its type was not
evar-expanded enough.
|
|
This removes quite a few unsafe casts. Unluckily, I had to reintroduce
the old non-module based names for these data structures, because I could
not reproduce easily the same hierarchy in EConstr.
|
|
|
|
|
|
|
|
|
|
This allows to factorize code and prevents the unnecessary use of back and
forth conversions between the various types of terms.
Note that functions from typing may now raise errors as PretypeError rather
than TypeError, because they call the proper wrapper. I think that they were
wrongly calling the kernel because of an overlook of open modules.
|
|
|
