| Age | Commit message (Collapse) | Author |
|---|
|
Ack-by: Zimmi48
Reviewed-by: ppedrot
|
|
Reviewed-by: herbelin
|
|
Reviewed-by: gares
|
|
|
|
This change improves the relaxed ambiguous path condition of coercions (#9743)
to check that any circular inheritance path of `C >-> C` is definitionally equal
to the identity function of the class `C`. Moreover, for a new inheritance path
`p : C >-> D` and existing (valid) one `q : C >-> D`, the new mechanism does not
report the ambiguity of `p` and `q` if they have a common element, that is to
say:
`p = p1 @ [c] @ p2` and `q = q1 @ [c] @ q2`
for some coercion `c` and inheritance paths `p1`, `p2`, `q1`, and `q2`.
In that case, convertibility of `p1` and `q1`, also, `p2` and `q2` should be
checked; thus, checking the ambiguity of `p` and `q` is redundant with them.
If the new mechanism does not report any ambiguous path, the inheritance graph
must be coherent [Barthe 1995, Sect. 3.2] [Saïbi 1997, Sect. 7]:
1. for any circular path `p : C >-> C`, `p` is definitionally equal to the
identity function, and
2. for any two paths `p, q : C >-> D`, `p` and `q` are convertible.
[Barthe 1995] Gilles Barthe, Implicit coercions in type systems, In: TYPES '95,
LNCS, vol 1158, Springer, 1996, pp 1-15.
[Saïbi 1997] Amokrane Saïbi, Typing algorithm in type theory with inheritance,
In: POPL '97, ACM, 1997, pp 292-301.
|
|
|
|
c.f. https://dev.azure.com/coq/coq/_build/results?buildId=6485&view=logs&jobId=2d2b3007-3c5c-5840-9bb0-2b1ea49925f3&j=2d2b3007-3c5c-5840-9bb0-2b1ea49925f3&t=77aad734-2057-5694-9ae2-ee1f5f26eae8
|
|
Apparently `expr 1 \+ 1` is fine on Linux but not cygwin/Windows, where
it fails with "syntax error". Similarly for `-` and `/`.
|
|
This reverts commit ec505a2fa67b0776b624be54417e06c6512f1734.
A better fix is coming
|
|
Apparently the bogomips produced by cygwin are extra-bogo.
|
|
Ack-by: Zimmi48
Reviewed-by: ppedrot
|
|
Reviewed-by: maximedenes
|
|
PR #9725 fixes completness bugs introduces some inefficiency. The
current PR intends to fix the inefficiency while retaining
completness. The fix removes a pre-processing step and instead relies
on a more elaborate proof format introducing positivity constraints on
the fly.
Solve bootstrapping issues: RMicromega <-> Rbase <-> lia.
Fixes #11063 and fixes #11242 and fixes #11270
|
|
Check that we don't regress on PR #10762 example
Fix regression discovered by Arthur in PR #10762
Fix script of #10298 which was relying on breaking semantics for `eapply`
Add doc
Add comment in clenvtac
Actually, always mark shelved goals as unresolvable
Update doc to reflect semantics w.r.t. shelved subgoals
|
|
only-printing notations
Ack-by: cpitclaudel
Reviewed-by: ejgallego
|
|
containing letins.
Reviewed-by: ppedrot
|
|
|
|
Reviewed-by: ejgallego
|
|
|
|
Fixes #10971
|
|
Reviewed-by: Zimmi48
Reviewed-by: ejgallego
|
|
Failing on CProdN([],...) was maybe a bit too radical.
|
|
- Warn in some places where {x:T} is not assumed to occur (e.g. in
argument of an application, or of a match).
- Warn when an implicit argument occurs several times with the same name.
- Accept local anonymous {_:T} with explicitation possible using name `arg_k`.
We obtain this by using a flag (impl_binder_index) which tells if we
are in a position where implicit arguments matter and, if yes, the
index of the next binder.
|
|
We renounce to the ad hoc rule preferring a notation w/o delimiter
for a term with coercions stripped over a notation for the
fully-applied terms with coercions not removed.
Instead, we interleave removal of coercions and search for notations:
we prefer a notation for the fully applied term, and, if not, try to
remove one coercion, and try again a notation for the remaining term,
and if not, try to remove the next coercion, etc.
Note: the flatten_application could be removed if prim_token were able
to apply on a prefix of an application node.
|
|
Reviewed-by: JasonGross
Reviewed-by: ejgallego
Reviewed-by: maximedenes
|
|
Ack-by: SkySkimmer
Reviewed-by: Zimmi48
Reviewed-by: maximedenes
|
|
Reviewed-by: gares
Reviewed-by: herbelin
|
|
And simplify a lot the compatibility infrastructure following this.
Update dev/tools/update-compat.py
Remove much complexity.
Co-authored-by: Jason Gross <jgross@mit.edu>
|
|
|
|
Reviewed-by: Zimmi48
Reviewed-by: silene
|
|
Reviewed-by: SkySkimmer
Reviewed-by: cpitclaudel
|
|
The manual was already saying that it was deprecated, but no warning was
emitted.
Fixes #10572
|
|
library
Reviewed-by: ejgallego
|
|
|
|
c.f. discussion at
https://github.com/coq/coq/pull/11177#issuecomment-559139477
|
|
In particular, display how long they took in bogomips-adjusted
centiseconds.
|
|
|
|
universes
Reviewed-by: Zimmi48
Reviewed-by: mattam82
Reviewed-by: ppedrot
|
|
Reviewed-by: herbelin
|
|
This increases backwards compatibility. If desired, we can add a tactic
notation to simplify the spec of `rapply` in the future if we want.
|
|
Also add a tactic notation so that it takes in uconstrs by default.
Also add some basic tests for `rapply`.
Also document rapply in the manual
|
|
Co-Authored-By: Hugo Herbelin <herbelin@users.noreply.github.com>
|
|
refined version of #8890 which prevents #11033.
Ack-by: Zimmi48
Reviewed-by: ejgallego
Ack-by: gares
|
|
Reviewed-by: ejgallego
|
|
Using the parameter universes in the constructor causes implicit
equality constraints, so those universes may not be template
polymorphic.
A couple types in the stdlib were erroneously marked template, which
is now detected. Removing the marking doesn't actually change
behaviour though.
Also fixes #10504.
|
|
Reviewed-by: ejgallego
Reviewed-by: herbelin
|
|
|
|
This is to hopefully prevent regressions on
https://github.com/coq/coq/issues/11150 and
https://github.com/coq/coq/issues/6502.
|
|
This is better than expecting other tests to fail if we mess up again.
|
|
We restrict #8890 so that it looks for a notation only for the fully
applied coercion.
|
