| Age | Commit message (Collapse) | Author |
|---|
|
We interpret `{|proj=pat|}` as `@C _ pat` instead of `C _ pat` (where
the `_` stands for the parameters).
Fix #12534
|
|
|
|
Re-raising inside exception handlers must be done with care in order
to preserve backtraces; even if newer OCaml versions do a better job
in automatically spilling `%reraise` in places that matter, there is
no guarantee for that to happen.
I've done a best-effort pass of places that were re-raising
incorrectly, hopefully I got the logic right.
There is the special case of `Nametab.error_global_not_found` which is
raised many times in response to a `Not_found` error; IMHO this error
should be converted to something more specific, however the scope of
that change would be huge as to do easily...
|
|
Reviewed-by: Matafou
Ack-by: SkySkimmer
Reviewed-by: gares
|
|
|
|
This is extracted from #9710, where we need the environment anyway to compute
iota rules on inductive types with let-bindings. The commit is self-contained,
so I think it could go directly in to save me a few rebases.
Furthermore, this is also related to #11707. Assuming we split cbn from the
other reduction machine, this allows to merge the "local" machine with
the general one, since after this PR they will have the same type. One less
reduction machine should make people happy.
|
|
constructor.
Moreover, the link to the constructor was hiding other contents of the
tuple.
|
|
This allows to have the "No local fields allowed in a record
construction" error applicable to all fields and not only the first
one. Formerly, this was wrongly raising an error "This record contains
fields of both T and T".
|
|
We basically avoid a detour via intern_applied_reference.
In particular, this stops dumpglobbing the name of the "constructor"
of the record which in practice does not appear in the source.
|
|
This provides linking, appropriate coloring and appropriate hovering
in coqdoc documents.
In particular, this fixes #7697.
|
|
|
|
No change of semantics.
|
|
|
|
|
|
Co-authored-by: Jasper Hugunin <jasper@hugunin.net>
Co-authored-by: Gaëtan Gilbert <gaetan.gilbert@skyskimmer.net>
|
|
Four types of numerals are introduced:
- positive natural numbers (may include "_", e.g. to separate thousands, and leading 0)
- integer numbers (may start with a minus sign)
- positive numbers with mantisse and signed exponent
- signed numbers with mantisse and signed exponent
In passing, we clarify that the lexer parses only positive numerals,
but the numeral interpreters may accept signed numerals.
Several improvements and fixes come from Pierre Roux. See
https://github.com/coq/coq/pull/11703 for details. Thanks to him.
|
|
Add headers to a few files which were missing them.
|
|
On the principle that a notation to a constant inherits the implicit
arguments of the constant, a non-applied notation should inherit its
next maximal implicit arguments.
|
|
|
|
|
|
Exception: when the notation is to some @id.
Formerly, this was ignored for all kinds of string notations.
|
|
This is a change of semantics.
|
|
This was done for abbreviations but not string notations. This adopts
the policy proposed in #11091 to make parsing and printing consistent.
|
|
We do two changes:
- We distinguish between a notion of format generically attached to
notations and a new notion of format attached to interpreted
notations. "Reserved Notation" attaches a format
generically. "Notation" attaches the format specifically to the given
interpretation, and additionally, attaches it generically if it is the
first time the notation is defined.
- We warn before overriding an explicitly reserved generic format, or
a specific format.
|
|
The standard use is to repeat the option keywords in lowercase, which
is basically useless.
En passant add doc entry for Dump Arith.
|
|
- Implicit arguments in the return clause and in the branches
of a match were not checked.
- Implicit arguments in each declaration of intern_context were not
restarted.
- Additionally, in intern_context, we take into account ids from the
env on top of which intern_context is called.
- A better approximation of the check for manual implicit in notations
improved (though not fully correct because the exact context of
interpretation of a binder in a notation with recursive binders is
not known).
We also rename impl_binder_index into binder_block_names in anticipation of
checking redundancies also for non-implicit arguments.
|
|
WithoutTypeConstraint says it can be a term. In particular, if it has
manual implicit arguments, these will be allowed only in leading
lambdas. I.e. it can start with "fun {x}" but not with "forall {x}".
New constructor UnknownIfTermOrType allows to be both a term or a
type. In particular, if it allowed start both with "fun {x}" and
"forall {x}".
|
|
+ tests in the test-suite for non max local implicit arguments
|
|
|
|
|
|
- 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.
|
|
This moves the encoding of "n" as "arg_n" closer to the user interface level.
Note however that Constrintern.build_impl is not able yet to use ExplByPos.
See further commits.
|
|
`Fixpoint` with a let binder
Reviewed-by: Zimmi48
Reviewed-by: ejgallego
|
|
We also remove trailing whitespace.
Script used:
```bash
for i in `find . -name '*.ml' -or -name '*.mli' -or -name '*.mlg'`; do expand -i "$i" | sponge "$i"; sed -e's/[[:space:]]*$//' -i.bak "$i"; done
```
|
|
|
|
We move `binder_kind` to the pretyping AST, removing the last data
type in the now orphaned file `Decl_kinds`.
This seems a better fit, as this data is not relevant to the lower
layers but only used in `Impargs`.
We also move state keeping to `Impargs`, so now implicit declaration
must include the type. We also remove a duplicated function.
|
|
Not pretty, but it had to be done some day, as `Globnames` seems to be
on the way out.
I have taken the opportunity to reduce the number of `open` in the
codebase.
The qualified style would indeed allow us to use a bit nicer names
`GlobRef.Inductive` instead of `IndRef`, etc... once we have the
tooling to do large-scale refactoring that could be tried.
|
|
|
|
Not sure what this was for.
|
|
local fix/cofix (#10197)
Reviewed-by: maximedenes
|
|
|
|
Reviewed-by: herbelin
|
|
|
|
|
|
This replaces the mismatched context error, which occurred when
Instance := {} was used with strictly more fields than declared.
Since we later check that field names match those declared for the
instance, now that we reject duplicates we know that there are no
extra fields.
|
|
After removing `Instance : !type` implicit_application is only used in
constrintern. We propagate constant arguments ?allow_partial and
combine_params_freevar.
Also remove unused functions.
|
|
Conversely, Type existential variables now (explicitly) cover the Set
case.
Similarly for Prop and SProp.
|
|
We consistently use:
- UUnknown: to mean a rigid anonymous universe
(written Type in instances and Type as a sort)
[was formerly encoded as [] in Type's argument]
- UAnonymous: to mean a flexible anonymous universe
(written _ in instances and Type@{_} as a sort)
[was formerly encoded as [None] in Type's argument]
- UNamed: to mean a named universe or universe expression
(written id or qualid in instances and Type@{id} or Type@{qualid} or more
generally Type@{id+n}, Type@{qualid+n}, Type@{max(...)} as a sort)
There is a little change of syntax: "_" in a "max" list of universes
(e.g. "Type@{max(_,id+1)}" is not anymore allowed. But it was
trivially satisfiable by unifying the flexible universe with a
neighbor of the list and the syntax is anyway not documented.
There is a little change of semantics: if I do id@{Type} for an
abbreviation "id := Type", it will consider a rigid variable rather
than a flexible variable as before.
|
|
* use mixfix `(p1 | … | pn)` notation for nested disjunctive patterns,
rather than infix `|`, making pattern syntax consistent with term
syntax.
* disable extending `pattern` grammar with notation incompatible with
the nested disjunctive pattern syntax `(p1 | … | pn)`, such as the `(p
| q)` divisibility notation used by `Numbers`.
* emit a (disabled by default) `disj-pattern-notation` warning when such
`Notation` is attempted.
* update documentation accordingly; document incompatibilities in
`changelog`.
* comment special treatment of `(num)` in grammar.
* update file extensions in `Pcoq` header comment.
* correct the keyword declarations to reflect the contents of the
grammar files; perhaps there should be an option to disable implicit
keyword extension in a `.mlg` file, so that these lists could actually
be checked.
* parse the `|}` manifest record terminator as `|` followed by `}`,
eliminating the `|}` token which conflicts with notations that use `|`
as a terminator (such as, absolute value, norm, or cardinal in
MathComp). Since `|` is now an `operconstr` _and_ `pattern` terminator,
`bar_cbrace` rule checks for contiguous symbols, this change entails no
visible behaviour change.
|
|
|
