| Age | Commit message (Collapse) | Author |
|---|
|
This commit was motivated by true spurious conversions arising in my
`to_constr` debug branch.
The changes here need careful review as the tradeoffs are subtle and
still a lot of clean up remains to be done in `vernac/*`.
We have opted for penalize [minimally] the few users coming from true
`Constr`-land, but I am sure we can tweak code in a much better way.
In particular, it is not clear if internalization should take an
`evar_map` even in the cases where it is not triggered, see the
changes under `plugins` for a good example.
Also, the new return type of `Pretyping.understand` should undergo
careful review.
We don't touch `Impargs` as it is not clear how to proceed, however,
the current type of `compute_implicits_gen` looks very suspicious as
it is called often with free evars.
Some TODOs are:
- impargs was calling whd_all, the Econstr equivalent can be either
+ Reductionops.whd_all [which does refolding and no sharing]
+ Reductionops.clos_whd_flags with all as a flag.
|
|
We follow the suggestions in #402 and turn uses of `Loc.located` in
`vernac` into `CAst.t`. The impact should be low as this change mostly
affects top-level vernaculars.
With this change, we are even closer to automatically map a text
document to its AST in a programmatic way.
|
|
The motivations are:
- To reflect the concrete syntax more closely.
- To factorize the different places where "contexts" are internalized:
before this patch, there is a different treatment of `Definition f
'(x,y) := x+y` and `Definition f := fun '(x,y) => x+y`, and a hack
to interpret `Definition f `pat := c : t`. With the patch, the fix
to avoid seeing a variable named `pat` works for both `fun 'x =>
...` and `Definition f 'x := ...`.
The drawbacks are:
- Counterpart to reflecting the concrete syntax more closerly, there
are more redundancies in the syntax. For instance, the case `CLetIn
(na,b,t,c)` can appears also in the form `CProdN (CLocalDef
(na,b,t)::rest,d)` and `CLambdaN (CLocalDef (na,b,t)::rest,d)`.
- Changes in the API, hence adaptation of plugins referring to `constr_expr` needed.
|
|
There is no way today to distinguish primitive projections from
compatibility constants, at least in the case of a record without
parameters.
We remedy to this by always using the r.(p) syntax when printing
primitive projections, even with Set Printing All.
The input syntax r.(p) is still elaborated to GApp, so that we can preserve
the compatibility layer. Hopefully we can make up a plan to get rid of that
layer, but it will require fixing a few problems.
|
|
|
|
This is useful as it allows to reflect program_mode behavior as an attribute.
|
|
Over the time, `Command` grew organically and it has become now one of
the most complex files in the codebase; however, its functionality is
well separated into 4 key components that have little to do with each
other.
We thus split the file, and also document the interfaces. Some parts
of `Command` export tricky internals to use by other plugins, and it
is common that plugin writers tend to get confused, so we are more
explicit about these parts now.
This patch depends on #6413.
|
|
We remove a lot of uses of `evar_map` ref in `vernac`, cleanup step
desirable to progress with EConstr there.
|
|
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'd like to handle proofs functionally we thus recommend not to use
printing functions without an explicit context.
We also adapt most of the code, making more explicit where the
printing environment is coming from.
An open task is to refactor some code so we gradually make the
`Pfedit.get_current_context ()` disappear.
|
|
The internal detype function takes an additional arguments dictating
whether it should be eager or lazy.
We introduce a new type of delayed `DAst.t` AST nodes and use it for
`glob_constr`.
Such type, instead of only containing a value, it can contain a lazy
computation too. We use a GADT to discriminate between both uses
statically, so that no delayed terms ever happen to be
marshalled (which would raise anomalies).
We also fix a regression in the test-suite:
Mixing laziness and effects is a well-known hell. Here, an exception
that was raised for mere control purpose was delayed and raised at a
later time as an anomaly. We make the offending function eager.
|
|
This was already the case, but the API was not exposing this.
|
|
The use of template polymorphism in constants was quite limited, as it
only applied to definitions that were exactly inductive types without any
parameter whatsoever. Furthermore, it seems that following the introduction
of polymorphic definitions, the code path enforced regular polymorphism as
soon as the type of a definition was given, which was in practice almost
always.
Removing this feature had no observable effect neither on the test-suite,
nor on any development that we monitor on Travis. I believe it is safe to
assume it was nowadays useless.
|
|
|
|
We aditionally return the abstract universe context inside the option. This
is relatively painless as most uses were using the option as a boolean.
|
|
|
|
|
|
Unluckily, this forces replacing a lot of code in plugins, because the API
defined the type of goals and tactics in Proof_type, and by the no-alias rule,
this was the only one. But Proof_type was already implicitly deprecated, so
that the API should have relied on Tacmach instead.
|
|
|
|
|
|
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.
|
|
|
|
a flag suspectingly renamed in a clearer way
|
|
As per https://github.com/coq/coq/pull/716#issuecomment-305140839
Partially using
```bash
git grep --name-only 'anomaly\s*\(~label:"[^"]*"\s*\)\?\(Pp.\)\?(\(\(Pp.\)\?str\)\?\s*".*[^\.!]")' | xargs sed s'/\(anomaly\s*\(~label:"[^"]*"\s*\)\?\(Pp.\)\?(\(\(Pp.\)\?str\)\?\s*".*\s*[^\.! ]\)\s*")/\1.")/g' -i
```
and
```bash
git grep --name-only ' !"' | xargs sed s'/ !"/!"/g' -i
```
The rest were manually edited by looking at the results of
```bash
git grep anomaly | grep '\.ml' | grep -v 'anomaly\s*\(~label:"[^"]*"\s*\)\?\(Pp\.\)\?(\(\(Pp.\)\?str\)\?\s*".*\(\.\|!\)")' | grep 'anomaly\($\|[^_]\)' | less
```
|
|
This highlights that this is a binary mode changing the interpretation
of "?x" rather than additionally allowing patvar.
|
|
This module collects the functions of Nameops which are about Name.t
and somehow standardize or improve their name, resulting in particular
from discussions in working group.
Note the use of a dedicated exception rather than a failwith for
Nameops.Name.out.
Drawback of the approach: one needs to open Nameops, or to use long
prefix Nameops.Name.
|
|
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.
|
|
|
|
Following @gasche idea, and the original intention of #402, we switch
the main parsing AST of Coq from `'a Loc.located` to `'a CAst.ast`
which is private and record-based.
This provides significantly clearer code for the AST, and is robust
wrt attributes.
|
|
This completes the Loc.ghost removal, the idea is to gear the API
towards optional, but uniform, location handling.
We don't print <unknown> anymore in the case there is no location.
This is what the test suite expects.
The old printing logic for located items was a bit inconsistent as
it sometimes printed <unknown> and other times it printed nothing as
the caller checked for `is_ghost` upstream.
|
|
Now it is a private field, locations are optional.
|
|
|
|
This is the second patch, which is a bit more invasive. We reasoning
is similar to the previous patch.
Code is not as clean as it could as we would need to convert
`glob_constr` to located too, then a few parts could just map the
location.
|
|
This is first of a series of patches, converting `constrexpr` pattern
data type from ad-hoc location handling to `Loc.located`.
Along Coq, we can find two different coding styles for handling
objects with location information: one style uses `'a Loc.located`,
whereas other data structures directly embed `Loc.t` in their
constructors.
Handling all located objects uniformly would be very convenient, and
would allow optimizing certain cases, in particular making located
smarter when there is no location information, as it is the case for
all terms coming from the kernel.
`git grep 'Loc.t \*'` gives an overview of the remaining work to do.
We've also added an experimental API for `located` to the `Loc`
module, `Loc.tag` should be used to add locations objects, making it
explicit in the code when a "located" object is created.
|
|
record fields.
|
|
|
|
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.
|
|
This is a patch fulfilling the relevant remark of Maxime that an
explicit information at the ML type level would be better than "cast
surgery" to carry the optional type of a let-in.
There are a very few semantic changes.
- a "(x:t:=c)" in a block of binders is now written in the more
standard way "(x:=c:t)"
- in notations, the type of a let-in is not displayed if not
explicitly asked so.
See discussion at PR #417 for more information.
|
|
RawLocal -> CLocal
|
|
This is a bit long, but it is to keep a symmetry with constr_expr.
|
|
|
|
Deprecating abstract_constr_expr in favor of mkCLambdaN,
prod_constr_expr in favor of mkCProdN.
Note: They did not do exactly the same, the first ones were
interpreting "(x y z:_)" as "(x:_) (y:_) (z:_)" while the second ones
were preserving the original sharing of the type, what I think is the
correct thing to do.
So, there is also a "fix" of semantic here.
|
|
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.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
