| Age | Commit message (Collapse) | Author |
|---|
|
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.
|
|
`internal_ghost` was an artifact to ease porting of the ml4 rules. Now
that the location is optional we can finally get rid of it.
|
|
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.
|
|
|
|
|
|
Today, both modes are controlled by a single flag, however this is a
bit misleading as is_silent really means "quiet", that is to say `coqc
-q` whereas "verbose" is Coq normal operation.
We also restore proper behavior of goal printing in coqtop on quiet
mode, thanks to @Matafou for the report.
|
|
record fields.
|
|
There were actually no syntax for it, and I'm still unsure what good
syntax to give to it, even more that it would be useful to have one.
|
|
We remove the camlp4 compatibility layer, and try to clean up
most structures. `parsing/compat` is gone.
We added some documentation to the lexer/parser interfaces that are
often obscured by module includes.
|
|
|
|
This is the good parts of PR #360.
IIUC, these vernacular were meant mostly for debugging and they are
not supposed to be of any use these days.
Back and join are still there not to break the testing infrastructure,
but some day they should go away.
|
|
|
|
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.
|
|
|
|
|
|
|
|
Note: This reveals a little bug yet to fix in g_vernac.ml4. In
Definition f '((x,y):id nat * id nat) '((x',y'):id nat * id nat) := Eval unfold id in x+y = x'+y'.
the "id" are wrongly unfolded and in
Definition f '(x,y) '(x',y') := x+y = x'+y' : Prop.
an unexpected cast remains in the body of f.
|
|
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.
|
|
|
|
For legacy reasons, pretty printing required to provide a "tag"
interpretation function `pp_tag`. However such function was not of much
use as the backends (richpp and terminal) hooked at the `Format.tag`
level.
We thus remove this unused indirection layer and annotate expressions
with their `Format` tags.
This is a step towards moving the last bit of terminal code out of the
core system.
|
|
This is what has always been used, so it doesn't represent a functional
change.
This is just a preliminary patch, but many more possibilities could be
done wrt tags.
|
|
No functional change.
|
|
|
|
|
|
Was a bug introduced in 0ad6edc1.
|
|
|
|
We introduce a bit of compatibility parsing code to print deprecation
warnings.
|
|
|
|
|
|
This reverts commit b00e039b957b8428c21faec5c76f3a3484cde2cf, reversing
changes made to ca9e00ff9b2a8ee17430398a5e0bef2345c39341.
It turns out that calling from fake_ide the STM commands that were removed
by this PR requires an extension of the XML protocol. So postponing the
integration.
|
|
I think these commands never make a lot of sense on scripts other than
debugging and we have better methods now.
The last remaining command, used for the tty emulation has been renamed
to VtBack, but it should go away at some point too once the legacy
interfaces are removed.
|
|
Was PR#339: Documenting type class options, typeclasses eauto
|
|
This was for the translator and is not relevant for the beautifier.
|
|
Was PR#340: Fix various shortcomings of the warnings infrastructure.
|
|
In addition to a priority, cleanup the interfaces for passing this
information as well. The pattern, if given, takes priority over the
inferred one.
We only allow Existing Instances gr ... gr | pri. for now, without
pattern, as before.
Make the API compatible to 8.5 as well.
|
|
With multiple arguments list, repeating the "/" modifier used to be
mandatory. So instead of forbidding it, we issue a deprecation warning.
|
|
- The flags are now interpreted from left to right, without any other
precedence rule. The previous one did not make much sense in interactive
mode.
- Set Warnings and Set Warnings Append are now synonyms, and have the
"append" semantics, which is the most natural one for warnings.
- Warnings on unknown warnings are now printed only once (previously the
would be repeated on further calls to Set Warnings, sections closing,
module requiring...).
- Warning status strings are normalized, so that e.g. "+foo,-foo" is reduced
to "-foo" (if foo exists, "" otherwise).
|
|
|
|
Was PR#319: More error tagging, try to fix bug 5135
|
|
Was PR#337: Fix arguments
|
|
The main point of this change is to fix #3035: Avoiding trailing
arguments in the Arguments command, and related issues occurring in
HoTT for instance. When the "assert" flag is not specified, we now
accept prefixes of the list of arguments.
The semantics of _ w.r.t. to renaming has changed. Previously, it meant
"restore the original name inferred from the type". Now it means "don't
change the current name".
The syntax of arguments is now restricted. Modifiers like /, ! and
scopes are allowed only in the first arguments list.
We also add *a lot* of missing checks on input values and fix various
bugs.
Note that this code is still way too complex for what it does, due to
the complexity of the implicit arguments, reduction behaviors and renaming
APIs.
|
|
|
|
+ a few improvements on 5f1dd4c40 (lexing of { and }).
|
