| Age | Commit message (Collapse) | Author |
|---|
|
Closed in 4a8950ec7a0d9f2b216e67e69b446c064590a8e9
|
|
Closed in 69c4d0fd7b8325187e8da697a9c283594047d.
I used [Timeout 2] to distinguish between stack overflow and immediate
return.
|
|
Closed in f65fa9de8a4c9c12d933188a755b51508bd51921
I used [Timeout 2 Fail] to test the difference between immediate failure
and stack overflow. Hopefully this is robust enough.
|
|
It was fixed in c3feef4ed5dec126f1144dec91eee9c0f0522a94.
The test case uses [Timeout 2] to test for "Coq runs instantaneously
rather than running out of memory". Hopefully this is robust enough.
|
|
|
|
I have a fix, I'm reviewing it because there may be other bugs around.
|
|
with -R. (Fix for Rocq/Rational.)"
This reverts commit 7d3ce4012a53b123dac95381bf46aac65f865d69.
Conflicts:
CHANGES
|
|
|
|
|
|
|
|
|
|
non-ML applications. Control channel can be also ignored.
|
|
|
|
|
|
applications of I31 constructor as lazy.
|
|
|
|
This completes the port of the native compiler to retroknowledge.
However, some testing and optimizations are still to be done.
|
|
|
|
|
|
|
|
dependencies.
|
|
|
|
|
|
This option complements -I-as and -R. As the two other options, it adds a
new loadpath, but contrarily to them, files are not looked into the
directory unless fully qualified.
|
|
Inside a section, Let followed by a Proof. Qed. are handled as regular
definitions, hence they have universe constraints coming from the type
and from the body. Only the former set was returned to libstack, but
both sets were added to the global universe graph. Hence, at section
closing time, an incomplete set of universe constraints was added back
to the global graph (End Section replays the libstack) and hence saved
in the .vo file. coqchk was right in reporting missing constraints.
|
|
|
|
(Fix for Rocq/Rational.)
|
|
|
|
They were just passed along in the tactics.
|
|
|
|
|
|
the current state without having to use both get, bind and set.
|
|
|
|
- Option -I no longer handles loadpath, only mlpath. This is the same
behavior as coq_makefile. Option -I-as is unchanged.
- Option -R no longer recursively adds to mlpath; only the root directory
is added.
- user-contrib/ and xdg directories are no longer recursively added to
the loadpath.
- theories/ and plugins/ are no longer recursively added to the loadpath
when option -nois is passed.
- All the preconfigured directories are still recursively added to the
mlpath though.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
when defining cache hash tables in Closure. Why it was working in 3.12
is a mystery to me.
|
|
refinement of evars (in filter_candidates).
Incidentally introduced a copy of type "option", "update", which
highlights the specific intention of "updating" or not.
|
|
|
|
test-suite. (Fix for bug #3024)
|
|
I added a .merlin in ide/ who inherits everything from the root .merlin and also
adds the dependency to lablgtk, which I removed from the root file. These way
people not working on that part of the code won't get bothered if they don't
have that package.
I removed the S/B entry for plugins which was useless, indeed there is no ML
file in that directory and merlin doesn't scan the subdirectories recursively
(as you know).
I also removed the S/B entry for checker since most of the files of this
directory are also present in kernel and that was the cause of a lot errors on
merlin's side (think "inconsistent assumptions").
On top of that, no part of the tree depends on checker (I back that assertion by
a grep of the *.d files of the tree) so these lines in the .merlin were actually
useless. The only part of the tree where you need to know what's in checker/ is
when you are working in checker/ itself, but since merlin automatically adds the
directory of the file under edition in its source and load paths nothing else is
needed.
There might still be problems after this patch, but they should be less of them.
Considering my poor knowledge of the codebase there might be other conflicts I
have missed.
|
|
|
