| Commit message (Collapse) | Author | Age |
|
|
|
| |
Should fix #6177, which was triggered by lonely .ml files.
|
|
|
|
|
| |
Julien Narboux confirmed that it was dead code (GeoProof is not to be
confused with GeoCoq).
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
We use the win32 specific function only if WIN32 is defined
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16055 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Threads were only there to handle blocking dialogs with the different
coqtops. But programming with threads have drawbacks : complex mutex
infrastructure, possible deadlocks, etc. In particular gtk functions
are not meant to be called from a thread which isn't the gtk main loop,
(unless some gtk mutex have been taken). This seem to pose problem
specifically in win32 (and macosx ?), hence the use of the
GtkThread.(a)sync hack for scheduling code for execution in the gtk
main loop.
Instead, we now use the Glib.Io module to install a callback that will
be runned when some answer of coqtop is available on the channel.
This implies using now a continuation-passing style: for instance,
instead of two sequential requests to coqtop, we'll now have the
2nd request inside the callback handling the answer to the 1st request.
Remarks:
- Also use asynchronous i/o for external commands (editor, coqc, make...).
Launching an external editor or browser won't freeze coqide anymore.
- Reworked handling of coqtop process, especially when closing them.
A responsive coqtop should now hara-kiri immediatly when its input
channel is closed. Otherwise we try later a soft kill, then some
hard kills if necessary. If nothing work we warns the user.
When quitting coqide, all this might induce a small delay (2s at worse).
- Be careful now to avoid "long" computations (or blocking i/o) in
a coqide function. Experimentally, it seems that loading/saving a .v
file is quick enough. If necessary, we could use asynchronous i/o
also for writing the .v, but for loading I've no clue.
- In the Coqide module, we ensure that the current continuation k
will indeed be run at the end thanks to an abstract return type
(void = opaque copy of unit).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16049 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We now try harder to handle ourselves gtk messages (e.g. Gtk-WARNING ...).
This way, we could reroute them nicely in w32, and pop-up the critical ones.
Moreover, the code rerouting debug messages to a log file in w32 was using
!Ideutils.debug before its initialization. Now, when a log file is used, its
name is displayed in the about messages.
Btw, some code cleaning in coqide_main
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16037 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
| |
Ctrl-C now triggers a clean (interactive) quit.
The other catchable signals still trigger an emergency (non-interactive) quit.
Btw, stop trying to relaunch the gtk loop in case of failure
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16036 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15922 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15814 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15715 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15354 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15352 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15325 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
| |
it. Now CoqIDE is not anymore totally irrespectful of the local configuration of themes, in particular w.r.t. to menu fonts.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15251 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
| |
CoqIDE to a segfault. Gtk was not initialized while it was trying to declare accel keys.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15216 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
| |
If it is AUTO then we keep the heuristic to change coqide by coqtop in Sys.executable_name.
If it fails coqtop location must be given by the users.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@15188 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
| |
Beware, it means that files position is not relative to coqtop position
but is given by XDG_DATA_DIRS and XDG_CONFIG_DIRS.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14822 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14715 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14695 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
| |
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14442 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
| |
to specify where to look the project file
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14441 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
| |
We can be easily substitute Gdk.Windowing by a glance of configure work...
This reverts commit 8b6f6b1c4b60e74dccd5d8c49bdd433e19d53bf4.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14208 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
| |
Because of lablgtk issues, accel_maps can't be customized well on MacOS
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14180 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We use GenerateConsoleCtrlEvent(CTRL_C_EVENT,...) after having
attached coqide to the console of the coqtop we want to interrupt.
Two caveats:
- This code isn't compatible with Windows < XP SP1.
- It relies on the fact that coqide is now a true GUI app, without
console by default. If for some reason the console of coqide is
restored (for instance via mkwinapp -unset), strange behavior
of the interrupt button is to be expected, at the very least all
instances of coqtop will get Ctrl-C instead of a precise one.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14077 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Now that coqide is a console-free win32 app, writing to
nonexistent stdout/stderr lead to Sys_error. To avoid that:
- We reroute coqide's stdout/stderr to either a pipe that will stay
unread (by default), or to a temp log file (in debug mode).
- When doing create_process, avoid referring to Unix.stderr:
anything printed by coqtop to its stderr will be merged in the
regular channel.
- On the coqtop side, remove the awkward fix consisting in a \r printed
on stderr apparently to fix coqide.byte. This fix is probably obsolete
since the separation of coqide and coqtop as distinct processes.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14047 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is implemented as a C external launching the TerminateProcess
of the Win32 API. This should be considered as quite experimental
(cf. the way we handle pid in the comment of ide_win32_stubs.c).
I don't know how to emulate an interrupt (Ctrl-C), for now the two
button "Restart" and "Interrupt" have the same semantics on win32
(kill the subprocess and start at top).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14044 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
| |
In Win32, these messages may trigger some Sys_error if we try to
turn coqide into a true windows app (no console). To be continued...
The best way would be probably to re-route the whole stdout and stderr
to something else via dup2, but to what ? A log file ?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14043 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is an adaptation of commit r13750 of branch 8.3
- coqlib is currently computed relatively of Sys.executable_name,
hence no need to set it manually
- in Win32, better detection of user home dir : in System.ml,
if HOME isn't set, we look now for HOMEDRIVE\HOMEPATH, and then
for USERPROFILE
- concerning PATH, in Win32 we now add coqbin (or the location of
coqide) to PATH during the initialization.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14041 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
used
This is an adaptation of commit r13751 of branch 8.3
Even if coqide.exe keeps its console by default for the moment,
we try to allow turning it off (for instance via the mkwinapp tool) :
for that we need to be cautious about the use of functions
like prerr_endline. Since stderr doesn't exists in this settings,
such functions trigger a Sys_error "bad file descriptor".
This patch protects many access to std** by a (try ... with _ ->()).
Nota: with camlp5 < 6.02.1, Print Grammar was also generating
such a Sys_error.
TODO: we should try to figure a way of displaying messages (both
debug and early/late error message). A log file ? A popup ?
diplay in the response buffer ?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@14040 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
| |
You can quit coqide from the dock and reboot/shutdown will ask you if
you want to save your unsavec files.
Asks for a re"configure".
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13974 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Run "coqide -coqtop someothercoqtop" if you want to use a toplevel
which isn't the one coming alongside coqide. To be documented,
to be improved (maybe an field in coqide's preferences ?).
coqide -h should display this kind of ide-specific option.
* Since we now use create_process instead of open_process, we don't
use /bin/sh, hence running Filename.quote on args was actually
wrong.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13932 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Avoid using Util which depends on Compat and hence Camlp4
- Instead, a small Minilib module specific to coqide, which
duplicate 5 functions from Util (50 lines)
- some dead code removal
- the coqlib variable is asked to coqtop
- remove obsolete Util.check_for_interrupt
This way, coqide only depends on 3 files outside ide/ :
Coq_config, Flags, Ide_intf. Makefile and ocamlbuild are adapted
accordingly.
TODO: how should we signal coqide error, warnings, etc ?
For the moment, some Printf.eprintf, some failwith.
To uniformize later...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13930 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- We now use create_process instead of open_process, and stores
the answered pid.
- The "Stop" button now sends a Sigint signal to this coqtop pid.
- The "Goto Start" button now works even if a computation is ongoing,
a new process is spawned and the previous one is killed -9, and
then a waitpid is done to avoid having zombies around.
Note that currently a vm_compute won't be stopped by a "Stop",
but only by a "Goto Start". This can be quite confusing. How to
properly document that "Goto Start" has the side effect of being
a kill ? Maybe we could check someday if the Ctrl-C has been
successful, and kill -9 if not ? Or maybe make coqide aware
of a flag "we_are_vm_computing" and then kill -9 instead of Ctrl-C
in this case ?
TODO:
- for the moment a forced "Goto Start" displays an unfriendly anomaly
- check if all this works under Windows (probably but not sure).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13926 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
if `pkg-config --exists ige-mac-integration`, coqide.opt will be
able to open files by double-clik in finder on Darwin.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13779 85f007b7-540e-0410-9357-904b9bb8a0f7
|