Re: [Cocci] Parsing C source files by Coccinelle
SF Markus Elfring (2016/03/08 10:20 +0100): > Would the Coccinelle software benefit from the tool "Menhir" if it will > take care for the parsing of C source files, too? From a performance point of view it is not sure there would be any gain. But obviously it would make more sense to use only one parser generator rather than two. However a bug in Menhir currently prevents its usage to generate the C parser. Once that bug will be fixed then using Menhir to generate the C parser will become an option. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Coccinelle Compilation Problem: configure file
Dear Aftab, Many thanks for your interest in Coccinelle. > Hi Julia Lawall and the Coccinelle Team, > > I really appreciate your response. I was able to generate the configure > file using *>./autogen*, > and all the installation steps. Thank you so much. Unfortunately, now I'm > unable to execute spatch. > > On executing this: > > *spatch -sp_file demos/simple.cocci demos/simple.c -o /tmp/new_simple.c* > > I get "spatch is not installed". Odd. What does which spatch say? Can you look at the file it points to? This file should be a shell script that tries to find the real spatch executable. If you are a bit familiar with shell scripts, perhaps you can go through this file and see whether the executables it tries to look for are there? > Also when I execute the following to test spatch, > > > *COCCINELLE_HOME=/home/aftab/workspace/coccinelle-master ./scripts/spatch > -sp_file demos/simple.cocci demos/simple.c -o /tmp/new_simple.c* > I get, > > *bash: ./scripts/spatch: No such file or directory* You can try to replace ./scripts/spatch by ./spatch in this command, I think it should work better. > I would like to let you know that I performed all the installation steps > (make, > make install), and I noticed no errors. > However, I am sceptical about why "make install" is carrying out the > following operations, > > > > *rm -f /usr/local/lib/coccinelle/spatchrm -f > /usr/local/lib/coccinelle/spatch.optrm -f /usr/local/bin/pycocci* > > I wonder whether this is causing the problem. I would be most glad if you > could please help me install spatch. I don't think these commands can cause a problem: they are supposed to make sure there is nothing left from previous installs. Feel free to send us the output of make install to make sure there is no problem there. > > Just in case, I installed the latest versions of OCaml and Menhir using > OPAM. Okay. FYI, Coccinelle is also packaged for opam. It is not the latest version but depending on what you intend to do it might be good enough, at least for a start. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Parsing C source files by Coccinelle
Hi, SF Markus Elfring (2016/03/07 19:39 +0100): > > I am interested in exploring how Coccinelle is parsing C programs, > > The software library "http://cristal.inria.fr/~fpottier/menhir/; > is involved to some degree. No, not for parsing C. Currently, Menhir (an advanced alternative to ocamlyacc) is used only to generate the SmPL parser, not the C parser. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] The Coccinelle repository on GitHub
Hi, SF Markus Elfring (2016/02/05 22:11 +0100): > > The repository on GitHub is now a mirror of the main developemnt repository. > > This evolution is a nice improvement. - Thank you. > > I am very curious on how our constructive dialogue will evolve further. > > Is it correct that no additional (topic) branches are provided so far? Yes. The team does not use such branches to collaborate, at a team level. > Are there any plans to adjust the affected software development process > a bit more? Not that I know. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] The Coccinelle repository on GitHub
SF Markus Elfring (2016/02/05 22:42 +0100): > >> Are there any plans to adjust the affected software development process > >> a bit more? > > > > Not that I know. > > Will any special branches be introduced in the near future? Not that I know. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Request 339055 commented by jengelh (submit devel:tools/coccinelle)
Hi, Sorry for the delayed response. Luis R. Rodriguez (2015/10/22 15:18 +0200): > On Thu, Oct 22, 2015 at 09:51:17AM +0200, Jan Engelhardt wrote: > > > > On Thursday 2015-10-22 07:23, SF Markus Elfring wrote: > > > > > >> Since coccinelle.spec does not invoke autoreconf > > > > > >How does the script > > >"https://github.com/coccinelle/coccinelle/blob/a46bef70162d17cec6b0fc6101d737989f735ee4/autogen; > > >fit to your view? > > > > 1. Running `aclocal; autoconf` is not enough. I spot a Makefile.am > > in the source tree, so you more or less need `autoreconf -fi` > > instead in the "autogen" script. Well, the Makefile.am is indeed present but not actually used so are you really sure the change you suggest is required? Is there any specific problem you are trying to solve with the current code? > > 2. My point was that autogen / aclocal / autoconf / etc. > > only needs to be run if there is no "configure" script present. > > Because there is a configure script in the released tarballs > > (at least there was so far), there is no need for coccinelle.spec to run > > autogen. And if autogen/aclocal/autoconf/etc. is not run, we do not need to > > BuildRequire it. > > The latest tarballs do not require it but the next releases will, so configure > will not be carried, so this was more of a heads up note / pro-active > patch. Well, so far my idea was to not provide configure in a public coccinelle repository but to provide it in tarballs, following an approach which I believe is common to many open-source projects, i.e. not version-controlling any generated file but still distribute them in tarballs. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] version 1.0.3 released
Julia Lawall (2015/10/28 08:33 +0100): > spgen should have a man page too. Not sure what is the status of that, > though. There is one in docs/spgen.1. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] version 1.0.3 released
Hello, Julia Lawall (2015/10/26 22:12 +0100): > * The dependency on ocamlp4 is removed, except when support for python is > included. Read camlp4 rather than ocamlp4. To be more precise, Coccinelle relies on a library called pycaml to support Python scripts. camlp4 is required only to compile this library. So if the library has been installed independently, e.g. through the distribution's packaging system (as can be done on Debian since pycaml is packaed), then camlp4 is not required. If the library has not been installed and Python scripting support is requested, then Coccinelle's build systemwill attempt to compile the bundled version of the camlp4 library, which will require camlp4. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] version 1.0.3 released
Dear Richard, Many thanks for your report! In coccinelle's source directory, could you please try the following: make distclean ./autogen ./configure ... hth, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Compiling Coccinelle with specific OCaml versions on openSUSE
Julia Lawall (2015/10/14 16:49 +0200): > No problem with opam, but this problem will recur, so I would really > prefer to have it solved. Sure! The two things are totally distinct and the reported problem definitely needs to be solved. So far, the issue is that I was absolutely not able to reproduce it, so any help on this would be greatly appreciated. The help could be a detailed procedure explaining how to reproduce the problem, an access to a machine where the problem occurs or anything else you guys could think about. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Compiling Coccinelle with specific OCaml versions on openSUSE
SF Markus Elfring (2015/10/14 12:32 +0200): > > Error: The files /usr/lib64/ocaml/obj.cmi > >and /usr/lib64/ocaml/menhirLib/menhirLib.cmi > >make inconsistent assumptions over interface Obj > > Makefile:90: recipe for target 'parser_cocci_menhir.cmx' failed > > I stumbled on the following message since I installed > the package "OCaml 4.02.3-1.1" also on my openSUSE system. > > > Error: Files type_cocci.cmx and menhirLib.cmx >make inconsistent assumptions over implementation Printf > Makefile:89: die Regel für Ziel „cocci_parser.cmxa“ scheiterte > > > How do you think about to circumvent this software dependency hiccup > by the OPAM interface for a while? > https://opam.ocaml.org/packages/coccinelle/coccinelle.1.0.2/ Sure, in my opinion, opam is a great solution worth trying for everybody. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Compiling latest coccinelle with opensuse
Luis, As the output of your rpm command shows, it seems your system has a mixture of packages installed, some from ocaml 4.02.1, others from ocaml 4.02.3. So, if completely uninstalling menhir, making sure all its iles have been removed from your system and then recompiling it is notenough, one other thing that may be worse trying is to also uninstall every OCaml related packages and then re-installing them. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Compiling latest coccinelle with opensuse
Hi, Luis: here is what I am guessing. At some point in the past you have installed OCaml and then compiled Menhir from sources with that version of OCaml. Meanwhile the OCaml related packages got updated but Menhir has not been recompiled with the updated version of OCaml, hence the inconsistency. My suggestion would thus be to make sure you recompile Menhir with the currently installed version of OCaml. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Compiling latest coccinelle with opensuse
Julia Lawall (2015/10/14 23:48 +0200): > I have the impression that he managed to have menhirLib without installing > menhir when he used the github version. Is that possible? Perhaps... > Is there a bundle for menhirLib? As far as I understand, the menhir bundle provides both the tool and MenhirLib. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Compiling latest coccinelle with opensuse
Dear Luis, Great you could finally compile coccinelle. And many thanks for having explained all the procedure to get docker working, it's really great and will be very helpful. I won't be able to look into the two problems (menhir and Common.Union_set) before next week but will do it as soonas possible. Regarding your suggestion to distribute the files generated by Menhir: I don't think this will solve the problem; Indeed, the problem does not occur when generating these files but when compiling them. Actually, there are two things that are used from Menhir. First, Menhir is used as a commandline tool. It takes a .mly file as inputandproduces corresponding .ml and .mli files. Thisis roughtly speaking equivalentto what yacc or bison do: taking a .y file as input and producing a .c as output. Second, the .ml file that has been generated by Menhir uses a library whichis called MenhirLib and to which the .cmo or .cmx files obtained by compilin the .ml file will have to be linked. That is where the problem you described appears andthat's why I don't think including the generated .ml files woudlhelpin any way. I hope I managed to be clear. :) Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Common.union_setError (was: Compiling latest coccinelle with opensuse)
Julia Lawall (2015/10/14 23:47 +0200): > OK, menhirLib is always needed with Coccinelle. Maybe Coccinelle manages > to use its own version, if you don't have one installed. I had thought > that Sebastien said that the menhir bundle was not being activated, but > perhaps I misunderstood. I'm myself confused, sorry about that Julia. There definitely is an issue in Coccinelle's uild system regading this. What I noiced (using INria's tree) is that Coccinelle is not able to use its bundled menhir. So if no Menhir is installed, things go wrong. I don't remember exactly what goes wrong but there's deinitely something that needs to be fixed here. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
[Cocci] Can we stop supporting OCaml 3.11?
Dear all, At the moment, the oldest version of OCaml which can be used to compile Coccinelle is 3.11. However, it turns out that OCaml 3.11.2 (released in January 2010) has a bug that pevents from compiling the optimized version o Coccinelle. We would like to drop our support of OCaml 3.11 and start with OCaml 3.12.0 that has been released in August 2010. Is anybody still using OCaml 3.11? Thanks a lot for your reponse, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] [PATCH 1/2] autotools: unify the way to determine version information
Hi, Nikolay Orlyuk (2015/07/23 07:18 +0300): Do we still need VERSION=$(shell ./version.sh | tr -d '\n') ? I'm pretty sure that VERSION=$(shell ./version.sh) will be enough. Same for configure.ac. Moreover why script version.sh can't be simply? #!/bin/sh tr -d '\n' ./version Other option would be to introduce subst. variable in configure.ac and use it in configurable Makefile. This approach might be more widespread. Yes, I believe this second approach would be better. Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] version 1.0.0
Dear Markus, SF Markus Elfring (2015/04/20 10:25 +0200): Comments and bug reports are welcome. Thanks for your update on development progress. Does this software version contain also any improvements for the documentation which were occasionally missed in previous releases? I'm not sure there has been a lot done on documentation so far, but Julia will certainly correct this point if I'm wrong. However, improving the user documentation is part of what I am supposdd to do on Coccinelle dugring this year. That's why, I'd be interested in any feedback. In particular, if you could let me know which are the areas which, according to you, are not so well documented, that woudl be helpful. Best wishes, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] coccinelle for osx
Hi, Could you please let us know the error message you get when trying to install findlib? Thanks, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Build errors with coccinelle 1.0.0-rc24 on OpenBSD and Linux
Dear Douglas,
Thanks a lot for the report.
Doug Hogan (2015/02/08 13:49 -0800):
There are two build errors with this release. I think both are due to
assuming you have parmap installed rather than using the bundled
version.
As far as I know, the way it is supposed to work is rather the
opposite: the build system is supposed to unconditionnally use the
bundled version. It is also supposed to ignore an installed version,
which can of course be considered as not very optimal, but that's how
the situation is supposed to be currently.
I get the same errors on OpenBSD and Linux. I have built
and used coccinelle in the past on both systems so I think this is a
new problem with this release.
Indeed, as Julia mentionned, Parmap was not used befre so that's indeed
a new problem.
For OpenBSD, I'm running OpenBSD -current with OCaml 4.02.1 (ports).
For Linux, I'm running Debian testing with OCaml 4.01.0.
I'm also using Debian testing, just re-downloaded the archive of
1.0-rc24 rom Coccinlle's web site, unpacked, ran ./configure and make,
but onmy system it worked,so I'm unable o reproduce the problem so far.
Here's the build failure. This was on OpenBSD, but I get the same build
failure on Linux. This was without any changes other than running
./configure:
gmake[6]: Entering directory
'/home/build/tmp/coccinelle-1.0.0-rc24/bundles/parmap/parmap-1.0-rc5-patched'
...
ocamlc -c bytearray_stubs.c
ocamlc -c -cc gcc -D_GNU_SOURCE -o setcore_stubs.o -fPIC setcore_stubs.c
setcore_stubs.c: In function 'setcore':
setcore_stubs.c:55: error: expected expression before 'else'
Makefile:54: recipe for target 'setcore_stubs.o' failed
gmake[6]: *** [setcore_stubs.o] Error 2
gmake[6]: Leaving directory
'/home/build/tmp/coccinelle-1.0.0-rc24/bundles/parmap/parmap-1.0-rc5-patched'
...
Parmap has a bug that prevents it from building on any OS that doesn't
have mach/threads_policy.h which includes OpenBSD and Linux. Here's
the syntax fix:
---
coccinelle-1.0.0-rc24.orig/bundles/parmap/parmap-1.0-rc5-patched/setcore_stubs.c
Wed Jan 7 16:58:46 2015
+++
coccinelle-1.0.0-rc24/bundles/parmap/parmap-1.0-rc5-patched/setcore_stubs.c
Sun Feb 8 13:15:29 2015
@@ -51,8 +51,8 @@ CAMLprim value setcore(value which) {
fprintf(stderr,MAC OS X: Failed pinning to cpu %d, trying
%d/2\n,w, w);
w=w/2;
}
-#endif
else
+#endif
{ //fprintf(stderr,Succeeded pinning to cpu %d\n,w);
finished=1;
}
I indeed don't have this mc/thread_policy.h and it indeed seems it is
not present in any Debian package. So at the moment I don't understand
why things work here and not for you.
It may also depend on how HAVE_DECL_SCHED_SETAFFINITY is handled, since
that block ends with an if statemen. So presumably, that block is
included on you system and not on mine and may be why you get the error
and I don't.
Even with this fix, it fails to build on both OSes. It's the same
failure with both OpenBSD and Linux (Unbound module Parmap).
gmake[4]: Leaving directory
'/home/build/tmp/coccinelle-1.0.0-rc24/bundles/parmap'
...
/usr/local/bin/ocamlopt.opt -unsafe -cclib -lparmap_stubs -I
/home/build/tmp/coccinelle-1.0.0-rc24/bundles/menhirLib/ -I
/home/build/tmp/coccinelle-1.0.0-rc24/bundles/pycaml/ -I /usr/local/lib/ocaml
-I /usr/local/lib/ocaml/pcre -I bundles/parmap/parmap-1.0-rc5-patched/_build
-I commons -I commons/ocamlextra -I ctl -I engine -I extra -I globals -I
ocaml -I parsing_c -I parsing_cocci -I popl09 -I python -c main.ml
File ./main.ml, line 956, characters 35-60:
Error: Unbound module Parmap
Makefile:660: recipe for target 'main.cmx' failed
gmake[3]: *** [main.cmx] Error 2
gmake[3]: Leaving directory '/home/build/tmp/coccinelle-1.0.0-rc24'
...
On Debian, if I install libparmap-ocaml and libparmap-ocaml-dev, then
I can build coccinelle. I presume that's because it doesn't use this
bundled version and the system version is picked up in a different
manner than the bundled version.
It may be more subtle, e.g. it uses some files of the installed version
and some files of the bundled version.
Sébastien.
___
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Possible to install on OSX?
Hi, I still have to be able to compile with python support. But coccinelle itself is working great. Good. I’ll try to follow your previous indications on that to solve it myself. Sure. A quick google suggests that port search python may help, perhapswith a | grep dev I will get back to you after some time If I am not able to do it. ;-) Okay, thanks. Actually I'd be interested in feed bakc if youmanage tomake things work, too! :) Thanks very much for your patience, kindness and supportiveness. My pleasure! Best wishes, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Possible to install on OSX?
Eliseo Martínez (2015/01/20 15:22 +0100): Here it is. I’ve renamed it to depend.txt in order to attach it without problems. Thanks. It seems the sed script does still not work properly on your system. The commons/.depend file you have sent is generated by the following command: /opt/local/bin/ocamldep -I ocamlextra -I ocollection common.mli oarray.mli oassoc.mli objet.mli ocollection.mli ograph_extended.mli ograph.mli ograph_simple.mli osequence.mli oset.mli parser_combinators.mli commands.ml common.ml common_extra.ml interfaces.ml objet.ml ocollection.ml seti.ml oset.ml oassoc.ml osequence.ml ograph.ml ocollection/oseti.ml ocollection/oseth.ml ocollection/osetb.ml ocollection/osetpt.ml ocollection/oassocb.ml ocollection/oassoch.ml ocollection/oassoc_buffer.ml ocollection/oassoc_cache.ml ocollection/oassocid.ml oarray.ml ocollection/ograph2way.ml ograph_simple.ml ograph_extended.ml glimpse.ml parser_combinators.ml | sed s/bytes\.cm[a-z]\+//g .depend Do you know sed enough to play a bit with this command in a terminal and see whether youcan make it produce a .depend where there is no occurence of bytes.cm something? A good test is to replace the .depend in the command by | grep bytes. If the sed bit works the modified command should display nothing at all. Many thanks for your patience and cooperation! Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Possible to install on OSX?
Hi again, There is no commons/.depend in the clean checkout state I start from. Oh, okay. Just in case, I run it again, checking there was no commons/.depend. Same result. Logs attached. Okay. After you have run make there should be a commons/.depend, though. Could you please send it? Thanks! Sébastien. On 20 Jan 2015, at 14:22, Sébastien Hinderer sebastien.hinde...@inria.fr wrote: Hi, I’ve removed gnused, and applied this last patch on top of the previous one you sent yesterday. Still the error about wrong assumptions in bytes interface. Can you please do rm commons/.depend and run make again? Thanks, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Possible to install on OSX?
Dear Eliseo, Many thanks for the information. Is it correct that the version of sed you are using is not GNU sed? In that case, would it be possible for you to try installing gnu sed, please? Once it's done and you are sure that simply running sed runs gnu SED, can you please make a make distclean ./configure make and see whether that works? Thanks, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
Re: [Cocci] Error: Unbound value Common.union_set
Dear Emil, I'm an engineer working with Julia on Coccinelle. Many thanks to you for having reported this issue. It is due to a change in the OCaml compiler between versions 4.01 and 4.02. We will do our best to fix Coccinelle as soon as possible so that it also works with OCaml 4.02, but meanwhile, you can always make it work with an older version of OCaml such as 4.01. Thismeans to recompile coccinelle e.g.with OCaml 4.01. Hope this helps, Sébastien. ___ Cocci mailing list Cocci@systeme.lip6.fr https://systeme.lip6.fr/mailman/listinfo/cocci
