Source code static analysis (Linux source packages C/C++ autotools cmake make m4...)

[Dmitry Ponyatov]

Is opencog has some module for parsing and source code transforming and static analysis ?

I'd like to have some tool able to process typycal Linux source package, and extract info on dependencies between used function signatures, #defines, source/object files, and configure options (for embedded Linux build primarily)

[Linas Vepstas]

no.  why would you think it did?

On Sun, Jun 11, 2017 at 12:58 AM, Dmitry Ponyatov <dpon...@gmail.com> wrote:

Is opencog has some module for parsing and source code transforming and static analysis ?

I'd like to have some tool able to process typycal Linux source package, and extract info on dependencies between used function signatures, #defines, source/object files, and configure options (for embedded Linux build primarily)

--
You received this message because you are subscribed to the Google Groups "opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencog+unsubscribe@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/227a32a4-d995-4188-bf1f-9d80f049fc68%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Dmitry Ponyatov]

no.  why would you think it did?

Hypergraph rewrite model looks native for source code processing and analisys

[Linas Vepstas]

On Wed, Jun 21, 2017 at 11:31 AM, Dmitry Ponyatov <dpon...@gmail.com> wrote:

no.  why would you think it did?

Hypergraph rewrite model looks native for source code processing and analisys

That is true, but the people who need that already have it ... so there are graph rewrite engines in gcc, in llvm/clang, in guile, and I assume in java and in perl and probably many other languages and compilers.

I tried to lift some of the general generic ideas from compiler design, and stick them into atomese, but atomese is kind-of much too fat and too slow for use in any ordinary compiler or code analysis tool. Its really meant for probabilistic knowledge representation.

Now, maybe you could do a supercompiler in opencog, but ven then it would be more efficient to create some custom hand-tuned code for that.

--linas