[cocci] How do you concatinate string constants in SmPL with python?

[cocci] How do you concatinate string constants in SmPL with python?

[Eric Wheeler]

Hello all,

I would like to do something like this:

@ r @
constant C;
@@
-perror(C);
+pr_err(C ": %s\n", strerror(errno));

Thus in python I need something like C = C + C ": %s\n".

I've tried a few things, but I don't think I'm anywhere close to getting 
it right.  Can someone help with how this should be setup?

Also is there a reference for the `coccinelle.` and `cocci.` python calls 
that seem to be available in a script:python section?

Thanks for your help!


--
Eric Wheeler

Re: [cocci] How do you concatinate string constants in SmPL with python?

[Julia Lawall]

On Sat, 2 Apr 2022, Eric Wheeler wrote:

> Hello all,
>
> I would like to do something like this:
>
> @ r @
> constant C;
> @@
> -perror(C);
> +pr_err(C ": %s\n", strerror(errno));
>
> Thus in python I need something like C = C + C ": %s\n".

I don't understand this.

>
> I've tried a few things, but I don't think I'm anywhere close to getting
> it right.  Can someone help with how this should be setup?
>
> Also is there a reference for the `coccinelle.` and `cocci.` python calls
> that seem to be available in a script:python section?

There is Coccilib.3cocci, but that is for ocaml.

My understanding is that there are reflective features in python that let
you find out what functions are available.  python/python_documentation.md
in the source tree also has some information.  Or look at the code in
python/coccilib.

julia

Re: [cocci] How do you concatinate string constants in SmPL with python?

[Eric Wheeler]

On Sat, 2 Apr 2022, Julia Lawall wrote:
> On Sat, 2 Apr 2022, Eric Wheeler wrote:
> > I would like to do something like this:
> >
> > @ r @
> > constant C;
> > @@
> > -perror(C);
> > +pr_err(C ": %s\n", strerror(errno));
> >
> > Thus in python I need something like C = C + C ": %s\n".
> 
> I don't understand this.

For example:

-perror("unable to open file");
+pr_err("unable to open file: %s\n", strerror(errno));

I'm trying to append ": %s\n" to the original string (C) so we can process 
the error using pr_err() instead of perror().  Since SmPL does not support 
string concatination (so far as I am aware), I would need to use python.

But I'm not sure how to structure the python bits so they interact with 
cocci.  I've looked at some examples but I don't see how to change the 
value of C to something else, such as C = C + ": %s\n" where + is the 
append operator in python.

--
Eric Wheeler


> 
> >
> > I've tried a few things, but I don't think I'm anywhere close to getting
> > it right.  Can someone help with how this should be setup?
> >
> > Also is there a reference for the `coccinelle.` and `cocci.` python calls
> > that seem to be available in a script:python section?
> 
> There is Coccilib.3cocci, but that is for ocaml.
> 
> My understanding is that there are reflective features in python that let
> you find out what functions are available.  python/python_documentation.md
> in the source tree also has some information.  Or look at the code in
> python/coccilib.
> 
> julia
> 

Re: [cocci] How do you concatinate string constants in SmPL with python?

[Julia Lawall]

On Sat, 2 Apr 2022, Eric Wheeler wrote:

> On Sat, 2 Apr 2022, Julia Lawall wrote:
> > On Sat, 2 Apr 2022, Eric Wheeler wrote:
> > > I would like to do something like this:
> > >
> > > @ r @
> > > constant C;
> > > @@
> > > -perror(C);
> > > +pr_err(C ": %s\n", strerror(errno));
> > >
> > > Thus in python I need something like C = C + C ": %s\n".
> >
> > I don't understand this.
>
> For example:
>
> -perror("unable to open file");
> +pr_err("unable to open file: %s\n", strerror(errno));
>
> I'm trying to append ": %s\n" to the original string (C) so we can process
> the error using pr_err() instead of perror().  Since SmPL does not support
> string concatination (so far as I am aware), I would need to use python.
>
> But I'm not sure how to structure the python bits so they interact with
> cocci.  I've looked at some examples but I don't see how to change the
> value of C to something else, such as C = C + ": %s\n" where + is the
> append operator in python.

You should check demos/pythontococci.  You need to make three rules, one
to match, one to use python to create the new string you want (in the
place of "something" in the demo), and one to rematch and generate the
code.

julia



>
> --
> Eric Wheeler
>
>
> >
> > >
> > > I've tried a few things, but I don't think I'm anywhere close to getting
> > > it right.  Can someone help with how this should be setup?
> > >
> > > Also is there a reference for the `coccinelle.` and `cocci.` python calls
> > > that seem to be available in a script:python section?
> >
> > There is Coccilib.3cocci, but that is for ocaml.
> >
> > My understanding is that there are reflective features in python that let
> > you find out what functions are available.  python/python_documentation.md
> > in the source tree also has some information.  Or look at the code in
> > python/coccilib.
> >
> > julia
> >
>

Re: [cocci] How do you concatenate string constants with SmPL?

[Markus Elfring]

[-- Attachment #1: Type: text/plain, Size: 1461 bytes --]


> For example:
>
> -perror("unable to open file");
> +pr_err("unable to open file: %s\n", strerror(errno));
>
> I'm trying to append ": %s\n" to the original string (C) so we can process
> the error using pr_err() instead of perror().


It is a pity that the following transformation approach failed with the message
“plus: parse error: …”.

@replacement@
@@
-perror
+pr_err
       ("unable to open file"
+       ": %s\n", strerror(errno)
       );


> Since SmPL does not support string concatination (so far as I am aware),


It would be nice if adjustments will become more convenient also for string literals.



> I would need to use python.


You can choose from the supported scripting languages.



> But I'm not sure how to structure the python bits so they interact with cocci.


Something can be achieved by using three SmPL rules for the mentioned situation.

1. Search for your update candidates.
2. Specify SmPL variables in a script rule so that data generation will be performed.
3. Integrate data into the desired source code adjustment.


Can you get further development ideas from my SmPL script example according to
the discussion topic “Can Coccinelle concatenate or append to existing string literals?”?
https://lore.kernel.org/cocci/caf9ec38-c627-711d-885d-b975cc914fca@web.de/
https://sympa.inria.fr/sympa/arc/cocci/2022-03/msg00058.html

Regards,
Markus

[-- Attachment #2: Type: text/html, Size: 3075 bytes --]

Re: [cocci] Completing documentation for SmPL scripting interfaces

[Markus Elfring]

>> Also is there a reference for the `coccinelle.` and `cocci.` python calls
>> that seem to be available in a script:python section?
> There is Coccilib.3cocci, but that is for ocaml.
>
> My understanding is that there are reflective features in python that let
> you find out what functions are available.  python/python_documentation.md
> in the source tree also has some information.  Or look at the code in
> python/coccilib.


Would you like to improve the documentation for these programming interfaces?
https://gitlab.inria.fr/coccinelle/coccinelle/-/blob/5069eaeadd731ecdd99e7a6f4465c286a2792354/python/python_documentation.md#fields-and-methods-exposed-via-cocci-object
https://github.com/coccinelle/coccinelle/blob/31bf3aefce245e43cc58f6b6ff49fc9b8e9084ff/python/python_documentation.md#fields-and-methods-exposed-via-cocci-object


Should the method “cocci.make_expr()” be explained also?

Regards,
Markus

Re: [cocci] Completing documentation for SmPL scripting interfaces

[Nicolas Palix]

On 03/04/2022 16:55, Markus Elfring wrote:
>>> Also is there a reference for the `coccinelle.` and `cocci.` python calls
>>> that seem to be available in a script:python section?
>> There is Coccilib.3cocci, but that is for ocaml.
>>
>> My understanding is that there are reflective features in python that let
>> you find out what functions are available.  python/python_documentation.md
>> in the source tree also has some information.  Or look at the code in
>> python/coccilib.
> 
> 
> Would you like to improve the documentation for these programming interfaces?
> https://gitlab.inria.fr/coccinelle/coccinelle/-/blob/5069eaeadd731ecdd99e7a6f4465c286a2792354/python/python_documentation.md#fields-and-methods-exposed-via-cocci-object
> https://github.com/coccinelle/coccinelle/blob/31bf3aefce245e43cc58f6b6ff49fc9b8e9084ff/python/python_documentation.md#fields-and-methods-exposed-via-cocci-object
> 
> 
> Should the method “cocci.make_expr()” be explained also?

An example:

https://gitlab.inria.fr/coccinelle/coccinelle/-/commit/d0666cf3f6faa4516bd42bac1f319d6bc894e32d
> 
> Regards,
> Markus
> 


-- 
Nicolas Palix

Re: [cocci] Evolution according to SmPL scripting interfaces and regular expressions

[Markus Elfring]

>> Should the method “cocci.make_expr()” be explained also?
>
> An example:
>
> https://gitlab.inria.fr/coccinelle/coccinelle/-/commit/d0666cf3f6faa4516bd42bac1f319d6bc894e32d


Thanks for your contribution “Scripting: Add example of make_expr”.


I noticed also your contribution “Regexp: Better error report with PCRE”
because of this reference.

Will your development branch “npalix-regexp” trigger any more software evolution
also according to improved handling of regular expressions?

Regards,
Markus