Re: [Cocci] Checking uniqueness for source code positions during SmPL data processing

[Julia Lawall <julia.lawall@lip6.fr>]

[-- Attachment #1: Type: text/plain, Size: 2531 bytes --]



On Mon, 22 Apr 2019, Markus Elfring wrote:

> Hello,
>
> I reported that I am trying a specific source code analysis out again.
> Information can be imported also into database tables for such a purpose.
> I observed a primary key constraint violation for my data processing attempt.
> Useful background information can be found for a topic like
> “Checking the handling of unique keys/indexes”.
> https://groups.google.com/d/msg/sqlalchemy/klmUwiirIQw/LDeeRTcshQ4J
>
> A corresponding aspect can trigger an usual development challenge.
> The transaction fails if questionable data were detected. It seems to be hard
> to find the single inappropriate data set out by SQL programming interfaces.

I'm not going to debug anything that involves external tools, ie your
database.

Note however that by converting from * to printing, you have converted the
...s in your searching rule from "exists" to "forall" as the quantifier
over the paths.  You may want to put exists in the header of the searching
rule.

julia


>
> Thus I developed the attached script variant for the semantic patch language.
> Another test result points interesting details out, doesn't it?
>
>
> elfring@Sonne:~/Projekte/Linux/next-patched> time spatch ~/Projekte/Coccinelle/janitor/list_duplicate_statement_pairs_from_if_branches5.cocci drivers/media/dvb-frontends/stv0297.c
> …
> A duplicate key was passed.
> function: stv0297_readreg
> file: drivers/media/dvb-frontends/stv0297.c
> line: 87
> column: 4
> Traceback (most recent call last):
>   File "<string>", line 4, in <module>
>   File "<string>", line 26, in store_statements
> RuntimeError
> exn while in timeout_function
> Error in Python script, line 34, file …
>
> real	0m0,606s
> user	0m0,541s
> sys	0m0,037s
>
>
> By the way: I would like to point out once more that the code from
> the SmPL rule “initialize” is 18 lines long and the definition for
> the function “store_statements” starts at line 4 originally.
>
>
> The implementation of the function “stv0297_readreg” contains two statements
> which are repeated in three if branches for the desired exception handling.
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/media/dvb-frontends/stv0297.c?id=085b7755808aa11f78ab9377257e1dad2e6fa4bb#n66
>
>
> Now I wonder about the shown software behaviour again when the corresponding
> source code position should be unique based on the specified data fields.
> How can affected software areas be improved further?
>
> Regards,
> Markus
>

[-- Attachment #2: Type: text/plain, Size: 136 bytes --]

_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci