* [Cocci] Searching designated initialisers with SmPL
@ 2019-10-27 15:12 Markus Elfring
2019-10-27 16:14 ` Julia Lawall
0 siblings, 1 reply; 6+ messages in thread
From: Markus Elfring @ 2019-10-27 15:12 UTC (permalink / raw)
To: Coccinelle
Hello,
I hoped to achieve something together with the semantic patch language
by the following search pattern.
@display@
identifier action, driver;
@@
static struct platform_driver driver =
{
<+...
*.remove = action
...+>
};
Unfortunately, I stumble on another error message.
elfring@Sonne:~/Projekte/Coccinelle/janitor> spatch --parse-cocci show_designated_initialisers6.cocci
…
minus: parse error:
File "show_designated_initialisers6.cocci", line 7, column 1, charpos = 91
around = '.',
whole content = *.remove = action
How would you like to clarify the software behaviour for
such an use case?
Regards,
Markus
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [Cocci] Searching designated initialisers with SmPL
2019-10-27 15:12 [Cocci] Searching designated initialisers with SmPL Markus Elfring
@ 2019-10-27 16:14 ` Julia Lawall
2019-10-27 16:47 ` Markus Elfring
2019-10-27 16:50 ` Julia Lawall
0 siblings, 2 replies; 6+ messages in thread
From: Julia Lawall @ 2019-10-27 16:14 UTC (permalink / raw)
To: Markus Elfring; +Cc: Coccinelle
[-- Attachment #1: Type: text/plain, Size: 888 bytes --]
On Sun, 27 Oct 2019, Markus Elfring wrote:
> Hello,
>
> I hoped to achieve something together with the semantic patch language
> by the following search pattern.
>
> @display@
> identifier action, driver;
> @@
> static struct platform_driver driver =
> {
> <+...
> *.remove = action
> ...+>
> };
>
>
> Unfortunately, I stumble on another error message.
>
> elfring@Sonne:~/Projekte/Coccinelle/janitor> spatch --parse-cocci show_designated_initialisers6.cocci
> …
> minus: parse error:
> File "show_designated_initialisers6.cocci", line 7, column 1, charpos = 91
> around = '.',
> whole content = *.remove = action
>
>
> How would you like to clarify the software behaviour for
> such an use case?
I don't know why you would even expect this to work. .remove is not a
mainingful construct in the C language. It is part of one, but it is not
a construct by itself.
julia
[-- Attachment #2: Type: text/plain, Size: 136 bytes --]
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [Cocci] Searching designated initialisers with SmPL
2019-10-27 16:14 ` Julia Lawall
@ 2019-10-27 16:47 ` Markus Elfring
2019-10-27 16:50 ` Julia Lawall
1 sibling, 0 replies; 6+ messages in thread
From: Markus Elfring @ 2019-10-27 16:47 UTC (permalink / raw)
To: Julia Lawall; +Cc: Coccinelle
> I don't know why you would even expect this to work.
I suggest to take another look at related information sources.
https://en.cppreference.com/w/c/language/struct_initialization
> .remove is not a mainingful construct in the C language.
I got an other impression from the declaration of a known data structure.
https://elixir.bootlin.com/linux/v5.4-rc2/source/include/linux/platform_device.h#L188
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/include/linux/platform_device.h?id=153a971ff578e9c8c5eadc9463c1d73a6adc8693#n190
> It is part of one, but it is not a construct by itself.
I am trying to improve source code analysis also around the usage
of function pointers.
But I have found some interesting information by the following search approach.
@find@
identifier action, driver;
@@
static struct platform_driver driver =
{
...,
.remove = action,
...
};
@display@
identifier find.action;
type T;
@@
T action(...)
{
<+...
*kfree(...);
...+>
}
So I guess that the clarification can become more interesting also for
the influence of commas and related SmPL constructs.
Regards,
Markus
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [Cocci] Searching designated initialisers with SmPL
2019-10-27 16:14 ` Julia Lawall
2019-10-27 16:47 ` Markus Elfring
@ 2019-10-27 16:50 ` Julia Lawall
2019-10-27 17:08 ` Markus Elfring
1 sibling, 1 reply; 6+ messages in thread
From: Julia Lawall @ 2019-10-27 16:50 UTC (permalink / raw)
To: Markus Elfring; +Cc: Coccinelle
On Sun, 27 Oct 2019, Julia Lawall wrote:
>
>
> On Sun, 27 Oct 2019, Markus Elfring wrote:
>
> > Hello,
> >
> > I hoped to achieve something together with the semantic patch language
> > by the following search pattern.
> >
> > @display@
> > identifier action, driver;
> > @@
> > static struct platform_driver driver =
> > {
> > <+...
> > *.remove = action
> > ...+>
> > };
Sorry, I thought that the above code was in a function definition.
It should be:
static struct platform_driver driver = { . remove = acrtion, };
There is no need for any form of ...
On the other hand the trailing comma is required.
julia
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [Cocci] Searching designated initialisers with SmPL
2019-10-27 16:50 ` Julia Lawall
@ 2019-10-27 17:08 ` Markus Elfring
2019-10-27 17:17 ` Julia Lawall
0 siblings, 1 reply; 6+ messages in thread
From: Markus Elfring @ 2019-10-27 17:08 UTC (permalink / raw)
To: Julia Lawall; +Cc: Coccinelle
> static struct platform_driver driver = { . remove = acrtion, };
>
> There is no need for any form of ...
I feel still unsure about the omission of SmPL ellipses
for the search of designated initialisers.
> On the other hand the trailing comma is required.
Thanks for such information.
Can this view change for the application of simpler search patterns?
Regards,
Markus
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [Cocci] Searching designated initialisers with SmPL
2019-10-27 17:08 ` Markus Elfring
@ 2019-10-27 17:17 ` Julia Lawall
0 siblings, 0 replies; 6+ messages in thread
From: Julia Lawall @ 2019-10-27 17:17 UTC (permalink / raw)
To: Markus Elfring; +Cc: Coccinelle
On Sun, 27 Oct 2019, Markus Elfring wrote:
> > static struct platform_driver driver = { . remove = acrtion, };
> >
> > There is no need for any form of ...
>
> I feel still unsure about the omission of SmPL ellipses
> for the search of designated initialisers.
>
>
> > On the other hand the trailing comma is required.
>
> Thanks for such information.
>
> Can this view change for the application of simpler search patterns?
No.
julia
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2019-10-27 17:18 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-10-27 15:12 [Cocci] Searching designated initialisers with SmPL Markus Elfring
2019-10-27 16:14 ` Julia Lawall
2019-10-27 16:47 ` Markus Elfring
2019-10-27 16:50 ` Julia Lawall
2019-10-27 17:08 ` Markus Elfring
2019-10-27 17:17 ` Julia Lawall
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).