From mboxrd@z Thu Jan  1 00:00:00 1970
From: johannes@sipsolutions.net (Johannes Berg)
Date: Thu, 26 Jan 2017 16:52:31 +0100
Subject: [Cocci] modifying initializers with spatch?
In-Reply-To: <alpine.DEB.2.20.1701261644140.3109@hadrien>
References: <1485433691.14760.1.camel@sipsolutions.net>
 <alpine.DEB.2.20.1701261415040.3109@hadrien>
 <1485437521.14760.5.camel@sipsolutions.net>
 <alpine.DEB.2.20.1701261518560.3109@hadrien>
 <1485440882.14760.8.camel@sipsolutions.net>
 <alpine.DEB.2.20.1701261644140.3109@hadrien>
Message-ID: <1485445951.14760.16.camel@sipsolutions.net>
To: cocci@systeme.lip6.fr
List-Id: cocci@systeme.lip6.fr


> The problem is that Coccinelle lets you put fewer field initializers
> than
> are actually available.??The reasoning is that they are optional for
> fields, and the user doesn't really know how many will be present in
> each
> case.??{} is the degenerate case where none are mentioned.??But that
> should not apply to your case, 1) because you are removing the
> braces, and
> 2) because you aren't doing field initializations anyway.??I'll have
> to look into it.

Ok, thanks.

I actually had a separate thought, that I could perhaps do something
like


{
... when != { ... }
T x = {
-  .field = E,
};
... when != S
    when any
++  x.field = E;


To get rid of all the initializers, but (as I'm still on the phone) I
haven't tried it yet :)

Btw, how was the ++ supposed to work? I had to run this more than once
to actually make it catch multiple instances of = {0}. Also, in the
case above, even if it works, the order of initializations might be
inverted if I run the spatch multiple times, so not sure I should do
that.

johannes