> > I am trying to move local variable initialization of some variable to
> > first statement. …
I find such a source code transformation task also interesting.
But it points some software development challenges out also for the safe
application
of the semantic patch language for this use case.
> put \(<+...a...+>\&E\) in place of E
How do you think about the following SmPL script variant?
@movement@
expression E;
identifier V1, V2, Vx;
statement S;
type T1, T2;
@@
T1 V1;
... when any
T2 V2
- = \( <+... Vx ...+> \& E \)
;
... when any
when != S
++ V2 = E;
elfring@Sonne:~/Projekte/Coccinelle/janitor> spatch --parse-cocci
convert_variable_initialisation_to_assignment.cocci
…
plus: parse error:
File "convert_variable_initialisation_to_assignment.cocci", line 10, column
9, charpos = 117
around = '\(',
whole content = - = \( <+... Vx ...+> \& E \)
I have tried another SmPL code variation out then.
-T2 V2 = \( <+... Vx ...+> \& E \)
+T2 V2
;
elfring@Sonne:~/Projekte/Coccinelle/janitor> spatch --parse-cocci
convert_variable_initialisation_to_assignment2.cocci
…
14: no available token to attach to
How will the software evolution be continued here?
Regards,
Markus
_______________________________________________
Cocci mailing list
[email protected]
https://systeme.lip6.fr/mailman/listinfo/cocci
