On 01/15/2010 07:27 AM, Julia Lawall wrote:
This in turn leads to many permutations of bindings of the position
variables. Because some of the relevant code is under ifs, not all paths
contain the same set of positions.
Is that fixable? Or something to keep in the back of the mind to use<... ...>
only scarcely when absolutely necessary?
No, I don't think it is fixable in general, although it might be fixable
Too bad as the general case is interesting. For this particular case
there is a way better solution that finishes the whole Wine tree in 6
minutes (plus 3 more minutes for the generated C files).
in this case where the<... ...> takes up the entire body of the function.
In this case it was the "natural solution" to grab the function name. I
tried first to use p[0].function in the hope it might be there too like
.line and .file but it wasn't. I had seen the function name saved into p
when running with -debug but not knowing python I didn't know yet about
the wonders of dir() and type() ;).
Actually, I think I had an optimization like that, but it might have
turned out to be a bad idea. But anyway, not using it is always faster
than using it.
I said it before but this was really a good learning opportunity. Thanks!
bye
michael
_______________________________________________
Cocci mailing list
[email protected]
http://lists.diku.dk/mailman/listinfo/cocci
(Web access from inside DIKUs LAN only)
