> Please actually try things out before declaring them to be useless. This feedback provides also another opportunity for collateral evolution in some directions. I am curious on how involved uncertainty can be fixed around possibly different interpretation for provided software functionality.
The SmPL construct “... when …” is mentioned in an area (of the section “Basic transformations” in the software documentation) which is introduced with the wording “The grammar for the minus or plus slice of a transformation is as follows:”. https://github.com/coccinelle/coccinelle/blob/c6d7554edf7c4654aeae4d33c3f040e300682f23/docs/manual/cocci_syntax.tex#L1033 I got the impression that the corresponding meaning is not explained in this information source so far. The published example “Reference counter: the of_xxx API” can be interesting then to some degree for the explanation of the discussed development efforts. https://github.com/coccinelle/coccinelle/blob/175de16bc7e535b6a89a62b81a673b0d0cd7075c/docs/manual/examples.tex#L320 If the available application documentation is still too limited (and incomplete because it is also work in progress), it is probably usual that SmPL code occasionally tries to express expectations which are not covered by an evolving software implementation. How would you like to improve the situation further? * Is it certain that a search is performed only for the source code “x == NULL” (and corresponding isomorphisms) by the SmPL constraint “when != true” (after a successful null pointer check was detected in this use case)? * Would you like to test any functionality which should work in different ways than you might see from the original OCaml source code? https://github.com/coccinelle/coccinelle/issues/134 Regards, Markus
