Attention is currently required from: jolly, laforge.
dexter has posted comments on this change by jolly. ( https://gerrit.osmocom.org/c/onomondo-ipa/+/43054?usp=email )
Change subject: V1.2: Add function ES10b.GetConnectivityParameters ......................................................................
Patch Set 7: Code-Review+1
(4 comments)
Patchset:
PS7: I think the return code should be 0 in case no httpParams are deposited. The caller then gets a null pointer in *http_params and the return code 0 would tell the caller that the command was successful but there were simply no httpParams deposited.
PS7:
Refer to Change-Id: I1ac4cf8980af501bb119208213242468ee4f3c06
Also good to know: The asn1c compiler writes a comment in each generated source file: "Generated by asn1c-0.9.29 (http://lionet.info/asn1c)"
File src/ipa/libipa/ipad.c:
https://gerrit.osmocom.org/c/onomondo-ipa/+/43054/comment/a8e7e60a_4c02dcc9?... : PS7, Line 366: res->res->choice.connectivityParameters.httpParams->buf); maybe the sheer absence of httpParams is not an error. Its an optional fields. Maybe some cards just return an emtpy struct if the httpParams are not set.
https://gerrit.osmocom.org/c/onomondo-ipa/+/43054/comment/c1bef69a_da962df8?... : PS7, Line 369: } maybe it makes sense to evaluate the GetConnectivityParametersResponse_PR_connectivityParametersError. If res->res->choice.connectivityParametersError is ConnectivityParametersError_parametersNotAvailable, we may return 0, since there are no parameters available, which is technically not an error. In all other cases we would return -EINVAL. Then the caller can know if he couldn't get the parameters just because there are none or if the reason was a real error.
-
dexter