Attention is currently required from: neels.
fixeria has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/36456?usp=email )
Change subject: msc: rework f_expect_paging(): handle mismatch and timeout ......................................................................
Patch Set 2:
(3 comments)
Commit Message:
https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/36456/comment/4b37daad_03125... PS2, Line 13:
you could add […]
Done
File msc/BSC_ConnectionHandler.ttcn:
https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/36456/comment/d071326a_4e324... PS2, Line 1362: [g_pars.ran_is_geran] BSSAP.receive(tr_BSSMAP_Paging(g_pars.imsi, tmsi));
(again the weird mix of g_pars. […]
Resolved in the new patchset.
https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/36456/comment/adb01b48_85731... PS2, Line 1366: function f_expect_paging(template OCT4 tmsi := *, float Tval := 4.0)
it would be very elegant if we could have the mechanism that {expects a message, but fails on a diff […]
Yes, a generic API for expecting the given message template would be nice. The problem with paging, though, is that it's not just one template but actually two different PDU types and thus two templates. So a generic `f_expect_foo()` API would need to accept two templates: one for GERAN and one for UTRAN.
While this approach might work, I believe having `f_expect_paging[_tmsi]()` and smaller altsteps (see next change) is a pretty good abstraction already, despite not being implemented using a generic `f_expect_foo()` API.
Imagine doing this every time you expect paging:
``` f_expect_foo(geran := tr_BSSMAP_Paging(g_pars.imsi, g_pars.tmsi), utran := tr_RANAP_Paging(cs_domain, imsi_hex2oct(g_pars.imsi))); ```
Calling `f_expect_paging()` is a lot shorter and self-explanatory.