Attention is currently required from: jolly, pespin.
5 comments:
File s1gw/S1GW_ConnHdlr.ttcn:
Patch Set #2, Line 1210: in ERabIdxList erabs_release := {})
I vote for stopping adding more "in" churn :D
For `ERabIdxList`, yes, we don't really need `in/out`. However we should keep using them for `ERabList`, to let the caller know if the given list is going to be modified or not.
Patch Set #2, Line 1720: in ERabIdxList erabs_forward,
No more "in" churn please.
Why are you so obsessed with this?
function f_ConnHdlr_handover_cmd(MME_UE_S1AP_ID mme_ue_id,
ENB_UE_S1AP_ID enb_ue_id,
in ERabIdxList erabs_forward,
Nitpick: in the existing API, arguments `mme_ue_id` and `enb_ue_id` usually follow `ERabList`/`ERabIdxList` arguments. For the sake of consistency, I suggest:
```suggestion
function f_ConnHdlr_handover_cmd(ERabIdxList erabs_forward,
ERabIdxList erabs_release,
MME_UE_S1AP_ID mme_ue_id,
ENB_UE_S1AP_ID enb_ue_id)
```
File s1gw/S1GW_Tests.ttcn:
Patch Set #2, Line 911: for (var integer i := 0; i < lengthof(erabs_forward); i := i + 1) {
I see this logic repeated in several places, so I think it's worth having a function in `S1GW_ConnHdlr` (next to `f_ERabList_find_by_pdu()`):
```
function f_ERabList_compose(ERabIdxList idx_list)
runs on ConnHdlr return ERabList {
var ERabList erabs;
for (var integer i := 0; i < lengthof(idx_list); i := i + 1) {
var ERabIdx idx := idx_list[i];
erabs[i] := g_pars.erabs[idx];
}
return erabs;
}
```This way here you could do:
```
if (lengthof(erabs_forward) > 0) {
var ERabList erabs := f_ERabList_compose(erabs_forward);
f_ConnHdlr_erab_release_cmd(erabs, g_pars.mme_ue_id, g_pars.idx);
f_ConnHdlr_erab_release_rsp(erabs, g_pars.mme_ue_id, g_pars.idx);
}
```
tabs vs spaces
To view, visit change 41094. To unsubscribe, or for help writing mail filters, visit settings.