Follow-up from "Server: add missing mandatory attributes + fixes to pass UACTT tests (nano scope)"
The following discussions from !297 (merged) should be addressed:
-
@v.monfort started a discussion: (+1 comment) The dynamic address space XML parser was not updated. It shall be done to keep it compatible with the modifications done in Python address space XML parser.
-
@V.Lacroix started a discussion: duplicated line with L78.
-
@V.Lacroix started a discussion: Nothing on serverTimestamp ?
-
@V.Lacroix started a discussion: Replace with a_Value_SourceTimestamp : a_NodeClass~[{e_ncl_Variable}] --> t_DateTime ?
-
@V.Lacroix started a discussion: Can be rewritten as follows: ResponseWrite_StatusCode[cast_t_WriteValue[1..nb_WriteValue]] <: {e_sc_bad_out_of_memory, e_sc_bad_attribute_id_invalid, e_sc_bad_index_range_invalid, e_sc_bad_node_id_unknown, e_sc_bad_not_writable, e_sc_bad_type_mismatch, e_sc_bad_user_access_denied, e_sc_bad_index_range_no_data}
-
@V.Lacroix started a discussion: Why is the Ok status not in the list.
-
@V.Lacroix started a discussion: Why did you removed this part which seems more precise ?
-
@V.Lacroix started a discussion: Invariant is inconsistent with initialization. To be discussed.
-
@V.Lacroix started a discussion: copy paste error on this line.
-
@V.Lacroix started a discussion: copy paste error.
-
@V.Lacroix started a discussion: Are you sure this type of comment is a good idea ?
-
@V.Lacroix started a discussion: Why is this declaration here ?
-
@V.Lacroix started a discussion: All the constants set to -1 can be removed from the model.
-
@V.Lacroix started a discussion: Do you confirm it is not possible to read the value attribute of a VariableType ?
-
@V.Lacroix started a discussion: Maybe update the comment of browseSubTree in common.py ?
-
@V.Lacroix started a discussion: Replace with:
if ua_node.value is None and ua_node.nodeid.ns not in (None,0):
-
@V.Lacroix started a discussion: why not a
const
here ? -
@V.Lacroix started a discussion: l_compat_with_conv is not used ?
-
@V.Lacroix started a discussion: Replace with
bres := bool( (p_expValueRank = -3 & p_concValueRank : {-1,1}) or (p_expValueRank = -2) or (p_expValueRank = -1 & p_concValueRank = -1) or (p_expValueRank = 0 & p_concValueRank >= 1) or (p_expValueRank >= 1 & p_expValueRank = p_concValueRank) )
-
@V.Lacroix started a discussion: To be suppressed ?
-
@V.Lacroix started a discussion: To be checked with B experts: link between
is_included_ValueRank
and this variable is probably missing to perform the proof. -
@V.Lacroix started a discussion: Duplicated line. Same issue for the two followings functions.
-
@V.Lacroix started a discussion: Why not a static const here ?
-
@V.Lacroix started a discussion: I think we can easily reduce the size of this array. For example, why don't we set an array of int16 giving us the integer nodeId for each index and build the SOPC_NodeId in a function or even a macro ?
-
@V.Lacroix started a discussion: This condition can be removed.
-
@V.Lacroix started a discussion: Suggestion: replace these numbers by macros or an enumerate.
-
@V.Lacroix started a discussion: replace
== None
byis None
replace== False
byis False
-
@V.Lacroix started a discussion: Ref is not None
? -
@V.Lacroix started a discussion: Are you sure this condition is necessary giving the second one ?
-
@V.Lacroix started a discussion: Always True. Add parenthesis or separate into two different implications.
-
@V.Lacroix started a discussion: I think it would be better if function names start with a verb to differentiate them from constants or enumerates.
-
@V.Lacroix started a discussion: be authorized ?
-
@V.Lacroix started a discussion: Why do the service read use iterators which seem dedicated to write request ?
-
@V.Lacroix started a discussion: You are not supposed to keep the source timestamp in this case ?
-
@V.Lacroix started a discussion: Why index_range_statuscode here ?
-
@V.Lacroix started a discussion: to return