Fix infection origin: remove unnecessary UC check

[acavelan]

Fix related to: #402

The UC check is not necessary for InfectionOrigin. This can be used both inside an intervention and for Uncomplicated Care in the health system.

[b-s-code]

I've given this a brief review and I can see that this change makes the code consistent with the base class' method documentation 👍

openmalaria/model/Clinical/CMDecisionTree.h

Lines 69 to 78 in b31d36a

	/** Create a user-configured decision from an XML node.
	*
	* Memory management is handled internally (statically).
	*
	* @param node XML element describing the tree
	* @param isUC If isUC is false and a "case type" decision is created, an
	* xml_scenario_error exception is thrown. CMHostData::pgState is only
	* used by the "case type" decision, so if isUC is false when creating the
	* tree, pgState does not need to be set when executing the tree. */
	static const CMDecisionTree& create( const ::scnXml::DecisionTree& node, bool isUC );

I have some questions slightly beyond the scope of this PR.

I don't suppose you know why the case type decision tree isn't meant to work with complicated cases? I don't yet understand the full context surrounding this code.

I had a look at the schema to see if the docs there explained it. Interestingly, it seems like:

• The schema actually permits a case type decision tree with a complicated case.
• The exception being thrown in CMDTCaseType::create is where the rule against using a case type decision tree with a complicated case is enforced?

As far as I can tell, the reason that the schema permits using a case type decision tree with a complicated case is this line below.

openmalaria/schema/healthSystem.xsd

Line 494 in b31d36a

<xs:element name="complicated" type="om:DecisionTree"/>

Maybe it is simpler to enforce the rule in CMDTCaseType::create than in the schema..

[b-s-code]

(Happy for this PR to be merged regardless of what the answers to the above questions may be.)

[acavelan]

Exactly, the DecisionTree is defined once in the xml (in a recursive manner) and the same definition is used in both the intervention section and the health system section. So it's easier to work with this one definition and enforce the extra rule in the C++ code.

The CaseType is defined as: "A switch which chooses a branch deterministically, based on whether the patient was treated recently (second line) or not (first line). For uncomplicated cases only." This only works for health system treatments, not treatments given through interventions, which are a bit more flexible.

[b-s-code]

Makes sense - thanks for clarifying.