Intention behind Parser::parseModel, updateModel, loadModel functions?

[kerimoyle]

I'm struggling to understand the need for three different functions here:

• parseModel which always creates a new Model instance from a string,
• updateModel which is only ever called from parseModel, and simply calls
• loadModel which always creates new entities inside the newly created model.

I don't get it ... why do we have an updateModel method, is private, used exactly once (in parseModel), and doesn't do what its name suggests (it naively appends, rather than updating) ... what am I missing? Why do we have it?

I'd like to change the updateModel function to one which:

• actually updates items rather than simply adding new copies of everything,
• is publicly available so the Annotator can use it.

This is related to #669 as it would be cleanest to make use of the serialisation functions in the Printer so that even automatically-generated ids are the same, whether they are constructed in Annotator or the Printer.

Thoughts? @nickerso @agarny @hsorby

[nickerso]

Not sure I see the connection between #669 and parsing? Or quite what it means for automatically generated ids to be the same?

[kerimoyle]

One option in 669 is to set the automatic ids in the Annotator using the Printer and then reparsing into a model. If that were the case then I'd want the parser to "update" rather than "append to" the model. We don't currently have that functionality (except it sounds like we do when you read the function names in the parser).