Things to be improved

[adrianq]

• Proper exception handling
• Files in src/test/resources/tmp shouldn't be in the repo
• When a model is not supported return a report. (this was working but I have to remove it during the refactor task)
• All the files for testing should be contain in the export projects. At the moment some of them are in the project while some others are in neuroml.model
• Remove all main methods from writers. They should go either to Travis or to jUnit tests.
• These two lines

MinimalMessageHandler.setVeryMinimal(true);
 E.setDebug(false);

can be found in several writers. Is this sth we can generalize? Can we move them to some abstract class or utils?

• Validate if outputFolder exists and generate proper exception if this is not the case
• Refactor remaining hardcoded file extension to new Formats enum class (some of them have already been moved)
• org.neuroml.import has a dependency with org.neuroml.export just because a util method. Shall we remove this dependency and add a new dependency between org.neuroml.model and org.neuroml.export? Maybe this util method can be moved to org.neuroml.model
• move outputFileName to an abstract class
• add a unit test for jNeuroml
• Refactor neuron code
• Refactor infowriter code
• Improve and generalize logging system
• Documentation
• Improve junit tests. They should extend from a super class as functionality is similar in most of the cases or at least create a util class. Also we should improve what we are checking/validating at the moment. Most cases we just validate that the file exists but some extra validation would be nice
• Implement getSupportedOutputs functionality for a specific file. This method will be used from Geppetto to check which conversions can be done in neuroml.export lib.

[sanjayankur31]

I'll go through this list and see what remains to be done. On my backlog now.

[sanjayankur31]

A lot of this may not be relevant now, so I'm leaving this issue open but dropping it for now.