chathurawidanage / cor-map Goto Github PK

Currently, both programs involved in a relationship are filtered on the same enrollment start and end dates

Currently, each legend item must select only one program. If a TEI is enrolled in multiple programs, the first one found is used to set the color of the node in the visualization. Instead, it should be possible to support legend items for combinations or programs - (such as case + contact or case only) as well as by program stage (suspected, confirmed, contacted, etc)

Currently it is only possible to filter by enrollment date, but org unit filtering would be very beneficial for reducing the scope of the visualization

Currently this can be configured but it is not well-tested

Instead of displaying the entire relationship network, for contact tracing use-cases it is important to support a search-and-trace visualization type. This is also much more performant, since it doesn't require downloading large numbers of TEIs and filtering them client-side.

There is an example of this type of visualization in the Contact Tracer tab of the R Shiny app shared by Brian O'Donnell

The attributes displayed in the details dialog are not particularly novel - we should perform some basic relationship analytics (such as simply the number of relationships, perhaps including relationship program indicators) and display the result in the dialog

Currently it is only possibly to save visualizations in the user data store for the logged-in user. It would be very useful to add support for sharing of saved visualizations with other users and user groups. This should be implemented generically in the datastore service

(minor, under-the-hood enhancement)

Use dhis2/app-service-datastore and dhis2/ui for saving/loading objects and for general layout building blocks, rather than copying those libraries locally.