Currently, a single script ./manipulation/0-ellis-island.R implements parsing of the models from both physical-cognitive and physical-physical track. The switching between tracks occurs by turning off/on relevant lines of code. While the parsing functions applied are the same, the logistics of the data flow appears to favor separate Ellis Islands (and potentially more, if other tracks are added).
Creating individual Ellis Islands for individual tracks appears to offer greater clarity of the data flow and more localized debugging. Note, that this may not be relevant under the conditions of "clean" model results supplied by Generation 2 Portland scripts.
Currently, a single script
./manipulation/0-ellis-island.Rimplements parsing of the models from bothphysical-cognitiveandphysical-physicaltrack. The switching between tracks occurs by turning off/on relevant lines of code. While the parsing functions applied are the same, the logistics of the data flow appears to favor separate Ellis Islands (and potentially more, if other tracks are added).Creating individual Ellis Islands for individual tracks appears to offer greater clarity of the data flow and more localized debugging. Note, that this may not be relevant under the conditions of "clean" model results supplied by Generation 2 Portland scripts.