review of indexing across R and C++

[ajrominger]

We should just dedicate time to going over indexing once and for all! a few questions:

• [x] when roleModel objects are initialized species IDs are indexed starting at 1, is that what we want?
• [x] are species IDs indexed starting at 0 in C++, how is conversion from R --> C++ done?
• [x] are 0 indexed spp IDs passed back to R and augmented by 1 so they're 1 indexed again?
• [x] the python code deals with converting 1 indexed to 0 indexed, should we avoid that be keeping it 0 indexed throughout?

[jidec]

the approach currently is that all objects living in R should be 1 indexed and all objects living in Cpp or Python are 0 indexed

currently indSpecies, which contains indices for species should start at 1 while in R and start at 0 while in Cpp. We subtract one in C++ right after indSpecies enters C++ (in roleDataCpp constructor) and add 1 in runRole immediately after the model is run

the edges in the phylo also need to be indexed starting at 1 in R and starting at 0 in Cpp, so we also decrement in C++ in the roleDataCpp constructor and increment the edges by 1 in runRole immediately after the model is run

we may want to do this for tipNames, which contains species indices as strings of the form "t1","t2", etc. but I haven't bothered because tipNames is currently not used at all within Cpp.

[ajrominger]

all sounds good! as discussed, let's leave tipNames as they are for now. i'm leaving this issue open in case we need to deal with tipNames in the near future