add functionality for csv files created from RiverWare/RiverSMART

[rabutler]

Now that RiverWare and RiverSMART can create csv files, we want to be able to handle those too. Probably includes a lot, but highest priority is ability to pass that data frame and an slot agg list to the aggregation functions and get the same data frame back as we would from an rdf.

[rabutler]

Will need to be able to read in single csv files and csv files that have been combined in RiverSMART

1. create readRwCsv()
   • it will read in the csv
   • remove spaces from column names
   • check that it has the expected column names (required and optional)
   • possibly could have it make a new object that contains tables to convert slots to variable names, etc.
2. rdfToTbl(read.rdf2()) to read in an rdf and create a dataframe/tbl that is the same as the dataframe returned by read_rw_csv()

[rabutler]

• [x] rw_read_rw_csv()
• [x] rw_read_rw_csv() should convert the timestep to match that from the rdf
• [x] rw_read_rw_csv() unit tests
• [x] rdfToTbl()
• [x] change name to rw_rdf_to_tbl()
• [x] rw_read_rw_csv() needs an option to add Year and Month columns.
• [x] rw_rdf_to_tbl() unit tests
• [x] new aggregation methods that take a data frame as input. #51
• [x] check that for a single scenario of data the new aggregation methods work for csv or rdf input moved to #51

[rabutler]

rdfToTbl(rdf, scenario = NULL)

if scenario is NULL, then do not add Scenario column

[rabutler]

For rw_rdf_to_tbl()

• [x] documentation
• [x] test with yearly data
• [x] decide how to represent timestep and if it should be different for yearly vs monthly
• [x] determine how to limit which variables are returned by function

---

It works with yearly data just fine. The Timestep is yyyy-12-31 for each value. Don't think we need to do anything special to handle annual data.

---

To limit the variables, we can use reqColumns() + maybe some others? And take this as an argument to the function. Need to figure out which values are always saved from our default RW csv output. But also need to be able to handle when those columns aren't there.

[rabutler]

csv output is in mm-dd-yyyy 23:59:00 format rdf contains timesteps in yyyy-m-dd 24:00 format

---

rw_read_rw_csv() now converts from mm-dd-yyyy format to yyyy-m-dd format to be closer to the tbl returned by rw_rdf_to_tbl(), but it left the hours-seconds alone. Since we'll probably convert to yearmon for comparison anyways, they should now be the same.