- Gains new function
isd_parse_csv()for working with ISD csv files rather than the compressed ASCII text files thatisd_parse()parses. This new function introduces two new package imports: data.table and lubridate. The internals of the parsing tools for the various data types have changed to accommodate both csv and ASCII text files (#16)
- towards working on integrating metadata for each of the data fields, we've incorporated a data.frame of metadata into the package. see
?isd_metadata. eventually we'd like to allow conversions and such on the data based on units (#12)
- fix to internal parsing of string; first remove REM section from the ADD section so that codes in REM that happen to match those in ADD aren't detected spuriously (#15) thanks @ezwelty
- New function
isd_transform()that transforms some variables in ISD output data - but not all. This functionality used to be done by default inisd_parse()andisd_parse_line()- but data transforming yanked out as separate step done withisd_transform()(#11) - Both
isd_parse()andisd_parse_line()gain new parameteradditionalto optionally include additional and remarks data sections in output (#10)
- Fixed a bug in
isd()in which a data section was not parsed correctly. (#9)
- Released to CRAN.