README.md

Data

Data for testing readstat-rs binary

Sources

• ahs2019n.sas7bdat → US Census data
  • http://www2.census.gov/programs-surveys/ahs/2019/AHS%202019%20National%20PUF%20v1.1%20Flat%20SAS.zip
  • Must be downloaded manually as currently ignored by git (i.e. has been added to the repository .gitignore file)
  • Renamed to be _ahs2019n.sas7bdat in order to be picked up by the _*.sas7bdat pattern in the .gitignore file
• all_types.sas7bdat → SAS dataset containing all SAS types
  • Created using create_all_types.sas
• cars.sas7bdat → SAS cars dataset
  • https://www.alanelliott.com/sas/ED2_FILES.html
• hasmissing.sas7bdat → SAS dataset containing missing values
  • https://www.alanelliott.com/sas/ED2_FILES.html
• intel.sas7bdat
  • https://www.alanelliott.com/sas/ED2_FILES.html
• messydata.sas7bdat
  • https://www.alanelliott.com/sas/ED2_FILES.html
• rand_ds.sas7bdat → Created using create_rand_ds.sas
  • Renamed to be _rand_ds.sas7bdat in order to be picked up by the _*.sas7bdat pattern in the .gitignore file
• rand_ds_largepage_err.sas7bdat → Created using create_rand_ds.sas with BUFSIZE set to 2M
  • Does not parse with version 1.1.6 of ReadStat
• rand_ds_largepage_ok.sas7bdat → Created using create_rand_ds.sas with BUFSIZE set to 1M
  • Parses with version 1.1.6 of ReadStat
• scientific_notation.sas7bdat → Used to test float parsing
  • Created using create_scientific_notation_ds.sas
  • Previously parsed floats with lexical by first converting to a string lexical::to_string(value) and then converting back (after truncating) with lexical::parse
  • Had an issue where large floats were being read correctly via ReadStat and the string conversion via lexical resulted in a string with scientific notation; after trying to parse back from a string to a float with lexical::parse there would be an error thrown
  • Fixed by d301a9f9ff8c5e3c34a604a16c095e99d205f624
• somedata.sas7bdat
  • https://www.alanelliott.com/sas/ED2_FILES.html
• somemiss.sas7bdat
  • https://www.alanelliott.com/sas/ED2_FILES.html