original_requirements.md

Original Requirements

1. Accept and validate a completed verbal learning template csv
2. Return scores conversions for each column, based on the column name; (see 9. below)
3. Ignore unknown columns
4. Generate errors on unacceptable input data
   • any input that isn't in valid csv format
   • any input that contains no single matching column header
5. Generate a blank template on demand
6. Work with Shiny
7. a function which returns a text object containing the template
8. a function which returns a stream io object container the template
9. column names matching the default column format will be calculated. Non-matching columns will be ignored

   • columns in scope match the regular expression

     r"^(cvlt|cvltc|ravlt|hvlt)_([a-z_])+$"

   • the column names shall be split on '_'

   • the total number of tokens after splitting on '_' determines what conversion to apply

     • matching column titles with four tokens
       1. instrument
       2. item
       3. trial
       4. value type
     • matching columns with three tokens
       1. instrument
       2. item
       3. value type
     • matching columns with two tokens
       1. instrument
       2. instrument metadata type

   • cvlt, cvltc, ravlt, hvlt are the instrument names

   • dr, imfr, sdcr, sdfr, ldfr, ldcr, recog, rep, int are the item names

   • t1, t2, t3, t4, t5, t13, t15, tb, b are the trial names

   • total, hits, c, i, fp are the value types

   • form, version are instrument metadata types

   • conditionals

     • if instrument is 'cvltc', ignore
     • if the item is in ['sdfr', 'ldfr']
     • if the trial isn't in ['totals', 't1', 't2', 't3', 't4', 't5', 'b']
10. a function which takes a string (matching the column formatting rules above), and returns a tuple of values representing the source value, and the desired target value, and raises an error on negative, non-integer, or out of range values.
11. a function which returns a text object containing the resulting converted data
12. a function which return a stream object containing the resulting converted data
13. a function which takes a text object, and returns a random sample of the converted dataset.