make_file: Implement the Specifications for Raw Files
In helseprofil/orgdata: Aggregating Original Data

make_file

R Documentation

Implement the Specifications for Raw Files

Description

Make a csv file with the specifications in the Access register database and implement them to the raw data of the selected group of files ie. (filgruppe). All files under the selected group will be affected unless the KOBLID with argument koblid is specified or select argument is used. Specifying koblid or select is useful especially for testing purposes.

This function is the most used function in KHelse for processing raw data. The function lag_fil() is an alias to make_file().

Usage

make_file(
  group = NULL,
  koblid = NULL,
  aggregate = NULL,
  save = FALSE,
  year.geo = NULL,
  implicitnull = NULL,
  row = NULL,
  base = NULL,
  parallel = deprecated(),
  raw = NULL,
  select = NULL
)

lag_fil(
  group = NULL,
  koblid = NULL,
  aggregate = NULL,
  save = FALSE,
  year.geo = NULL,
  implicitnull = NULL,
  row = NULL,
  base = NULL,
  parallel = deprecated(),
  raw = NULL,
  select = NULL
)

mf(
  group = NULL,
  koblid = NULL,
  aggregate = NULL,
  save = FALSE,
  year.geo = NULL,
  implicitnull = NULL,
  row = NULL,
  base = NULL,
  parallel = deprecated(),
  raw = NULL,
  select = NULL
)

Arguments

`group`	The name of filegroup as specified in filgruppe
`koblid`	`KOBLID` from table tbl_Koble
`aggregate`	Logical value. Default is `TRUE`. Aggregate data according to the specification in registration database. Global options with `orgdata.aggregate`.
`save`	Logical value. Default is `FALSE`. To save as `.csv` format file by activating `save_file()` function.
`year.geo`	Which reference year to use for geograhical coding. If it is missing then global option for `orgdata.year` will be used.
`implicitnull`	Logical value. Default is `TRUE` to add implicit null to the dataset. Global options with `orgdata.implicit.null`.
`row`	Select specific row(s) numbers only. Useful for debugging. Please read Debugging article for detail.
`base`	Logical value. If `TRUE` then use year in the original data as the base year to recode the geographical codes. Default is `FALSE` and use all available codes in geo codebook
`parallel`	Logical or numeric value. With logical value `TRUE` it will run with parallel using 50% ie. 0.5 of local cores. User can decide other percentage if needed. For example to use 75% of the cores then specify as `parallel = 0.75`. Nevertheless, maximum cores allowed is only 80%. Default value is `FALSE` ie. to use sequential processing
`raw`	Logical value. Default is `FALSE` as in config. If `TRUE` then read original raw data directly from source file even if the dataset is already available in DuckDB without the need to unmark `KONTROLLERT` in the Access database
`select`	Select number of valid files to process as an alternative to using `KOBLID`. To select the first 5 files then write `select=1:5`. Use `select="last"` to select the last or most recent file.

Examples

## Not run: 
dt <- make_file("ENPERSON")
dt <- make_file("ENPERSON", raw = TRUE) #Skip DuckDB and read directly from original files
dt <- make_file("ENPERSON", koblid = 120:125) #Select specific files only
dt <- make_file("ENPERSON", select = "last") #Select most recent file

## End(Not run)

helseprofil/orgdata documentation built on Feb. 3, 2025, 4:55 p.m.