For each file a scan read and flagged, writes a copy beside it with the
flagged fields redacted by pii_redact(): survey.dta becomes
survey_redacted.dta. Untouched columns are written as they were read;
Stata and SPSS variable and value labels are kept. In text files, such as
logs and scripts, matched email addresses, phone numbers and IP addresses
are replaced wherever they occur.
Usage
pii_redact_files(
path,
findings = pii_scan(path, config = config),
method = "placeholder",
key = NULL,
min_confidence = NULL,
suffix = "_redacted",
overwrite = FALSE,
config = pii_config()
)Arguments
- path
Files or directories, as for
pii_scan().- findings
The result of
pii_scan()onpath. Scanned if missing.- method
One method for every flagged column, or a named vector giving the method per finding type, with an optional
default.- key
A secret string that determines the pseudonyms; needed only for
"pseudonym". Keep it out of the shared files; use the same key to redact files that must still join.- min_confidence
The weakest findings to act on; the configuration's
min_confidence, or"medium", unless given.- suffix
Added to a file's name to name its redacted copy.
- overwrite
Whether to replace a redacted copy that already exists.
- config
The project's configuration, from
pii_config(); used for the scan whenfindingsis missing.
Value
A tibble with one row per file: source, status, the
destination written, and in detail what was redacted and what was
left.
Details
Every file is accounted for in the result. A file can be redacted,
partly redacted (something flagged could not be removed, such as a name
inside free text), clean (nothing to redact; no copy written),
unsupported (the format cannot be written back), skipped (it is itself a
redacted copy), not read, or failed.
Delimited text, Excel, Stata, SPSS (sav, zsav), SAS transport (xpt), R data and Parquet files can be written back, as can plain text. An Excel file is written as xlsx without its formatting or formulas and needs the 'writexl' package. SAS data files, SPSS portable files, JSON and documents are reported as unsupported.
The originals are not changed or moved. Until they are removed or kept out of version control, the personal data are still in the project.
Examples
dir <- tempfile()
dir.create(dir)
survey <- data.frame(
id = 1:20,
q12 = paste0("person", 1:20, "@example.org"),
village = rep(c("Palpur", "Netrad"), 10)
)
write.csv(survey, file.path(dir, "survey.csv"), row.names = FALSE)
writeLines("Call 98765 43210 after six.", file.path(dir, "notes.txt"))
report <- pii_redact_files(dir, method = c(email = "pseudonym", default = "drop"), key = "s3")
#> Wrote 2 redacted copies.
#> ℹ The originals are unchanged and still hold the personal data.
report
#> # A tibble: 2 × 4
#> source status destination detail
#> <chr> <chr> <chr> <chr>
#> 1 /tmp/RtmpeoiXV6/file1d797591ce43/notes.txt redacted /tmp/RtmpeoiXV6/f… repla…
#> 2 /tmp/RtmpeoiXV6/file1d797591ce43/survey.csv redacted /tmp/RtmpeoiXV6/f… redac…
read.csv(file.path(dir, "survey_redacted.csv"))[1:3, ]
#> id q12
#> 1 1 grufwb9@wmfxars.wuv
#> 2 2 lliqur6@mieytmk.bse
#> 3 3 lhqjro8@pfspxjo.cru