Checks what ARS requires and what the reporting event refers to: each
required field is there (an analysis's purpose above all: it is the
SAP's decision, so tfl_ars() leaves it blank when the spec does), each
id is used once in its class, and each reference (methodId,
analysisSetId, dataSubsetId, groupingId, an operation, an output or
analysis of the list of contents) names something that is there. With
schema = TRUE and the jsonvalidate package installed, the JSON is
also checked against CDISC's JSON Schema for ARS v1.0
(system.file("ars", "ars_ldm.json", package = "tflspec")).
Arguments
- ars
A
tfl_ars(), or the path of an ARS.jsonfile.- schema
Check against the JSON Schema too (needs jsonvalidate; skipped with a message when it is not installed).
- profile
"siera"checks also what siera needs to run it (three analyses an output, one analysis set an output and of one condition, a code template for each method, ids of letters, digits and_). Default: the profile thetfl_ars()was written with, else"cdisc".
Value
A data frame, one row per problem (none: zero rows): part (the
class and id), field and problem.
Examples
f <- system.file("ars", "ars_ldm.json", package = "tflspec")
file.exists(f)
#> [1] TRUE