Report Repeated Exposure Records and Read the Latest

Review ask

As a clinical trial statistician or programmer, answer two short questions in Your review: the task and the inputs and outputs. Agreeing with both approves this benchmark; disagreeing with either needs a comment. Propose a fix with the Edit button on any source file, or add it to your comments.

We suggest spending no more than 5 minutes on a review. The other tabs beside Your review are background for assessment, not under review.

Summary

EditUnder reviewWhat this benchmark means

README.md

Public entries: Python and R. Passing coverage requires an installed inventory.

Goal: derive each subject's date of last exposure (TRTEDT) from the exposure records, reporting the subjects with more than one record instead of stopping the run.

Input: one spec.yaml declaring two datasets. DM carries one row per subject. EX carries exposure records: subject, sequence number, start date and end date.

Variables:

TRTEDT is the end date of the subject's exposure record that starts last, the later sequence number breaking a tie; a subject with one record takes that record's end date. It is blank when the subject has no exposure record or the selected record has no end date.

Lookups: LAST_EX names each subject's latest exposure record. The subject is expected to appear once in the exposure records; one that appears more often is reported, and the latest of its records is read.

The run report (adsl-checks.csv) has one row per item the run handled, each naming the dataset (ARTIFACT), the kind of item (KIND), where the spec states it (SPEC_PATH) and the item's name (NAME, identifying an input, row source, reported event, column, or check): the run and its outcome (OUTCOME), the input datasets read, the subjects built, then the two counts of the exposure read: the subjects for which the latest record was chosen among several, and those for which no record was found. Each count names the column it feeds (TARGET) and the rule its answer stands in for (RULE). One row per output column follows, with its missing values and longest text, then the one check, which carries its label (ID), its RULE and SEVERITY, the condition it found (CONDITION), and how many subjects repeat out of how many subjects the exposure records hold (COUNT, TOTAL). Its KEYS list each repeated subject once, in the order its first exposure record appears, and its DETAILS name the subject columns, the exposure dataset and the lookup.

Note: the repeated subjects are reported before any row is built and change nothing that is read: every subject still takes its latest record. Without the order and the choice of the last record, a repeated subject would have no record to read, so the check could not warn and would have to stop the run.

YAML specification

1 spec file

spec.yaml

Edit48 lines

1yamaa: "1.0"2type: spec3model: ADaM4domain: ADSL5keys: [STUDYID, USUBJID]6input:7  DM: input/dm.csv8  EX: input/ex.csv9base: DM1011output:12  path: adsl.csv13  columns: [STUDYID, USUBJID, TRTEDT]14  # The warning's line in the run report lists the subjects with several15  # exposure records, which needs a report file.16  report: adsl-checks.csv1718intermediates:19  # Each subject's latest exposure record. The subject key is expected to20  # be unique across the exposure records, but a subject with several is21  # reported rather than refused, and the record that starts last is read:22  # the check warns only because the order and keep choose one record.23  - id: LAST_EX24    dataset: EX25    key: [STUDYID, USUBJID]26    order_by: [EX.EXSTDTC, EX.EXSEQ]27    keep: last28    verifications:29      - unique:30          id: one-exposure-record31          columns: [STUDYID, USUBJID]32          severity: warning3334columns:35  - name: STUDYID36    type: str37    label: Study Identifier38    derivation: DM.STUDYID3940  - name: USUBJID41    type: str42    label: Unique Subject Identifier43    derivation: DM.USUBJID4445  - name: TRTEDT46    type: date47    label: Date of Last Exposure to Treatment48    derivation: LAST_EX.EXENDTC

Inputs

Under review2 source files

input/dm.csv

Edit3 rows

input/dm.csv
STUDYID	USUBJID
YAMAA-01
YAMAA-01
YAMAA-01

input/ex.csv

Edit5 rows

input/ex.csv
STUDYID	EXSEQ	EXSTDTC	EXENDTC
YAMAA-01	1	2024-01-02	2024-01-30
YAMAA-01	1	2024-01-04	2024-01-31
YAMAA-01	1	2024-01-05	2024-02-03
YAMAA-01	2	2024-02-01	2024-02-28
YAMAA-01	2	2024-02-04	2024-03-01

Expected output

Under review13 expected rows

expected/adsl-checks.csv

Edit10 rows

expected/adsl-checks.csv
ARTIFACT	KIND	SPEC_PATH	NAME	ID	TARGET	RULE	SEVERITY	OUTCOME	CONDITION	COUNT	TOTAL	KEYS	DETAILS
adsl.csv	run	output	—	—	—	—	—	succeeded	—	3	—	—	{}
adsl.csv	input	input.DM	DM	—	—	—	—	—	—	3	—	—	{"bytes":55,"path":"input/dm.csv","profile":"csv"}
adsl.csv	input	input.EX	EX	—	—	—	—	—	—	5	—	—	{"bytes":223,"path":"input/ex.csv","profile":"csv"}
adsl.csv	rows	base	DM	—	—	—	—	—	—	3	3	—	{}
adsl.csv	handler	columns.TRTEDT.derivation.source.multiple_matches	multiple_matches	—	TRTEDT	LOOKUP-10	—	—	—	2	—	—	{}
adsl.csv	handler	columns.TRTEDT.derivation.source.absent	absent	—	TRTEDT	LOOKUP-24	—	—	—	0	—	—	{}
adsl.csv	column	columns.STUDYID	STUDYID	—	—	—	—	—	—	0	3	—	{"max_length":8}
adsl.csv	column	columns.USUBJID	USUBJID	—	—	—	—	—	—	0	3	—	{"max_length":3}
adsl.csv	column	columns.TRTEDT	TRTEDT	—	—	—	—	—	—	0	3	—	{}
adsl.csv	check	intermediates.LAST_EX.verifications[0].unique	unique	one-exposure-record	—	LOOKUP-17	warning	violated	duplicate_intermediate_records	2	3	[{"STUDYID":"YAMAA-01","USUBJID":"P03"},{"STUDYID":"YAMAA-01","USUBJID":"P02"}]	{"columns":["STUDYID","USUBJID"],"dataset":"EX","intermediate":"LAST_EX"}

expected/adsl.csv

Edit3 rows

expected/adsl.csv
STUDYID	USUBJID	TRTEDT
YAMAA-01		2024-01-30
YAMAA-01		2024-03-01
YAMAA-01		2024-02-28

Shaded columns are derived— = missing value

Human review and assessment

Dataset Name	Variable Name	Variable Label	Type	Codelist/Controlled Terms	Computational Method	Origin
ADSL	STUDYID	Study Identifier	character		Copy value from DM.STUDYID.	Predecessor
ADSL	USUBJID	Unique Subject Identifier	character		Copy value from DM.USUBJID.	Predecessor
ADSL	TRTEDT	Date of Last Exposure to Treatment	numeric	ISO 8601	Take EXENDTC from the last EX record matching on STUDYID and USUBJID, ordered by EX.EXSTDTC then EX.EXSEQ, and no match yields blank.	Derived

Version	Date	Author	Description
1.7	2026-10-20	yamaa docs build	Read a specification in the 1.0 surface as its 0.5 counterpart; state that missing values stay missing where a mapping declares no missing handler, and leave out a declared missing answer that restates a handler's default
1.6	2026-10-17	yamaa docs build	Describe collected date parts written as ISO 8601 date text and times of day counted in seconds since midnight
1.5	2026-10-17	yamaa docs build	Name collected precision in words instead of letter codes; state imputation fallbacks, minimum source precision, bounds ignored outside the collected period, and invalid text with no handler as an error; describe date and datetime shifts, imputation flag codes, and partial text giving no date part
1.4	2026-10-15	yamaa docs build	State datetime precision codes through the hour and minute, and complete only the time parts a collected datetime lacks
1.3	2026-09-29	yamaa docs build	Show study codelist names, external dictionaries, and ISO 8601 formats in variable rows; list the document's codelist values and metadata
1.2	2026-09-29	yamaa docs build	Classify ADaM copies from a declared input as Predecessor and recodes as Derived, and SDTM recodes of locally derived columns as Derived
1.1	2026-09-29	yamaa docs build	Resolve lookups into the donor record they select, and give every derivation shape a plain-language rule instead of its raw YAML
1.0		yamaa docs build	Initial generation from spec.yaml

R

Edit3 lines

1library(yamaa)23yamaa_spec("spec.yaml")$run()$output

Python

Edit3 lines

1import yamaa23yamaa.spec("spec.yaml").run().output

input/dm.csv

input/ex.csv

expected/adsl-checks.csv

expected/adsl.csv

Your review

Mapping spec

yamaa code

R

Python