Expression Forms

Review ask

As a clinical trial statistician or programmer, challenge anything in this benchmark: describe a problem in Comments, or propose a fix with the Edit button on any source file.

Task: Would a real study need this derivation, specified this way? Does it follow CDISC ADaM: variable names, controlled terms, record structure? Flag any rule that is ambiguous or that a sponsor would handle differently.
Inputs: Do they look like data collected in a real trial, and do they include the situations you would expect to meet in one?
Expected output: Would independent QC programming reproduce every value from these inputs and rules?

The other tabs beside Comments are background for assessment, not under review.

Summary

EditUnder reviewWhat this benchmark means

README.md

Goal: pin how each column's value is written and read: fixed values, values copied by naming their source column, age bands, a population flag chosen by a condition, first-available country selection, joining text, and the largest and smallest of several collected weights.

Input: one spec.yaml. DM carries one record per subject: sex, age, first-dose date, country, site country, and two collected weights, any of which may be blank.

Columns:

STUDYID holds a fixed study code written directly in the column.
USUBJID and SEX copy the collected values by naming the source column alone.
AGEGRP places each age in a band: under 18, 18 to 64, or 65 and over. Each band includes its lower edge, so an age of exactly 65 is in the oldest band and an age of exactly 18 is in the middle band; a 12-year-old exercises the youngest band. A subject with no age carries the fallback text NOT REPORTED.
SAFFL flags the safety population: Y for a subject with a first-dose date, N for a subject without one.
COUNTRY takes the subject's own country, or the site country when the subject's own entry is blank. When both are blank it carries the declared fallback text.
SUBJLBL joins the fixed word Subject to the subject number.
WTMAX carries the larger of the two collected weights, skipping a blank one; with both blank it stays blank.
WTMIN carries the smaller of the two collected weights, skipping a blank one; with both blank it stays blank.

YAML specification

1 spec file

spec.yaml

Edit93 lines

1schema_version: "1.0"2domain: ADSL3keys: [USUBJID]4input:5  DM: {path: input/dm.csv, types: {AGE: int, WT_SCR: int, WT_BL: int}}6base: DM78output:9  path: adsl.csv10  columns: [STUDYID, USUBJID, SEX, AGEGRP, SAFFL, COUNTRY, SUBJLBL, WTMAX,11            WTMIN]1213columns:14  # A literal returns its declared value unchanged.15  - name: STUDYID16    type: str17    label: Study Identifier18    derivation: {literal: YAMAA-01}1920  # A bare string is the source shorthand: it desugars to21  # {source: DM.USUBJID} before registry dispatch.22  - name: USUBJID23    type: str24    label: Unique Subject Identifier25    derivation: DM.USUBJID2627  - name: SEX28    type: str29    label: Sex30    derivation: DM.SEX3132  # Cut assigns a numeric source to one labeled band. Bands are closed on33  # the left, so an age of exactly 65 lands in 65+ and an age of exactly34  # 18 lands in 18-64; a missing age takes the declared missing value.35  - name: AGEGRP36    type: str37    label: Age Group38    derivation:39      cut:40        source: DM.AGE41        breaks: [18, 65]42        labels: ['<18', '18-64', '65+']43        missing: NOT REPORTED4445  # Each then and the trailing otherwise nests an expression. The first46  # true branch wins; a subject with no first-dose date falls through to47  # otherwise.48  - name: SAFFL49    type: str50    label: Safety Population Flag51    derivation:52      case:53        - when: "DM.RFXSTDTC IS NOT NULL"54          then: {literal: Y}55        - otherwise: {literal: N}5657  # First_available returns the first non-missing source, or the declared58  # missing value when every source is missing.59  - name: COUNTRY60    type: str61    label: Country62    derivation:63      first_available:64        sources: [DM.COUNTRY, DM.SITECNTY]65        missing: UNKNOWN6667  # str_concat.sources nests expressions: literals sit beside sources.68  - name: SUBJLBL69    type: str70    label: Subject Label71    derivation:72      str_concat:73        sources:74          - literal: "Subject "75          - source: DM.SUBJID7677  # Greatest returns the largest non-missing source, or missing when none78  # exist. The sources must be mutually comparable.79  - name: WTMAX80    type: int81    label: Highest Recorded Weight82    derivation:83      greatest:84        sources: [DM.WT_SCR, DM.WT_BL]8586  # Least returns the smallest non-missing source, or missing when none87  # exist.88  - name: WTMIN89    type: int90    label: Lowest Recorded Weight91    derivation:92      least:93        sources: [DM.WT_SCR, DM.WT_BL]

Inputs

Under review1 source file

input/dm.csv

Edit6 rows

input/dm.csv
STUDYID	SUBJID	SEX	AGE	RFXSTDTC	COUNTRY	SITECNTY	WT_SCR	WT_BL
YAMAA-01	001	M	65	2026-01-12	USA	USA	70	72
YAMAA-01	002	F	40	2026-01-19	—	CAN	—	68
YAMAA-01	003	M	—	—	—	—	—	—
YAMAA-01	004	F	25	2026-02-02	GBR	GBR	55	55
YAMAA-01	005	F	18	2026-02-10	USA	—	80	75
YAMAA-01	006	M	12	—	GBR	GBR	45	45

Expected output

Under review6 expected rows

expected/adsl.csv

Edit6 rows

expected/adsl.csv
STUDYID	SEX	AGEGRP	SAFFL	COUNTRY	SUBJLBL	WTMAX	WTMIN
YAMAA-01	M	65+	Y	USA	Subject 001	72	70
YAMAA-01	F	18-64	Y	CAN	Subject 002	68	68
YAMAA-01	M	NOT REPORTED	N	UNKNOWN	Subject 003	—	—
YAMAA-01	F	18-64	Y	GBR	Subject 004	55	55
YAMAA-01	F	18-64	Y	USA	Subject 005	80	75
YAMAA-01	M	<18	N	GBR	Subject 006	45	45

Shaded columns are derived— = missing value

Comments and assessment materials

Dataset Name	Variable Name	Variable Label	Type	Computational Method	Origin
ADSL	STUDYID	Study Identifier	character	Set constant value "YAMAA-01".	Assigned
ADSL	USUBJID	Unique Subject Identifier	character	Copy value from DM.USUBJID.	Predecessor
ADSL	SEX	Sex	character	Copy value from DM.SEX.	Predecessor
ADSL	AGEGRP	Age Group	character	Band DM.AGE at 18, 65 into "<18", "18-64" and "65+", intervals closed on the left; a missing source → "NOT REPORTED".	Derived
ADSL	SAFFL	Safety Population Flag	character	If DM.RFXSTDTC IS NOT NULL then "Y"; otherwise "N".	Derived
ADSL	COUNTRY	Country	character	First non-missing of DM.COUNTRY and DM.SITECNTY; all missing → "UNKNOWN".	Derived
ADSL	SUBJLBL	Subject Label	character	Concatenate "Subject " and DM.SUBJID.	Derived
ADSL	WTMAX	Highest Recorded Weight	numeric	Greatest of DM.WT_SCR and DM.WT_BL.	Derived
ADSL	WTMIN	Lowest Recorded Weight	numeric	Least of DM.WT_SCR and DM.WT_BL.	Derived

Version	Date	Author	Description
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

run.R

Edit4 lines

1library(yamaa)23adsl <- yamaa_domain("spec.yaml")$output4adsl

run.py

Edit4 lines

1import yamaa23adsl = yamaa.yamaa_domain("spec.yaml").output4adsl

input/dm.csv

expected/adsl.csv

Comments

Mapping spec

yamaa code

run.R

run.py