Age From Birthdate

1library(yamaa)23dm <- yamaa_domain("spec.yaml")$output4dm

1import yamaa23dm = yamaa.yamaa_domain("spec.yaml").output4dm

1# Reference solution for the yamaa benchmark sdtm-dm-age (R track).2#3# Written from the benchmark's full prompt and its inputs alone, as an agent4# would write it. Harbor's oracle agent runs it to check that the prompt can5# be solved and that the grader scores a correct answer 1.67library(dplyr, warn.conflicts = FALSE)8library(readr)9library(tidyr)10library(stringr)1112odm <- read_csv(13  "/app/input/odm.csv",14  col_types = cols(.default = col_character())15)1617is_leap <- function(year) {18  year %% 4 == 0 & (year %% 100 != 0 | year %% 400 == 0)19}2021whole_years <- function(birth, consent) {22  full <- str_detect(birth, "^\\d{4}-\\d{2}-\\d{2}$") &23    str_detect(consent, "^\\d{4}-\\d{2}-\\d{2}$")24  by <- as.integer(str_sub(birth, 1, 4))25  bm <- as.integer(str_sub(birth, 6, 7))26  bd <- as.integer(str_sub(birth, 9, 10))27  cy <- as.integer(str_sub(consent, 1, 4))28  cm <- as.integer(str_sub(consent, 6, 7))29  cd <- as.integer(str_sub(consent, 9, 10))30  # A February 29 birthday has its anniversary on February 28 in a31  # common year. A year-month or year-only birth date has no age.32  ann_m <- if_else(bm == 2 & bd == 29 & !is_leap(cy), 2L, bm)33  ann_d <- if_else(bm == 2 & bd == 29 & !is_leap(cy), 28L, bd)34  age <- cy - by - if_else(35    cm < ann_m | (cm == ann_m & cd < ann_d), 1L, 0L36  )37  if_else(full, age, NA_integer_)38}3940# Whole calendar years between the birth date and the consent date. A41# missing birth date leaves the birth date, age, and units empty.42dm <- odm |>43  filter(ItemGroupOID == "IG.DM") |>44  pivot_wider(45    id_cols = c(StudyOID, SubjectKey),46    names_from = ItemOID,47    values_from = Value,48    values_fn = first49  ) |>50  mutate(51    DOMAIN = "DM",52    STUDYID = StudyOID,53    USUBJID = SubjectKey,54    RFICDTC = na_if(`IT.DM.RFICDTC`, ""),55    BRTHDTC = na_if(`IT.DM.BRTHDT`, ""),56    AGE = whole_years(BRTHDTC, RFICDTC),57    AGEU = if_else(is.na(AGE), NA_character_, "YEARS")58  ) |>59  arrange(USUBJID) |>60  select(DOMAIN, STUDYID, USUBJID, RFICDTC, BRTHDTC, AGE, AGEU)6162dir.create("/app/output", showWarnings = FALSE)63write_csv(dm, "/app/output/dm.csv", na = "")

1# Reference solution for the yamaa benchmark sdtm-dm-age (Python track).2#3# Written from the benchmark's full prompt and its inputs alone, as an agent4# would write it. Harbor's oracle agent runs it to check that the prompt can5# be solved and that the grader scores a correct answer 1.67import re8from datetime import date9from pathlib import Path1011import polars as pl1213odm = pl.read_csv("/app/input/odm.csv", infer_schema=False)1415FULL = re.compile(r"^\d{4}-\d{2}-\d{2}$")161718def is_leap(year: int) -> bool:19    return year % 4 == 0 and (year % 100 != 0 or year % 400 == 0)202122def whole_years(birth: str, consent: str) -> int | None:23    # Whole calendar years between birth and consent; a year-month or24    # year-only birth date cannot give completed years.25    if not FULL.match(birth or "") or not FULL.match(consent or ""):26        return None27    by, bm, bd = int(birth[0:4]), int(birth[5:7]), int(birth[8:10])28    cy, cm, cd = int(consent[0:4]), int(consent[5:7]), int(consent[8:10])29    try:30        date(by, bm, bd)31        date(cy, cm, cd)32    except ValueError:33        return None34    # A February 29 birthday has its anniversary on February 28 in a35    # common year.36    if (bm, bd) == (2, 29) and not is_leap(cy):37        anniversary = (2, 28)38    else:39        anniversary = (bm, bd)40    return cy - by - (1 if (cm, cd) < anniversary else 0)414243wide = odm.filter(pl.col("ItemGroupOID") == "IG.DM").pivot(44    index=["StudyOID", "SubjectKey"],45    on="ItemOID",46    values="Value",47    aggregate_function="first",48)4950rows = []51for record in wide.to_dicts():52    rfic = record.get("IT.DM.RFICDTC") or None53    brth = record.get("IT.DM.BRTHDT") or None54    # A missing birth date leaves the birth date, age, and units empty.55    if not brth:56        brth, age, ageu = None, None, None57    else:58        age = whole_years(brth, rfic or "")59        ageu = "YEARS" if age is not None else None60    rows.append(61        {62            "DOMAIN": "DM",63            "STUDYID": record["StudyOID"],64            "USUBJID": record["SubjectKey"],65            "RFICDTC": rfic,66            "BRTHDTC": brth,67            "AGE": age,68            "AGEU": ageu,69        }70    )7172dm = pl.DataFrame(73    rows,74    schema={75        "DOMAIN": pl.String,76        "STUDYID": pl.String,77        "USUBJID": pl.String,78        "RFICDTC": pl.String,79        "BRTHDTC": pl.String,80        "AGE": pl.Int64,81        "AGEU": pl.String,82    },83).sort("USUBJID")8485Path("/app/output").mkdir(exist_ok=True)86dm.write_csv("/app/output/dm.csv")

1/* opensas reference solution for the yamaa benchmark sdtm-dm-age.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data odm;6  length StudyOID MetaDataVersionOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupOID ItemGroupRepeatKey ItemOID Value $1024;7  infile "/app/input/odm.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input StudyOID : $char1024. MetaDataVersionOID : $char1024. SubjectKey : $char1024. StudyEventOID : $char1024. StudyEventRepeatKey : $char1024. ItemGroupOID : $char1024. ItemGroupRepeatKey : $char1024. ItemOID : $char1024. Value : $char1024.;9run;1011proc sort data=odm out=ordered; by StudyOID SubjectKey; run;12data wide;13  length BRTHDTC RFICDTC $1024;14  retain BRTHDTC RFICDTC;15  set ordered;1617  by StudyOID SubjectKey;18  if first.SubjectKey then do;19    BRTHDTC=''; RFICDTC='';20  end;21    if ItemOID='IT.DM.BRTHDT' then BRTHDTC=Value;22  if ItemOID='IT.DM.RFICDTC' then RFICDTC=Value;23  if last.SubjectKey then output;24run;25data result; set wide; length DOMAIN $2 STUDYID USUBJID $1024 AGEU $8; DOMAIN='DM'; STUDYID=StudyOID; USUBJID=SubjectKey;26if lengthn(BRTHDTC)=10 and lengthn(RFICDTC)=10 then do; AGE=intck('year',input(BRTHDTC,yymmdd10.),input(RFICDTC,yymmdd10.),'continuous'); AGEU='YEARS'; end; run;27proc sql;28  create table final as select DOMAIN, STUDYID, USUBJID, RFICDTC, BRTHDTC, AGE, AGEU from result;29quit;30proc export data=final outfile="/app/output/dm.csv" dbms=csv replace; run;