Age From Birthdate

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

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

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 1.67"""Build the SDTM DM dataset (one record per subject) from the ODM extract.89Reads /app/input/odm.csv (long format: one row per collected item) and10writes /app/output/dm.csv with columns11DOMAIN, STUDYID, USUBJID, RFICDTC, BRTHDTC, AGE, AGEU.12"""1314import csv15import os16from calendar import isleap1718INPUT_DIR = "/app/input"19OUTPUT_DIR = "/app/output"2021BIRTH_ITEM = "IT.DM.BRTHDT"22CONSENT_ITEM = "IT.DM.RFICDTC"232425def parse_full_date(text):26    """Return (year, month, day) if text is a full ISO date, else None."""27    if not text:28        return None29    parts = text.strip().split("-")30    if len(parts) != 3:31        return None32    try:33        year, month, day = (int(p) for p in parts)34    except ValueError:35        return None36    return year, month, day373839def completed_years(birth, consent):40    """Whole calendar years between birth and consent dates.4142    One year fewer when the birthday falls after the consent date's43    month and day. A February 29 birthday has its anniversary on44    February 28 in a common year.45    """46    by, bm, bd = birth47    cy, cm, cd = consent48    if (bm, bd) == (2, 29) and not isleap(cy):49        anniversary = (2, 28)50    else:51        anniversary = (bm, bd)52    years = cy - by53    if anniversary > (cm, cd):54        years -= 155    return years565758def main():59    with open(os.path.join(INPUT_DIR, "odm.csv"), newline="", encoding="utf-8") as f:60        rows = list(csv.DictReader(f))6162    subjects = {}63    order = []64    for row in rows:65        key = (row["StudyOID"], row["SubjectKey"])66        if key not in subjects:67            subjects[key] = {"RFICDTC": "", "BRTHDTC": ""}68            order.append(key)69        item = row["ItemOID"]70        value = (row["Value"] or "").strip()71        if item == CONSENT_ITEM and not subjects[key]["RFICDTC"]:72            subjects[key]["RFICDTC"] = value73        elif item == BIRTH_ITEM and not subjects[key]["BRTHDTC"]:74            subjects[key]["BRTHDTC"] = value7576    out_rows = []77    for study_oid, subject_key in order:78        vals = subjects[(study_oid, subject_key)]79        rficdtc = vals["RFICDTC"]80        brthdtc = vals["BRTHDTC"]8182        age = ""83        birth = parse_full_date(brthdtc)84        consent = parse_full_date(rficdtc)85        if birth is not None and consent is not None:86            age = str(completed_years(birth, consent))87        ageu = "YEARS" if age else ""8889        out_rows.append(90            {91                "DOMAIN": "DM",92                "STUDYID": study_oid,93                "USUBJID": subject_key,94                "RFICDTC": rficdtc,95                "BRTHDTC": brthdtc,96                "AGE": age,97                "AGEU": ageu,98            }99        )100101    os.makedirs(OUTPUT_DIR, exist_ok=True)102    fieldnames = ["DOMAIN", "STUDYID", "USUBJID", "RFICDTC", "BRTHDTC", "AGE", "AGEU"]103    with open(os.path.join(OUTPUT_DIR, "dm.csv"), "w", newline="", encoding="utf-8") as f:104        writer = csv.DictWriter(f, fieldnames=fieldnames)105        writer.writeheader()106        writer.writerows(out_rows)107108109if __name__ == "__main__":110    main()

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;