Parent Linkage

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

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

1# Reference solution for the yamaa benchmark sdtm-suppmh-linkage (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)1011mh <- read_csv(12  "/app/input/mh.csv",13  col_types = cols(.default = col_character())14)15raw <- read_csv(16  "/app/input/mh_supp_raw.csv",17  col_types = cols(.default = col_character())18)1920# The parent sequence number, matched on subject and condition term.21parents <- mh |> select(STUDYID, USUBJID, MHSEQ, MHTERM)2223suppmh <- raw |>24  pivot_longer(25    cols = c(MHFAMHX, MHCONF),26    names_to = "QNAM",27    values_to = "QVAL"28  ) |>29  filter(!is.na(QVAL) & QVAL != "") |>30  inner_join(parents, by = c("STUDYID", "USUBJID", "MHTERM")) |>31  mutate(32    RDOMAIN = "MH",33    IDVAR = "MHSEQ",34    IDVARVAL = MHSEQ,35    QLABEL = case_when(36      QNAM == "MHFAMHX" ~ "Family History",37      QNAM == "MHCONF" ~ "Confirmed by Medical Records"38    ),39    QORIG = "Collected",40    QEVAL = NA_character_41  ) |>42  arrange(USUBJID, suppressWarnings(as.integer(MHSEQ)), QNAM) |>43  select(44    STUDYID, RDOMAIN, USUBJID, IDVAR, IDVARVAL, QNAM, QLABEL, QVAL, QORIG, QEVAL45  )4647dir.create("/app/output", showWarnings = FALSE)48write_csv(suppmh, "/app/output/suppmh.csv", na = "")

1# Reference solution for the yamaa benchmark sdtm-suppmh-linkage (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 SUPPMH from MH and MH_SUPP_RAW.89One SUPPMH record per collected qualifier on a condition. The parent MH10record is found by subject and condition term; its sequence number is11carried as text in IDVARVAL.12"""1314import csv15import os1617INPUT = "/app/input"18OUTPUT = "/app/output"1920QUALIFIERS = [21    ("MHFAMHX", "MHFAMHX", "Family History"),22    ("MHCONF", "MHCONF", "Confirmed by Medical Records"),23]2425COLUMNS = [26    "STUDYID", "RDOMAIN", "USUBJID", "IDVAR", "IDVARVAL",27    "QNAM", "QLABEL", "QVAL", "QORIG", "QEVAL",28]293031def read_csv(path):32    with open(path, newline="", encoding="utf-8") as handle:33        return list(csv.DictReader(handle))343536def main():37    mh = read_csv(os.path.join(INPUT, "mh.csv"))38    raw = read_csv(os.path.join(INPUT, "mh_supp_raw.csv"))3940    mhseq_by_key = {}41    for row in mh:42        key = (row["STUDYID"], row["USUBJID"], row["MHTERM"])43        mhseq_by_key[key] = row["MHSEQ"]4445    records = []46    for row in raw:47        key = (row["STUDYID"], row["USUBJID"], row["MHTERM"])48        mhseq = mhseq_by_key.get(key)49        if mhseq is None:50            continue51        for source, qnam, qlabel in QUALIFIERS:52            answer = (row.get(source) or "").strip()53            if not answer:54                continue55            records.append({56                "STUDYID": row["STUDYID"],57                "RDOMAIN": "MH",58                "USUBJID": row["USUBJID"],59                "IDVAR": "MHSEQ",60                "IDVARVAL": str(mhseq),61                "QNAM": qnam,62                "QLABEL": qlabel,63                "QVAL": answer,64                "QORIG": "Collected",65                "QEVAL": "",66            })6768    os.makedirs(OUTPUT, exist_ok=True)69    with open(os.path.join(OUTPUT, "suppmh.csv"), "w", newline="",70              encoding="utf-8") as handle:71        writer = csv.DictWriter(handle, fieldnames=COLUMNS)72        writer.writeheader()73        writer.writerows(records)747576if __name__ == "__main__":77    main()

1/* opensas reference solution for the yamaa benchmark sdtm-suppmh-linkage.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data mh;6  length STUDYID USUBJID MHTERM $1024;7  infile "/app/input/mh.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input STUDYID : $char1024. USUBJID : $char1024. MHSEQ MHTERM : $char1024.;9run;1011data mh_supp_raw;12  length STUDYID USUBJID MHTERM MHFAMHX MHCONF $1024;13  infile "/app/input/mh_supp_raw.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;14  input STUDYID : $char1024. USUBJID : $char1024. MHTERM : $char1024. MHFAMHX : $char1024. MHCONF : $char1024.;15run;1617proc sql; create table qualified as select a.*,b.MHSEQ from mh_supp_raw a inner join mh b on a.STUDYID=b.STUDYID and a.USUBJID=b.USUBJID and a.MHTERM=b.MHTERM; quit;18data result; set qualified; length RDOMAIN IDVAR IDVARVAL QNAM QLABEL QVAL QORIG QEVAL $1024; RDOMAIN='MH'; IDVAR='MHSEQ'; IDVARVAL=strip(put(MHSEQ,best32.)); QORIG='Collected'; QEVAL='';19if not missing(MHFAMHX) then do; QNAM='MHFAMHX'; QLABEL='Family History'; QVAL=MHFAMHX; output; end;20if not missing(MHCONF) then do; QNAM='MHCONF'; QLABEL='Confirmed by Medical Records'; QVAL=MHCONF; output; end; run;21proc sql;22  create table final as select STUDYID, RDOMAIN, USUBJID, IDVAR, IDVARVAL, QNAM, QLABEL, QVAL, QORIG, QEVAL from result;23quit;24proc export data=final outfile="/app/output/suppmh.csv" dbms=csv replace; run;