Text Cleanup

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

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

1# Reference solution for the yamaa benchmark adam-adae-text-cleanup (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(stringr)1011ae <- read_csv(12  "/app/input/ae.csv",13  col_types = cols(AESEQ = col_integer(), .default = col_character())14)1516adae <- ae |>17  mutate(18    # AE, a hyphen, and exactly three digits; missing is 0, any other shape19    # is -1, and the digits give the number.20    AEREFNUM = case_when(21      is.na(AESPID) | AESPID == "" ~ 0L,22      str_detect(AESPID, "^AE-[0-9]{3}$") ~ suppressWarnings(as.integer(str_sub(AESPID, 4, 6))),23      .default = -1L24    ),25    AETERMLO = tolower(AETERM),26    AERELLC = if_else(is.na(AEREL) | AEREL == "", "not reported", tolower(AEREL)),27    AREL = toupper(AERELLC)28  ) |>29  select(30    STUDYID, USUBJID, AESEQ, AESPID, AEREFNUM, AETERM, AETERMLO, AERELLC, AREL31  )3233dir.create("/app/output", showWarnings = FALSE)34write_csv(adae, "/app/output/adae.csv", na = "")

1# Reference solution for the yamaa benchmark adam-adae-text-cleanup (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 csv8import re910INPUT = "/app/input/ae.csv"11OUTPUT = "/app/output/adae.csv"1213COLUMNS = [14    "STUDYID",15    "USUBJID",16    "AESEQ",17    "AESPID",18    "AEREFNUM",19    "AETERM",20    "AETERMLO",21    "AERELLC",22    "AREL",23]2425REFNUM_PATTERN = re.compile(r"^AE-(\d{3})$")262728def aerefnum(aespid):29    if aespid is None or aespid == "":30        return 031    match = REFNUM_PATTERN.match(aespid)32    if match:33        return int(match.group(1))34    return -1353637def main():38    with open(INPUT, newline="", encoding="utf-8") as handle:39        rows = list(csv.DictReader(handle))4041    out_rows = []42    for row in rows:43        aespid = row.get("AESPID") or ""44        aeterm = row.get("AETERM") or ""45        aerel = row.get("AEREL") or ""46        aerellc = aerel.lower() if aerel != "" else "not reported"47        out_rows.append(48            {49                "STUDYID": row.get("STUDYID") or "",50                "USUBJID": row.get("USUBJID") or "",51                "AESEQ": row.get("AESEQ") or "",52                "AESPID": aespid,53                "AEREFNUM": aerefnum(aespid),54                "AETERM": aeterm,55                "AETERMLO": aeterm.lower(),56                "AERELLC": aerellc,57                "AREL": aerellc.upper(),58            }59        )6061    with open(OUTPUT, "w", newline="", encoding="utf-8") as handle:62        writer = csv.DictWriter(handle, fieldnames=COLUMNS)63        writer.writeheader()64        writer.writerows(out_rows)656667if __name__ == "__main__":68    main()

1/* opensas reference solution for the yamaa benchmark adam-adae-text-cleanup.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data ae;6  length STUDYID USUBJID AESPID AETERM AEREL $1024;7  infile "/app/input/ae.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input STUDYID : $char1024. USUBJID : $char1024. AESEQ AESPID : $char1024. AETERM : $char1024. AEREL : $char1024.;9run;1011data result;12  length AETERMLO AERELLC AREL $1024;13  set ae;14  if missing(AESPID) then AEREFNUM=0;15  else if prxmatch('/^AE-[0-9]{3}$/',strip(AESPID)) then AEREFNUM=input(substr(AESPID,4,3),best32.);16  else AEREFNUM=-1;17  AETERMLO=lowcase(AETERM); AERELLC=lowcase(AEREL);18  if missing(AEREL) then AERELLC='not reported';19  AREL=upcase(AERELLC);20run;2122proc sql;23  create table final as select STUDYID, USUBJID, AESEQ, AESPID, AEREFNUM, AETERM, AETERMLO, AERELLC, AREL from result;24quit;25proc export data=final outfile="/app/output/adae.csv" dbms=csv replace; run;