# Install: brew install duckdb (macOS) or see duckdb.org/docs/installation

cd /path/to/athena_download

for f in CONCEPT CONCEPT_ANCESTOR CONCEPT_RELATIONSHIP CONCEPT_SYNONYM \
         RELATIONSHIP VOCABULARY DOMAIN CONCEPT_CLASS DRUG_STRENGTH; do
  [ -f "$f.csv" ] && duckdb -c \
    "COPY (SELECT * FROM read_csv('$f.csv', delim='\t', header=true, quote='')) \
     TO '$f.parquet' (FORMAT PARQUET);"
done

# pip install duckdb

cd /path/to/athena_download

python3 -c "
import duckdb, os
for f in ['CONCEPT','CONCEPT_ANCESTOR','CONCEPT_RELATIONSHIP','CONCEPT_SYNONYM',
          'RELATIONSHIP','VOCABULARY','DOMAIN','CONCEPT_CLASS','DRUG_STRENGTH']:
    if os.path.exists(f+'.csv'):
        duckdb.sql(\"COPY (SELECT * FROM read_csv('\"+f+\".csv', delim='\t', header=true, quote='')) TO '\"+f+\".parquet' (FORMAT PARQUET)\")
        print(f'  {f}.csv -> {f}.parquet')
"