A cli tool to get sense data for a lemma and then find verses in which it occur using the macula-greek repo

get_lemma_senses.py

from collections import defaultdict
import sys
import csv


filename = 'macula-greek-SBLGNT.tsv'


lemmas_domain = defaultdict(set)
lemmas_senses = defaultdict(set)


with open(filename, 'r') as csvfile:
    reader = csv.reader(csvfile, delimiter='\t')
    for row in reader:
        if '!' not in row[1]:
            continue
        verse = row[1].split('!', maxsplit=1)[0]
        lemma = row[10]
        domain = row[22].split(' ')
        senses = row[23].split(' ')
        for x in domain:
            lemmas_domain[(lemma, x)].add(verse)
        for x in senses:
            lemmas_senses[(lemma, x)].add(verse)

lemma = sys.argv[1]
sense = sys.argv[2]


def stack_results(res):
    out = {}
    for r in res:
        book, cv = r.split(' ', maxsplit=1)
        if book in out:
            out[book].append(cv)
        else:
            out[book] = [cv]
    output = []
    for book, cvs in out.items():
        output.append(book + ' ' + ','.join(cvs))
    return ';'.join(output)


if sense == '*':
    print("displaying senses for " + lemma)
    senses = defaultdict(set)
    domains = defaultdict(set)
    for (l, x) in lemmas_senses.keys():
        senses[l].add(x)
    if lemma in senses:
        print(' '.join(senses[lemma]))
    exit()
elif ' ' in sense:
    verses = defaultdict(dict)
    for (l, x), vs in lemmas_senses.items():
        if sense in vs and l == lemma:
            print(f"{lemma}: {x}")
else:
    print(stack_results(lemmas_senses[(lemma, sense)]))