Calculate new languages for new MMTEB task

find_n_new_language.py

from mteb.tasks.BitextMining import *

new_task = "BibleNLPBitextMining"

### Gather existing BitextMining tasks
ignore = ["AbsTaskBitextMining", "Any", "CrosslingualTask", "TaskMetadata", "TEST_SAMPLES"]
ignore.append(new_task)
existing_tasks = [
    task for task in dir() if task[0].isupper() and task not in ignore
]

### Find all languages currently in BitextMining
existing_languages = []
for task in existing_tasks:
    existing_languages.extend(list(globals()[task]().languages))
existing_languages = set(existing_languages)

### Find new languages introduced by "BibleNLPBitextMining"
new_languages = []
new_task_languages = globals()[new_task]().languages
for lang in sorted(new_task_languages):
    if not lang in existing_languages:
        new_languages.append(lang)

### Report results
for idx, lang in enumerate(new_languages):
    print(f"New language {idx+1}: {lang}")

print(f"Total new languages: {len(new_languages)}")