import pytest

import stanza
import stanza.resources.prepare_resources as prepare_resources

from stanza.tests import *

pytestmark = [pytest.mark.travis, pytest.mark.pipeline]

def test_split_model_name():
    # Basic test
    lang, package, processor = prepare_resources.split_model_name('ro_nonstandard_tagger.pt')
    assert lang == 'ro'
    assert package == 'nonstandard'
    assert processor == 'pos'

    # Check that nertagger is found even though it also ends with tagger
    # Check that ncbi_disease is correctly partitioned despite the extra _
    lang, package, processor = prepare_resources.split_model_name('en_ncbi_disease_nertagger.pt')
    assert lang == 'en'
    assert package == 'ncbi_disease'
    assert processor == 'ner'

    # assert that processors with _ in them are also okay
    lang, package, processor = prepare_resources.split_model_name('en_pubmed_forward_charlm.pt')
    assert lang == 'en'
    assert package == 'pubmed'
    assert processor == 'forward_charlm'