mirror of
https://github.com/explosion/spaCy.git
synced 2024-11-16 06:37:04 +03:00
9 lines
257 B
Python
9 lines
257 B
Python
|
def test_tl_simple_punct(tl_tokenizer):
|
||
|
|
text = "Sige, punta ka dito"
|
||
|
|
tokens = tl_tokenizer(text)
|
||
|
|
assert tokens[0].idx == 0
|
||
|
|
assert tokens[1].idx == 4
|
||
|
|
assert tokens[2].idx == 6
|
||
|
|
assert tokens[3].idx == 12
|
||
|
|
assert tokens[4].idx == 15
|