def test_serialize_custom_tokenizer(en_vocab, en_tokenizer):
"""Test that custom tokenizer with not all functions defined or empty
properties can be serialized and deserialized correctly (see #2494,
#4991)."""
tokenizer = Tokenizer(en_vocab, suffix_search=en_tokenizer.suffix_search)
tokenizer_bytes = tokenizer.to_bytes()
Tokenizer(en_vocab).from_bytes(tokenizer_bytes)
# test that empty/unset values are set correctly on deserialization
tokenizer = get_lang_class("en")().tokenizer
tokenizer.token_match = re.compile("test").match
assert tokenizer.rules != {}
assert tokenizer.token_match is not None
assert tokenizer.url_match is not None
tokenizer.from_bytes(tokenizer_bytes)
assert tokenizer.rules == {}
assert tokenizer.token_match is None
assert tokenizer.url_match is None
tokenizer = Tokenizer(en_vocab,
rules={"ABC.": [{
"ORTH": "ABC"
}, {
"ORTH": "."
}]})
tokenizer.rules = {}
tokenizer_bytes = tokenizer.to_bytes()
tokenizer_reloaded = Tokenizer(en_vocab).from_bytes(tokenizer_bytes)
assert tokenizer_reloaded.rules == {}
def custom_tokenizer(nlp):
infix_re = re.compile(r'''[?;‘’`“”"'~]''')
prefix_re = compile_prefix_regex(nlp.Defaults.prefixes)
suffix_re = compile_suffix_regex(nlp.Defaults.suffixes)
tokenizer = Tokenizer(nlp.vocab)
tokenizer.prefix_search = prefix_re.search
tokenizer.suffix_search = suffix_re.search
tokenizer.infix_finditer = infix_re.finditer
tokenizer.token_match = None
return tokenizer
