Reformat using black

This commit is contained in:
Deniz Şafak
2026-01-09 01:36:14 +03:00
parent 3c800df450
commit 5ae153f841
48 changed files with 1530 additions and 894 deletions
+30 -12
View File
@@ -16,14 +16,22 @@ from abogen.normalization_settings import (
from abogen.spacy_contraction_resolver import resolve_ambiguous_contractions
SPACY_RESOLVER_AVAILABLE = bool(resolve_ambiguous_contractions("It's been a long time."))
SPACY_RESOLVER_AVAILABLE = bool(
resolve_ambiguous_contractions("It's been a long time.")
)
def _normalize_text(text: str, *, normalization_overrides: dict[str, object] | None = None) -> str:
def _normalize_text(
text: str, *, normalization_overrides: dict[str, object] | None = None
) -> str:
runtime_settings = get_runtime_settings()
if normalization_overrides:
runtime_settings = apply_normalization_overrides(runtime_settings, normalization_overrides)
config = build_apostrophe_config(settings=runtime_settings, base=DEFAULT_APOSTROPHE_CONFIG)
runtime_settings = apply_normalization_overrides(
runtime_settings, normalization_overrides
)
config = build_apostrophe_config(
settings=runtime_settings, base=DEFAULT_APOSTROPHE_CONFIG
)
return normalize_for_pipeline(text, config=config, settings=runtime_settings)
@@ -202,7 +210,9 @@ def test_contractions_can_be_kept_when_override_disabled() -> None:
def test_sibilant_possessives_remain_when_marking_disabled() -> None:
normalized = _normalize_text(
"The boss's chair wobbled.",
normalization_overrides={"normalization_apostrophes_sibilant_possessives": False},
normalization_overrides={
"normalization_apostrophes_sibilant_possessives": False
},
)
assert "boss's" in normalized
assert "boss iz" not in normalized.lower()
@@ -276,10 +286,13 @@ def test_spacy_disambiguates_she_would() -> None:
@pytest.mark.skipif(not SPACY_RESOLVER_AVAILABLE, reason="spaCy model unavailable")
def test_sample_sentence_handles_complex_contractions() -> None:
sample = "I've heard the captain'll arrive by dusk, but they'd said the same yesterday."
sample = (
"I've heard the captain'll arrive by dusk, but they'd said the same yesterday."
)
normalized = _normalize_text(sample)
assert (
"I have heard the captain will arrive by dusk, but they had said the same yesterday." == normalized
"I have heard the captain will arrive by dusk, but they had said the same yesterday."
== normalized
)
@@ -316,9 +329,12 @@ def mock_settings():
"normalization_footnotes": True,
"normalization_numbers_year_style": "american",
}
with patch("tests.test_text_normalization.get_runtime_settings", return_value=defaults):
with patch(
"tests.test_text_normalization.get_runtime_settings", return_value=defaults
):
yield
def test_currency_magnitude():
cases = [
("$2 million", "two million dollars"),
@@ -331,13 +347,15 @@ def test_currency_magnitude():
("$2.50", "two dollars, fifty cents"),
("$100", "one hundred dollars"),
]
settings = {
"normalization_numbers": True,
"normalization_currency": True,
"normalization_apostrophe_mode": "spacy"
"normalization_apostrophe_mode": "spacy",
}
for input_text, expected in cases:
normalized = _normalize_text(input_text, normalization_overrides=settings)
assert expected.lower() in normalized.lower(), f"Failed for {input_text}: got '{normalized}'"
assert (
expected.lower() in normalized.lower()
), f"Failed for {input_text}: got '{normalized}'"