From 64af952caf19cd9cb5d6c16cea63111001d004bd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Sol=C3=A8ne=20Tarride?= <starride@teklia.com>
Date: Mon, 18 Sep 2023 09:23:36 +0200
Subject: [PATCH] Update tests for data extraction
---
tests/test_extract.py | 8 +++-----
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/tests/test_extract.py b/tests/test_extract.py
index 4e2672e5..2d4653ef 100644
--- a/tests/test_extract.py
+++ b/tests/test_extract.py
@@ -186,7 +186,6 @@ def test_normalize_spaces(text, trimmed):
("\rcarriage_return", "carriage_return"),
("\r\ncarriage_return+linebreak", "carriage_return+linebreak"),
("\n\r\r\n\ncarriage_return+linebreak", "carriage_return+linebreak"),
- ("no|linebreaks", "no|linebreaks"),
),
)
def test_normalize_linebreaks(text, trimmed):
@@ -397,10 +396,9 @@ def test_extract(
VAL_DIR / "val-page_1-line_2.jpg",
VAL_DIR / "val-page_1-line_3.jpg",
output / "labels.json",
- # Language resources
- output / "language_model" / "corpus.txt",
- output / "language_model" / "lexicon.txt",
- output / "language_model" / "tokens.txt",
+ output / "language_corpus.txt",
+ output / "language_lexicon.txt",
+ output / "language_tokens.txt",
]
assert sorted(filter(methodcaller("is_file"), output.rglob("*"))) == expected_paths
--
GitLab