Restructure unit tests

tests/e2e/test_anchor_e2e.py

@@ -10,8 +10,8 @@ from genalog.text import alignment, anchor, preprocess
 @pytest.mark.parametrize(
     "gt_file, ocr_file",
     zip(
-        sorted(glob.glob("tests/text/data/gt_*.txt")),
-        sorted(glob.glob("tests/text/data/ocr_*.txt")),
+        sorted(glob.glob("tests/unit/text/data/gt_*.txt")),
+        sorted(glob.glob("tests/unit/text/data/ocr_*.txt")),
     ),
 )
 def test_align_w_anchor_and_align(gt_file, ocr_file):
@@ -37,8 +37,8 @@ def test_align_w_anchor_and_align(gt_file, ocr_file):
 @pytest.mark.parametrize(
     "gt_file, ocr_file",
     zip(
-        sorted(glob.glob("tests/text/data/gt_*.txt")),
-        sorted(glob.glob("tests/text/data/ocr_*.txt")),
+        sorted(glob.glob("tests/unit/text/data/gt_*.txt")),
+        sorted(glob.glob("tests/unit/text/data/ocr_*.txt")),
     ),
 )
 @pytest.mark.parametrize("max_seg_length", [25, 50, 75, 100, 150])

tests/e2e/test_document_generation.py

@@ -11,7 +11,7 @@ CONTENT = CompositeContent(
 UNSUPPORTED_CONTENT_FORMAT = ["foo bar"]
 UNSUPPORTED_CONTENT_TYPE = CompositeContent(["foo"], [ContentType.TITLE])
 
-CUSTOM_TEMPLATE_PATH = "tests/generation/templates"
+CUSTOM_TEMPLATE_PATH = "tests/unit/generation/templates"
 CUSTOM_TEMPLATE_NAME = "mock.html.jinja"
 CUSTOM_STYLE_TEMPLATE_NAME = "font_family.html.jinja"
 MULTI_PAGE_TEMPLATE_NAME = "multipage.html.jinja"

tests/e2e/test_pipeline.py

@@ -4,7 +4,7 @@ import pytest
 
 from genalog import pipeline
 
-EXAMPLE_TEXT_FILE = "tests/text/data/gt_1.txt"
+EXAMPLE_TEXT_FILE = "tests/unit/text/data/gt_1.txt"
 
 
 @pytest.fixture
@@ -36,7 +36,7 @@ def test_custom_generate_img(custom_analog_generator):
 
 
 def test_generate_dataset_multiprocess():
-    INPUT_TEXT_FILENAMES = glob.glob("tests/text/data/gt_*.txt")
+    INPUT_TEXT_FILENAMES = glob.glob("tests/unit/text/data/gt_*.txt")
     with pytest.deprecated_call():
         pipeline.generate_dataset_multiprocess(
             INPUT_TEXT_FILENAMES, "test_out", {}, [], "text_block.html.jinja"

tests/ocr/.env

@@ -1,8 +0,0 @@
-COMPUTER_VISION_ENDPOINT = "https://enki-vision.cognitiveservices.azure.com/"
-SEARCH_SERVICE_NAME = "ocr-ner-pipeline"
-SKILLSET_NAME = "testocrskillset"
-INDEX_NAME = "testocrindex"
-INDEXER_NAME = "testocrindexer"
-DATASOURCE_NAME = "syntheticimages"
-DATASOURCE_CONTAINER_NAME = "testocrimages"
-BLOB_NAME = "syntheticimages"

tests/unit/generation/test_document.py

@@ -12,12 +12,12 @@ MOCK_COMPILED_DOCUMENT = "<p>populated document</p>"
 MOCK_TEMPLATE = MagicMock()
 MOCK_TEMPLATE.render.return_value = MOCK_COMPILED_DOCUMENT
 
-IMG_BYTES = open("tests/generation/2x2.jpg", "rb").read()
+IMG_BYTES = open("tests/unit/generation/2x2.jpg", "rb").read()
 
 FILE_DESTINATION_PDF = "sample.pdf"
 FILE_DESTINATION_PNG = "sample.png"
 
-CUSTOM_TEMPLATE_PATH = "tests/generation/templates"
+CUSTOM_TEMPLATE_PATH = "tests/unit/generation/templates"
 CUSTOM_TEMPLATE_NAME = "mock.html.jinja"
 DEFAULT_TEMPLATE_NAME = "text_block.html.jinja"
 DEFAULT_PACKAGE_NAME = "genalog.generation"

tests/unit/text/test_alignment.py

@@ -5,8 +5,8 @@ from unittest.mock import MagicMock
 import pytest
 
 from genalog.text import alignment
-from tests.cases.text_alignment import ALIGNMENT_REGRESSION_TEST_CASES
-from tests.cases.text_alignment import PARSE_ALIGNMENT_REGRESSION_TEST_CASES
+from tests.unit.cases.text_alignment import ALIGNMENT_REGRESSION_TEST_CASES
+from tests.unit.cases.text_alignment import PARSE_ALIGNMENT_REGRESSION_TEST_CASES
 
 RANDOM_INT = randint(1, 100)
 MOCK_ALIGNMENT_RESULT = [("X", "X", 0, 0, 1)]

tests/unit/text/test_anchor.py

@@ -4,7 +4,7 @@ import warnings
 import pytest
 
 from genalog.text import alignment, anchor, preprocess
-from tests.cases.text_alignment import ALIGNMENT_REGRESSION_TEST_CASES
+from tests.unit.cases.text_alignment import ALIGNMENT_REGRESSION_TEST_CASES
 
 
 @pytest.mark.parametrize(
@@ -200,8 +200,8 @@ def test_find_anchor_recur_fixed_seg_len(
 @pytest.mark.parametrize(
     "gt_file, ocr_file",
     zip(
-        sorted(glob.glob("tests/text/data/gt_1.txt")),
-        sorted(glob.glob("tests/text/data/ocr_1.txt")),
+        sorted(glob.glob("tests/unit/text/data/gt_1.txt")),
+        sorted(glob.glob("tests/unit/text/data/ocr_1.txt")),
     ),
 )
 @pytest.mark.parametrize("max_seg_length", [75])

tests/unit/text/test_ner_label.py

@@ -1,7 +1,7 @@
 import pytest
 
 from genalog.text import ner_label
-from tests.cases.label_propagation import LABEL_PROPAGATION_REGRESSION_TEST_CASES
+from tests.unit.cases.label_propagation import LABEL_PROPAGATION_REGRESSION_TEST_CASES
 
 
 @pytest.mark.parametrize(

tests/unit/text/test_utf8.py

@@ -5,7 +5,7 @@ import pytest
 
 from genalog.text import alignment
 from genalog.text.alignment import GAP_CHAR
-from tests.cases.text_alignment import ALIGNMENT_REGRESSION_TEST_CASES
+from tests.unit.cases.text_alignment import ALIGNMENT_REGRESSION_TEST_CASES
 
 
 def random_utf8_char(byte_len=1):