my.pdfs: reorganize tests a bit, fix mypy

tests/bluemaestro.py

@@ -38,8 +38,8 @@ def test_old_db() -> None:
 
 @pytest.fixture(autouse=True)
 def prepare():
-    testdata = Path(__file__).absolute().parent.parent / 'testdata'
-    bmdata = testdata / 'hpi-testdata' / 'bluemaestro'
+    from .common import testdata
+    bmdata = testdata() / 'hpi-testdata' / 'bluemaestro'
     assert bmdata.exists(), bmdata
 
     class bluemaestro:

tests/common.py

@@ -1,4 +1,5 @@
 import os
+from pathlib import Path
 
 import pytest
 
@@ -18,3 +19,9 @@ def reset_modules() -> None:
     to_unload = [m for m in sys.modules if re.match(r'my[.]?', m)]
     for m in to_unload:
         del sys.modules[m]
+
+
+def testdata() -> Path:
+    d = Path(__file__).absolute().parent.parent / 'testdata'
+    assert d.exists(), d
+    return d

tests/location.py

@@ -32,10 +32,8 @@ def prepare(tmp_path: Path):
 
 
 def _prepare_google_config(tmp_path: Path):
-    testdata = Path(__file__).absolute().parent.parent / 'testdata'
-    assert testdata.exists(), testdata
-
-    track = one(testdata.rglob('italy-slovenia-2017-07-29.json'))
+    from .common import testdata
+    track = one(testdata().rglob('italy-slovenia-2017-07-29.json'))
 
     # todo ugh. unnecessary zipping, but at the moment takeout provider doesn't support plain dirs
     import zipfile

tests/pdfs.py

@@ -1,34 +1,34 @@
-#!/usr/bin/env python3
-
 import inspect
 from pathlib import Path
 import tempfile
 
 from my.pdfs import get_annots, annotated_pdfs
 
+from .common import testdata
 
-ROOT = Path(__file__).parent.absolute()
-EXPECTED_HIGHLIGHTS = set(['Since 1994, when we first began organizing web sites, we have enjoyed a rare opportunity to participate in the birth of a new discipline. ',
+EXPECTED_HIGHLIGHTS = {
+    'Since 1994, when we first began organizing web sites, we have enjoyed a rare opportunity to participate in the birth of a new discipline. ',
     'And yet, unlearn we must, ',
     '',
-                         ])
+}
 
-def test_get_annots():
+
+def test_get_annots() -> None:
     """
     Test get_annots, with a real PDF file
     get_annots should return a list of three Annotation objects
     """
-    annotations = get_annots(Path(ROOT / 'Information Architecture for the World Wide Web.pdf'))
+    annotations = get_annots(testdata() / 'pdfs' / 'Information Architecture for the World Wide Web.pdf')
     assert len(annotations) == 3
     assert set([a.highlight for a in annotations]) == EXPECTED_HIGHLIGHTS
 
 
-def test_annotated_pdfs_with_filelist():
+def test_annotated_pdfs_with_filelist() -> None:
     """
     Test annotated_pdfs, with a real PDF file
     annotated_pdfs should return a list of one Pdf object, with three Annotations
     """
-    filelist = [Path(ROOT / 'Information Architecture for the World Wide Web.pdf')]
+    filelist = [testdata() / 'pdfs' / 'Information Architecture for the World Wide Web.pdf']
     annotations_generator = annotated_pdfs(filelist=filelist, roots=None)
 
     assert inspect.isgeneratorfunction(annotated_pdfs)
@@ -36,6 +36,7 @@ def test_annotated_pdfs_with_filelist():
     highlights_from_pdfs = []
 
     for pdf_object in list(annotations_generator):
+        assert not isinstance(pdf_object, Exception)
         highlights_from_pdfs.extend([a.highlight for a in pdf_object.annotations])
 
     assert len(highlights_from_pdfs) == 3