update tests

tests/README.md

@@ -0,0 +1,9 @@
+# Tests
+
+This package contains tests for the translation file.
+Run them using pytest against the repository root.
+
+```shell
+pip install pytest
+pytest -v
+```

tests/__init__.py

@@ -1,4 +1,25 @@
 from pathlib import Path
 
-BASE_PATH = Path(__file__).resolve().parent.parent
-TRANSLATION_FILE = BASE_PATH/"master_dv.divehi.txt"
+
+class QuranCorpus:
+    """
+    Simple wrapper class to read
+    content from master_dv.divehi.txt
+    """
+
+    BASE_PATH = Path(__file__).resolve().parent.parent
+    TRANSLATION_FILE = BASE_PATH / "master_dv.divehi.txt"
+
+    @staticmethod
+    def read_all():
+        with open(QuranCorpus.TRANSLATION_FILE, "r") as f:
+            return f.read()
+
+    @staticmethod
+    def read_lines():
+        with open(QuranCorpus.TRANSLATION_FILE, "r") as f:
+            for line in f:
+                if line.startswith("\n") or line.startswith("#"):
+                    continue
+
+                yield line

tests/test_content_structure.py

@@ -1,10 +1,9 @@
-from . import TRANSLATION_FILE
+from . import QuranCorpus
 
 
 def test_line_count():
     """
     Number of lines in translation files must == 6236
     """
-    with open(TRANSLATION_FILE, "r") as f:
-        lines = [line for line in f.readlines() if not line.startswith("#")]
-        assert len(lines) == 6236, f"Expected 6236 lines, got {len(lines)}"
+    lines = list(QuranCorpus.read_lines())
+    assert len(lines) == 6236, f"Expected 6236 lines, got {len(lines)}"

tests/test_content_validity.py

@@ -1,15 +1,14 @@
 import re
-from . import TRANSLATION_FILE
+from . import QuranCorpus
 
 
 def test_no_bom():
     """
     Translation file should not contain BOM's
     as per Tanzil directive
     """
-    with open(TRANSLATION_FILE, "r") as f:
-        trans = f.read()
-        assert "\uFEFF" not in trans
+    trans = QuranCorpus.read_all()
+    assert "\uFEFF" not in trans
 
 
 def test_no_repeated_fili():
@@ -20,10 +19,9 @@ def test_no_repeated_fili():
     error_list = []
     pattern = re.compile("[\u07A6-\u07AB]{2}")
 
-    with open(TRANSLATION_FILE, "r") as f:
-        for line in f:
-            if re.findall(pattern, line):
-                sura, ayat, trans = line.split("|")
-                error_list.append(f"{sura}|{ayat}")
+    for line in QuranCorpus.read_lines():
+        if re.findall(pattern, line):
+            sura, ayat, trans = line.split("|")
+            error_list.append(f"{sura}|{ayat}")
 
     assert len(error_list) == 0, f"Repeated fili found on {error_list}"