fixes for edge cases

[librarian.git] / tests / test_text.py

diff --git a/tests/test_text.py b/tests/test_text.py
index 70dfb60..c566a48 100644 (file)
--- a/tests/test_text.py
+++ b/tests/test_text.py
@@ -3,32 +3,62 @@
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
+from __future__ import unicode_literals
+
+import unittest
 from librarian import NoDublinCore
-from librarian.parser import WLDocument
+from librarian.builders import builders
+from librarian.parser import WLDocument as LegacyWLDocument
+from librarian.document import WLDocument
 from nose.tools import *
-from utils import get_fixture
+from .utils import get_fixture
+
+
+class TextTests(unittest.TestCase):
+    maxDiff = None
+
+    def test_transform_legacy(self):
+        expected_output_file_path = get_fixture('text', 'asnyk_miedzy_nami_expected.txt')
+
+        text = LegacyWLDocument.from_file(
+            get_fixture('text', 'miedzy-nami-nic-nie-bylo.xml')
+        ).as_text().get_bytes().decode('utf-8')
+
+        with open(expected_output_file_path, 'rb') as f:
+            self.assertEqual(text, f.read().decode('utf-8'))
+
+    def test_transform(self):
+        expected_output_file_path = get_fixture('text', 'asnyk_miedzy_nami_expected.txt')
+
+        text = WLDocument(
+            filename=get_fixture('text', 'miedzy-nami-nic-nie-bylo.xml')
+        ).build(builders['txt']).get_bytes().decode('utf-8')
 
+        with open(expected_output_file_path, 'rb') as f:
+            self.assertEqual(text, f.read().decode('utf-8'))
 
-def test_transform():
-    expected_output_file_path = get_fixture('text', 'asnyk_miedzy_nami_expected.txt')
+    
+    def test_transform_raw(self):
+        expected_output_file_path = get_fixture('text', 'asnyk_miedzy_nami_expected_raw.txt')
 
-    text = WLDocument.from_file(
+        text = LegacyWLDocument.from_file(
             get_fixture('text', 'miedzy-nami-nic-nie-bylo.xml')
-        ).as_text().get_string()
+        ).as_text(flags=['raw-text']).get_bytes().decode('utf-8')
 
-    assert_equal(text, file(expected_output_file_path).read())
+        with open(expected_output_file_path, 'rb') as f:
+            self.assertEqual(text, f.read().decode('utf-8'))
 
 
 @raises(NoDublinCore)
 def test_no_dublincore():
-    WLDocument.from_file(
+    LegacyWLDocument.from_file(
             get_fixture('text', 'asnyk_miedzy_nami_nodc.xml')
         ).as_text()
 
 
 def test_passing_parse_dublincore_to_transform():
     """Passing parse_dublincore=False to the constructor omits DublinCore parsing."""
-    WLDocument.from_file(
+    LegacyWLDocument.from_file(
             get_fixture('text', 'asnyk_miedzy_nami_nodc.xml'),
             parse_dublincore=False,
         ).as_text()