add test for dotclear import (improve test coverage)

Add a check for line breaks in the importer code to better handle transition from section (posts and category)

pelican/tests/test_importer.py

@@ -4,12 +4,17 @@ from __future__ import unicode_literals, print_function
 import os
 import re
 
-from pelican.tools.pelican_import import wp2fields, fields2pelican, decode_wp_content
+from pelican.tools.pelican_import import dc2fields, wp2fields, fields2pelican, decode_wp_content
 from pelican.tests.support import (unittest, temporary_folder, mute,
                                    skipIfNoExecutable)
 
 CUR_DIR = os.path.dirname(__file__)
 WORDPRESS_XML_SAMPLE = os.path.join(CUR_DIR, 'content', 'wordpressexport.xml')
+
+# based on http://themes.dotaddict.org/files/public/downloads/lorem-backup.txt
+# suggested by http://docs.getpelican.com/en/2.8/importer.html
+DOTCLEAR_SAMPLE = os.path.join(CUR_DIR, 'content', 'lorem_backup_dotclear-2.1.5.txt')
+
 WORDPRESS_ENCODED_CONTENT_SAMPLE = os.path.join(CUR_DIR,
                                                 'content',
                                                 'wordpress_content_encoded')
@@ -22,6 +27,17 @@ try:
 except ImportError:
     BeautifulSoup = False  # NOQA
 
+@unittest.skipUnless(BeautifulSoup, 'Needs BeautifulSoup module')
+class DotClearImporter(unittest.TestCase):
+    def test_dotclear(self):
+        # Act
+        self.posts = list(dc2fields(DOTCLEAR_SAMPLE))
+
+        # Assert
+        self.assertEqual(26, len(self.posts))
+        self.assertEqual(u'Mon premier billet', self.posts[0][0])
+        #x = self.posts
+
 
 @skipIfNoExecutable(['pandoc', '--version'])
 @unittest.skipUnless(BeautifulSoup, 'Needs BeautifulSoup module')

pelican/tools/pelican_import.py

@@ -152,18 +152,21 @@ def dc2fields(file):
                  '"BeautifulSoup4" and "lxml" required to import Dotclear files.')
         sys.exit(error)
 
-
     in_cat = False
     in_post = False
     category_list = {}
     posts = []
 
     with open(file, 'r', encoding='utf-8') as f:
-
         for line in f:
             # remove final \n
             line = line[:-1]
 
+            # When there is a line break, interpret this as a section break
+            if line == '\r':
+                in_cat = False
+                in_post = False
+
             if line.startswith('[category'):
                 in_cat = True
             elif line.startswith('[post'):