Merge pull request #577 from davidjb/import-improvements-slug

Provide slug storage option for posts during Pelican import

docs/changelog.rst

@@ -9,6 +9,8 @@ Release history
 3.1 (2012-12-04)
 ================
 
+* Importer now stores slugs within files by default. This can be disabled with
+  the ``--disable-slugs`` option.
 * Improve handling of links to intra-site resources
 * Ensure WordPress import adds paragraphs for all types of line endings
   in post content

docs/importer.rst

@@ -39,29 +39,44 @@ Usage
 """""
 
 | pelican-import [-h] [--wpfile] [--dotclear] [--feed] [-o OUTPUT]
-|                [-m MARKUP][--dir-cat]
+|                [-m MARKUP] [--dir-cat] [--strip-raw] [--disable-slugs]
 |                input
 
+Positional arguments
+====================
+
+  input                 The input file to read
+
 Optional arguments
 """"""""""""""""""
 
   -h, --help            show this help message and exit
-  --wpfile              Wordpress XML export
+  --wpfile              Wordpress XML export (default: False)
-  --dotclear            Dotclear export
+  --dotclear            Dotclear export (default: False)
-  --feed                Feed to parse
+  --feed                Feed to parse (default: False)
   -o OUTPUT, --output OUTPUT
-                        Output path
+                        Output path (default: output)
-  -m MARKUP             Output markup
+  -m MARKUP, --markup MARKUP
+                        Output markup format (supports rst & markdown)
+                        (default: rst)
   --dir-cat             Put files in directories with categories name
+                        (default: False)
+  --strip-raw           Strip raw HTML code that can't be converted to markup
+                        such as flash embeds or iframes (wordpress import
+                        only) (default: False)
+  --disable-slugs       Disable storing slugs from imported posts within
+                        output. With this disabled, your Pelican URLs may not
+                        be consistent with your original posts. (default:
+                        False)
 
 Examples
 ========
 
-for WordPress::
+For WordPress::
 
     $ pelican-import --wpfile -o ~/output ~/posts.xml
 
-for Dotclear::
+For Dotclear::
 
     $ pelican-import --dotclear -o ~/output ~/backup.txt
 

pelican/tools/pelican_import.py

@@ -181,7 +181,7 @@ def feed2fields(file):
         yield (entry.title, entry.description, slug, date, author, [], tags, "html")
 
 
-def build_header(title, date, author, categories, tags):
+def build_header(title, date, author, categories, tags, slug):
     """Build a header from a list of fields"""
     header = '%s\n%s\n' % (title, '#' * len(title))
     if date:
@@ -192,10 +192,12 @@ def build_header(title, date, author, categories, tags):
         header += ':category: %s\n' % ', '.join(categories)
     if tags:
         header += ':tags: %s\n' % ', '.join(tags)
+    if slug:
+        header += ':slug: %s\n' % slug
     header += '\n'
     return header
 
-def build_markdown_header(title, date, author, categories, tags):
+def build_markdown_header(title, date, author, categories, tags, slug):
     """Build a header from a list of fields"""
     header = 'Title: %s\n' % title
     if date:
@@ -206,18 +208,21 @@ def build_markdown_header(title, date, author, categories, tags):
         header += 'Category: %s\n' % ', '.join(categories)
     if tags:
         header += 'Tags: %s\n' % ', '.join(tags)
+    if slug:
+        header += 'Slug: %s\n' % slug
     header += '\n'
     return header
 
-def fields2pelican(fields, out_markup, output_path, dircat=False, strip_raw=False):
+def fields2pelican(fields, out_markup, output_path, dircat=False, strip_raw=False, disable_slugs=False):
     for title, content, filename, date, author, categories, tags, in_markup in fields:
+        slug = not disable_slugs and filename or None
         if (in_markup == "markdown") or (out_markup == "markdown") :
             ext = '.md'
-            header = build_markdown_header(title, date, author, categories, tags)
+            header = build_markdown_header(title, date, author, categories, tags, slug)
         else:
             out_markup = "rst"
             ext = '.rst'
-            header = build_header(title, date, author, categories, tags)
+            header = build_header(title, date, author, categories, tags, slug)
 
         filename = os.path.basename(filename)
 
@@ -278,8 +283,8 @@ def fields2pelican(fields, out_markup, output_path, dircat=False, strip_raw=Fals
 
 def main():
     parser = argparse.ArgumentParser(
-        description="Transform feed, Wordpress or Dotclear files to rst files."
+        description="Transform feed, Wordpress or Dotclear files to reST (rst) "
-            "Be sure to have pandoc installed",
+                    "or Markdown (md) files. Be sure to have pandoc installed.",
         formatter_class=argparse.ArgumentDefaultsHelpFormatter)
 
     parser.add_argument(dest='input', help='The input file to read')
@@ -298,6 +303,11 @@ def main():
     parser.add_argument('--strip-raw', action='store_true', dest='strip_raw',
         help="Strip raw HTML code that can't be converted to "
              "markup such as flash embeds or iframes (wordpress import only)")
+    parser.add_argument('--disable-slugs', action='store_true',
+        dest='disable_slugs',
+        help='Disable storing slugs from imported posts within output. '
+             'With this disabled, your Pelican URLs may not be consistent '
+             'with your original posts.')
 
     args = parser.parse_args()
 
@@ -328,4 +338,5 @@ def main():
 
     fields2pelican(fields, args.markup, args.output,
                    dircat=args.dircat or False,
-                   strip_raw=args.strip_raw or False)
+                   strip_raw=args.strip_raw or False,
+                   disable_slugs=args.disable_slugs or False)

tests/test_importer.py

@@ -48,6 +48,26 @@ class TestWordpressXmlImporter(unittest.TestCase):
                          strip_raw=True))
             self.assertFalse(any('<iframe' in rst for rst in rst_files))
 
+    def test_can_toggle_slug_storage(self):
+
+        posts = list(self.posts)
+        r = lambda f: open(f).read()
+        silent_f2p = mute(True)(fields2pelican)
+
+        with temporary_folder() as temp:
+
+            rst_files = (r(f) for f in silent_f2p(posts, 'markdown', temp))
+            self.assertTrue(all('Slug:' in rst for rst in rst_files))
+            rst_files = (r(f) for f in silent_f2p(posts, 'markdown', temp,
+                         disable_slugs=True))
+            self.assertFalse(any('Slug:' in rst for rst in rst_files))
+
+            rst_files = (r(f) for f in silent_f2p(posts, 'rst', temp))
+            self.assertTrue(all(':slug:' in rst for rst in rst_files))
+            rst_files = (r(f) for f in silent_f2p(posts, 'rst', temp,
+                         disable_slugs=True))
+            self.assertFalse(any(':slug:' in rst for rst in rst_files))
+
     def test_decode_html_entities_in_titles(self):
         posts = list(self.posts)
         test_posts = [post for post in posts if post[2] == 'html-entity-test']