From 6f45344847a544542c54e8ddcf8fb03dd567af6c Mon Sep 17 00:00:00 2001
From: Charles Haley <>
Date: Sun, 3 Oct 2010 14:54:06 +0100
Subject: [PATCH] Add tweak to change articles in title sort pattern

---
 resources/default_tweaks.py             |  9 +++++++++
 src/calibre/ebooks/metadata/__init__.py | 10 +++++++++-
 2 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/resources/default_tweaks.py b/resources/default_tweaks.py
index 3b60e3410e..17ed0eeea7 100644
--- a/resources/default_tweaks.py
+++ b/resources/default_tweaks.py
@@ -83,6 +83,15 @@ title_series_sorting = 'library_order'
 # strictly_alphabetic, it would remain "The Client".
 save_template_title_series_sorting = 'library_order'
 
+# Set the list of words that are to be considered 'articles' when computing the
+# title sort strings. The list is a regular expression, with the articles
+# separated by 'or' bars. Comparisons are case insensitive, and that cannot be
+# changed. Changes to this tweak won't have an effect until the book is modified
+# in some way. If you enter an invalid pattern, it is silently ignored.
+# Default: '^(A|The|An)\s+'
+title_sort_articles='^(A|The|An)\s+'
+
+
 # Specify a folder that calibre should connect to at startup using
 # connect_to_folder. This must be a full path to the folder. If the folder does
 # not exist when calibre starts, it is ignored. If there are '\' characters in
diff --git a/src/calibre/ebooks/metadata/__init__.py b/src/calibre/ebooks/metadata/__init__.py
index 429ba06c6e..01e5190640 100644
--- a/src/calibre/ebooks/metadata/__init__.py
+++ b/src/calibre/ebooks/metadata/__init__.py
@@ -44,7 +44,15 @@ def author_to_author_sort(author):
 def authors_to_sort_string(authors):
     return ' & '.join(map(author_to_author_sort, authors))
 
-_title_pat = re.compile('^(A|The|An)\s+', re.IGNORECASE)
+try:
+    _title_pat = re.compile(tweaks.get('title_sort_articles',
+                                       r'^(A|The|An)\s+'), re.IGNORECASE)
+except:
+    print 'Error in title sort pattern'
+    import traceback
+    traceback.print_exc()
+    _title_pat = re.compile('^(A|The|An)\s+', re.IGNORECASE)
+
 _ignore_starts = u'\'"'+u''.join(unichr(x) for x in range(0x2018, 0x201e)+[0x2032, 0x2033])
 
 def title_sort(title):