Method to find identical books in worker processes without access to the db

src/calibre/db/cache.py

@@ -1774,10 +1774,21 @@ class Cache(object):
             return f.get_books_for_val(item_id_or_composite_value, self._get_proxy_metadata, self._all_book_ids())
         return self._books_for_field(f.name, int(item_id_or_composite_value))
 
+    @read_api
+    def data_for_find_identical_books(self):
+        ''' Return data that can be used to implement
+        :meth:`find_identical_books` in a worker process without access to the
+        db. See db.utils for an implementation. '''
+        at = self.fields['authors'].table
+        author_map = defaultdict(set)
+        for aid, author in at.id_map.iteritems():
+            author_map[icu_lower(author)].add(aid)
+        return (author_map, at.col_book_map.copy(), self.fields['title'].table.book_col_map.copy())
+
     @read_api
     def find_identical_books(self, mi, search_restriction='', book_ids=None):
         ''' Finds books that have a superset of the authors in mi and the same
-        title (title is fuzzy matched) '''
+        title (title is fuzzy matched). See also :meth:`data_for_find_identical_books`. '''
         from calibre.db.utils import fuzzy_title
         identical_book_ids = set()
         if mi.authors:

src/calibre/db/tests/reading.py

@@ -628,3 +628,18 @@ class ReadingTest(BaseTest):
         self.assertEqual('FMT2', cache.field_for('#ccf', 1))
     # }}}
 
+    def test_find_identical_books(self):  # {{{
+        ' Test find_identical_books '
+        from calibre.ebooks.metadata.book.base import Metadata
+        from calibre.db.utils import find_identical_books
+        # 'find_identical_books': [(,), (Metadata('unknown'),), (Metadata('xxxx'),)],
+        cache = self.init_cache(self.library_path)
+        data = cache.data_for_find_identical_books()
+        for mi, books in (
+                (Metadata('title one', ['author one']), {2}),
+                (Metadata(_('Unknown')), {3}),
+                (Metadata('title two', ['author one']), {1}),
+        ):
+            self.assertEqual(books, cache.find_identical_books(mi))
+            self.assertEqual(books, find_identical_books(mi, data))
+    # }}}

src/calibre/db/utils.py

@@ -53,6 +53,23 @@ def fuzzy_title(title):
         title = pat.sub(repl, title)
     return title
 
+def find_identical_books(mi, data):
+    author_map, aid_map, title_map = data
+    author_ids = set()
+    for a in mi.authors:
+        author_ids |= author_map.get(icu_lower(a), set())
+    book_ids = set()
+    for aid in author_ids:
+        book_ids |= aid_map.get(aid, set())
+    ans = set()
+    titleq = fuzzy_title(mi.title)
+    for book_id in book_ids:
+        title = title_map.get(book_id, '')
+        if fuzzy_title(title) == titleq:
+            ans.add(book_id)
+    return ans
+
+
 Entry = namedtuple('Entry', 'path size timestamp thumbnail_size')
 class CacheError(Exception):
     pass