Merge e860135 into 6370063

cnxarchive/database.py

@@ -60,6 +60,7 @@ def _read_sql_file(name):
     'get-in-book-search': _read_sql_file('get-in-book-search'),
     'get-in-book-search-full-page': _read_sql_file(
         'get-in-book-search-full-page'),
+    'get-collated-content': _read_sql_file('get-collated-content'),
     }
 
 
@@ -227,11 +228,11 @@ def get_module_ident_from_ident_hash(ident_hash, cursor):
     return module_ident
 
 
-def get_tree(ident_hash, cursor):
+def get_tree(ident_hash, cursor, as_collated=False):
     """Return a JSON representation of the binder tree for ``ident_hash``."""
     uuid, version = split_ident_hash(ident_hash)
     cursor.execute(SQL['get-tree-by-uuid-n-version'],
-                   (uuid, version,))
+                   (uuid, version, as_collated,))
     try:
         tree = cursor.fetchone()[0]
     except TypeError:  # NoneType
@@ -243,6 +244,16 @@ def get_tree(ident_hash, cursor):
         return tree
 
 
+def get_collated_content(ident_hash, context_ident_hash, cursor):
+    """Return collated content for ``ident_hash``."""
+    cursor.execute(SQL['get-collated-content'],
+                   (ident_hash, context_ident_hash,))
+    try:
+        return cursor.fetchone()[0]
+    except TypeError:  # NoneType
+        return
+
+
 def get_module_uuid(plpy, moduleid):
     """Retrieve page uuid from legacy moduleid."""
     plan = plpy.prepare("SELECT uuid FROM modules WHERE moduleid = $1;",

cnxarchive/sql/get-collated-content.sql

@@ -0,0 +1,15 @@
+-- ###
+-- Copyright (c) 2016, Rice University
+-- This software is subject to the provisions of the GNU Affero General
+-- Public License version 3 (AGPLv3).
+-- See LICENCE.txt for details.
+-- ###
+
+-- arguments[positional]: ident_hash:string; context_ident_hash:string
+SELECT f.file
+FROM files AS f
+  LEFT JOIN collated_file_associations cfa ON cfa.fileid = f.fileid
+  LEFT JOIN modules m ON m.module_ident = cfa.item
+  LEFT JOIN modules context ON context.module_ident = cfa.context
+WHERE m.uuid || '@' || concat_ws('.', m.major_version, m.minor_version) = %s AND
+      context.uuid || '@' || concat_ws('.', context.major_version, context.minor_version) = %s

cnxarchive/sql/get-tree-by-uuid-n-version.sql

@@ -5,5 +5,5 @@
 -- See LICENCE.txt for details.
 -- ###
 
--- arguments[positional]: id:string; version:string;
-SELECT tree_to_json(%s, %s, FALSE)::json;
+-- arguments[positional]: id:string; version:string; as_collated:boolean
+SELECT tree_to_json(%s, %s, %s)::json;

cnxarchive/sql/search/query.sql

@@ -64,6 +64,7 @@ FROM
   derived_weighted_query_results AS wqr
 WHERE
   wqr.module_ident = lm.module_ident
+  AND lm.portal_type != 'CompositeModule'
   {filters}
   {groupby}
 ORDER BY {sorts}

cnxarchive/tests/test_database.py

@@ -917,7 +917,6 @@ def test_module(self, cursor):
 
         cursor.execute('SELECT COUNT(*) FROM modules')
         old_n_modules = cursor.fetchone()[0]
-        self.assertEqual(old_n_modules, 19)
 
         # Insert a new version of an existing module
         cursor.execute('''\

cnxarchive/tests/test_views.py

@@ -494,12 +494,46 @@ def test_derived_collection(self):
             self.assertEqual(content['parent'][key],
                              COLLECTION_DERIVED_METADATA['parent'][key])
 
+    def test_content_collated_collection(self):
+        uuid = 'e79ffde3-7fb4-4af3-9ec8-df648b391597'
+        version = '6.1'
+
+        # Build the request environment.
+        self.request.matchdict = {'ident_hash': '{}@{}'.format(uuid, version)}
+        self.request.matched_route = mock.Mock()
+        self.request.matched_route.name = 'content'
+
+        # Call the view.
+        from ..views import get_content
+        content = get_content(self.request).json_body
+
+        # Check the tree.
+        self.assertEqual({
+            u'id': u'{}@{}'.format(uuid, version),
+            u'shortId': u'55_943-0@6.1',
+            u'title': u'College Physics',
+            u'contents': [
+                {u'id': u'209deb1f-1a46-4369-9e0d-18674cf58a3e@7',
+                 u'shortId': u'IJ3rHxpG@7',
+                 u'title': u'Preface'},
+                {u'id': u'174c4069-2743-42e9-adfe-4c7084f81fc5@1',
+                 u'shortId': u'F0xAaSdD@1',
+                 u'title': u'Collated page'},
+                ],
+            }, content['tree'])
+
     @testing.db_connect
     def _create_empty_subcollections(self, cursor):
-        cursor.execute("INSERT INTO trees VALUES (91, 53, NULL, 'Empty Subcollections', 1)")
-        cursor.execute("INSERT INTO trees VALUES (92, 91, NULL, 'empty 1', 1)")
-        cursor.execute("INSERT INTO trees VALUES (93, 91, NULL, 'empty 2', 2)")
-        cursor.execute("INSERT INTO trees VALUES (94, 53, NULL, 'Empty Subcollection', 3)")
+        cursor.execute("""\
+INSERT INTO trees (nodeid, parent_id, title, childorder, is_collated)
+    VALUES (9100, 91, 'Empty Subcollections', 1, true);
+INSERT INTO trees (nodeid, parent_id, title, childorder, is_collated)
+    VALUES (9200, 9100, 'empty 1', 1, true);
+INSERT INTO trees (nodeid, parent_id, title, childorder, is_collated)
+    VALUES (9300, 9100, 'empty 2', 2, true);
+INSERT INTO trees (nodeid, parent_id, title, childorder, is_collated)
+    VALUES (9400, 91, 'Empty Subcollection', 4, true);
+""")
 
     def test_empty_subcollection_content(self):
         self._create_empty_subcollections()
@@ -551,6 +585,11 @@ def test_empty_subcollection_content(self):
                     u'shortId': u'IJ3rHxpG@7',
                     u'title': u'Preface',
                     },
+                {
+                    u'id': u'174c4069-2743-42e9-adfe-4c7084f81fc5@1',
+                    u'shortId': u'F0xAaSdD@1',
+                    u'title': u'Collated page',
+                    },
                 {
                     u'id': u'subcol',
                     u'shortId': u'subcol',
@@ -616,6 +655,25 @@ def test_module_content(self):
         # Check the content is the html file.
         self.assertTrue(content_text.find('<html') >= 0)
 
+    def test_content_composite_module(self):
+        # Test for retrieving a a composite module.
+        uuid = '174c4069-2743-42e9-adfe-4c7084f81fc5'
+        version = '1'
+
+        # Build the request environment.
+        self.request.matchdict = {'ident_hash': '{}@{}'.format(uuid, version)}
+        self.request.matched_route = mock.Mock()
+        self.request.matched_route.name = 'content'
+
+        from ..views import get_content
+
+        content = get_content(self.request).json_body
+
+        # Check the content.
+        self.assertEqual(
+            '<html><body>test collated content</body></html>',
+            content['content'])
+
     def test_content_without_version(self):
         uuid = 'ae3e18de-638d-4738-b804-dc69cd4db3a3'
 
@@ -696,6 +754,48 @@ def test_content_not_found_w_invalid_uuid(self):
         self.assertRaises(IdentHashShortId, get_content,
                           self.request)
 
+    def test_content_collated_page_inside_book(self):
+        book_uuid = 'e79ffde3-7fb4-4af3-9ec8-df648b391597'
+        book_version = '6.1'
+        page_uuid = '209deb1f-1a46-4369-9e0d-18674cf58a3e'
+        page_version = '7'
+
+        # Build the request.
+        self.request.matchdict = {
+            'ident_hash': '{}@{}'.format(book_uuid, book_version),
+            'page_ident_hash': '{}@{}'.format(page_uuid, page_version),
+            'separator': ':',
+            }
+        self.request.matched_route = mock.Mock()
+        self.request.matched_route.name = 'content'
+
+        # Call the view.
+        from ..views import get_content
+        content = get_content(self.request).json_body
+
+        self.assertEqual(
+            '<html><body>Page content after collation</body></html>\n',
+            content['content'])
+
+    def test_content_uncollated_page(self):
+        page_uuid = '209deb1f-1a46-4369-9e0d-18674cf58a3e'
+        page_version = '7'
+
+        # Build the request.
+        self.request.matchdict = {
+            'ident_hash': '{}@{}'.format(page_uuid, page_version),
+            }
+        self.request.matched_route = mock.Mock()
+        self.request.matched_route.name = 'content'
+
+        # Call the view.
+        from ..views import get_content
+        content = get_content(self.request).json_body
+
+        self.assertNotEqual(
+            '<html><body>Page content after collation</body></html>\n',
+            content['content'])
+
     def test_content_page_inside_book_version_mismatch(self):
         book_uuid = 'e79ffde3-7fb4-4af3-9ec8-df648b391597'
         book_version = '7.1'

cnxarchive/utils/mimetype.py

@@ -23,6 +23,7 @@
 MIMETYPES = (MODULE_MIMETYPE, COLLECTION_MIMETYPE, FOLDER_MIMETYPE,)
 PORTALTYPE_TO_MIMETYPE_MAPPING = {
     'Module': MODULE_MIMETYPE,
+    'CompositeModule': MODULE_MIMETYPE,
     'Collection': COLLECTION_MIMETYPE,
     }
 

cnxarchive/views.py

@@ -26,7 +26,7 @@
 from . import cache
 # FIXME double import
 from . import database
-from .database import SQL, get_tree
+from .database import SQL, get_tree, get_collated_content
 from .search import (
     DEFAULT_PER_PAGE, QUERY_TYPES, DEFAULT_QUERY_TYPE,
     Query,
@@ -299,13 +299,24 @@ def _get_content_json(ident_hash=None):
             result = get_content_metadata(id, version, cursor)
             if result['mediaType'] == COLLECTION_MIMETYPE:
                 # Grab the collection tree.
-                result['tree'] = get_tree(ident_hash, cursor)
+                result['tree'] = get_tree(ident_hash, cursor, as_collated=True)
+                if not result['tree']:
+                    # If collated tree is not available, get the uncollated
+                    # tree.
+                    result['tree'] = get_tree(ident_hash, cursor)
 
                 if page_ident_hash:
                     for id_ in flatten_tree_to_ident_hashes(result['tree']):
                         id, version = split_ident_hash(id_)
                         if id == p_id and (
                            version == p_version or not p_version):
+                            content = get_collated_content(
+                                id_, ident_hash, cursor)
+                            if content:
+                                result = get_content_metadata(
+                                    p_id, p_version, cursor)
+                                result['content'] = content[:]
+                                return result
                             raise httpexceptions.HTTPFound(request.route_path(
                                 request.matched_route.name,
                                 ident_hash=join_ident_hash(id, version)))
@@ -886,6 +897,7 @@ def sitemap(request):
                             '[^0-9a-z]+', ' ', 'g')), ' +', '-', 'g'),
                         revised
                     FROM latest_modules
+                    WHERE portal_type != 'CompositeModule'
                     ORDER BY revised DESC LIMIT 50000""")
             res = cursor.fetchall()
             for ident_hash, page_name, revised in res: