Parsing: Add the 'content' attribute

Add the 'content' attribute to documents to allow access
to a document's parsed content for indexing and other purposes.

Fixes the document parsing indexing failing test.

Signed-off-by: Roberto Rosario <roberto.rosario.gonzalez@gmail.com>

HISTORY.rst

@@ -83,6 +83,8 @@
   and converteed before serializing them.
 - Add the 'ocr_content' attribute to documents to allow access
   to a document's OCR content for indexing and other purposes.
+- Add the 'content' attribute to documents to allow access
+  to a document's parsed content for indexing and other purposes.
 
 3.1.9 (2018-11-01)
 ==================

mayan/apps/document_parsing/apps.py

@@ -15,7 +15,7 @@ from common import (
     MayanAppConfig, menu_facet, menu_multi_item, menu_object, menu_secondary,
     menu_tools
 )
-from common.classes import ModelField
+from common.classes import ModelAttribute, ModelField
 from common.settings import settings_db_sync_task_delay
 from documents.search import document_search, document_page_search
 from documents.signals import post_version_upload
@@ -40,7 +40,7 @@ from .permissions import (
     permission_parse_document
 )
 from .signals import post_document_version_parsing
-from .utils import get_document_content
+from .utils import document_property_content, get_document_content
 
 logger = logging.getLogger(__name__)
 
@@ -95,7 +95,7 @@ class DocumentParsingApp(MayanAppConfig):
 
         Document.add_to_class('submit_for_parsing', document_parsing_submit)
         Document.add_to_class(
-            'content', get_document_content
+            'content', document_property_content
         )
         DocumentVersion.add_to_class(
             'content', get_document_content
@@ -104,6 +104,12 @@ class DocumentParsingApp(MayanAppConfig):
             'submit_for_parsing', document_version_parsing_submit
         )
 
+        ModelAttribute(
+            model=Document, name='content', description=_(
+                'The parsed content of the document.'
+            )
+        )
+
         ModelField(
             Document, name='versions__pages__content__content'
         )

mayan/apps/document_parsing/tests/literals.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-TEST_PARSING_INDEX_NODE_TEMPLATE = '{% if "sample" in document.latest_version.content|join:" "|lower %}sample{% endif %}'
+TEST_PARSING_INDEX_NODE_TEMPLATE = '{% if "sample" in document.content.lower() %}sample{% endif %}'

mayan/apps/document_parsing/utils.py

@@ -14,6 +14,11 @@ def get_document_content(document):
         try:
             page_content = page.content.content
         except DocumentPageContent.DoesNotExist:
-            pass
+            yield ''
         else:
             yield conditional_escape(force_text(page_content))
+
+
+@property
+def document_property_content(self):
+    return ' '.join(get_document_content(self))