Add support for reindexing document on content parsing changes.

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

HISTORY.rst

@@ -1,3 +1,8 @@
+3.2 (2018-XX-XX)
+================
+* Add support for reindexing document on content parsing
+  changes.
+
 3.1.4 (2018-10-XX)
 ==================
 * Fix the link to the documenation. Closes GitLab issue #516.

mayan/apps/document_parsing/apps.py

@@ -25,7 +25,8 @@ from navigation import SourceColumn
 
 from .events import event_parsing_document_version_submit
 from .handlers import (
-    handler_initialize_new_parsing_settings, handler_parse_document_version
+    handler_index_document, handler_initialize_new_parsing_settings,
+    handler_parse_document_version
 )
 from .links import (
     link_document_content, link_document_content_download,
@@ -37,6 +38,7 @@ from .permissions import (
     permission_content_view, permission_document_type_parsing_setup,
     permission_parse_document
 )
+from .signals import post_document_version_parsing
 from .utils import get_document_content
 
 logger = logging.getLogger(__name__)
@@ -178,8 +180,14 @@ class DocumentParsingApp(MayanAppConfig):
                 link_document_type_submit, link_error_list,
             )
         )
+
+        post_document_version_parsing.connect(
+            dispatch_uid='document_parsing_handler_index_document',
+            receiver=handler_index_document,
+            sender=DocumentVersion
+        )
         post_save.connect(
-            dispatch_uid='handler_initialize_new_parsing_settings',
+            dispatch_uid='document_parsing_handler_initialize_new_parsing_settings',
             receiver=handler_initialize_new_parsing_settings,
             sender=DocumentType
         )

mayan/apps/document_parsing/handlers.py

@@ -4,11 +4,19 @@ import logging
 
 from django.apps import apps
 
+from document_indexing.tasks import task_index_document
+
 from .settings import setting_auto_parsing
 
 logger = logging.getLogger(__name__)
 
 
+def handler_index_document(sender, **kwargs):
+    task_index_document.apply_async(
+        kwargs=dict(document_id=kwargs['instance'].document.pk)
+    )
+
+
 def handler_initialize_new_parsing_settings(sender, instance, **kwargs):
     DocumentTypeSettings = apps.get_model(
         app_label='document_parsing', model_name='DocumentTypeSettings'

mayan/apps/document_parsing/managers.py

@@ -10,6 +10,7 @@ from django.db import models
 
 from .events import event_parsing_document_version_finish
 from .parsers import Parser
+from .signals import post_document_version_parsing
 
 logger = logging.getLogger(__name__)
 
@@ -50,6 +51,10 @@ class DocumentPageContentManager(models.Manager):
                 target=document_version
             )
 
+            post_document_version_parsing.send(
+                sender=document_version.__class__, instance=document_version
+            )
+
 
 class DocumentTypeSettingsManager(models.Manager):
     def get_by_natural_key(self, document_type_natural_key):

mayan/apps/document_parsing/signals.py

@@ -0,0 +1,7 @@
+from __future__ import unicode_literals
+
+from django.dispatch import Signal
+
+post_document_version_parsing = Signal(
+    providing_args=('instance',), use_caching=True
+)

mayan/apps/document_parsing/tests/literals.py

@@ -0,0 +1,3 @@
+from __future__ import unicode_literals
+
+TEST_PARSING_INDEX_NODE_TEMPLATE = '{% if "sample" in document.latest_version.content|join:" "|lower %}sample{% endif %}'

mayan/apps/document_parsing/tests/test_indexing.py

@@ -0,0 +1,39 @@
+from __future__ import unicode_literals
+
+from django.test import override_settings
+
+from common.tests import BaseTestCase
+from documents.tests import DocumentTestMixin, TEST_HYBRID_DOCUMENT
+
+from document_indexing.models import Index, IndexInstanceNode
+from document_indexing.tests.literals import TEST_INDEX_LABEL
+
+from .literals import TEST_PARSING_INDEX_NODE_TEMPLATE
+
+
+@override_settings(DOCUMENT_PARSING_AUTO_PARSING=False)
+@override_settings(OCR_AUTO_OCR=False)
+class ParsingIndexingTestCase(DocumentTestMixin, BaseTestCase):
+    auto_upload_document = False
+    test_document_filename = TEST_HYBRID_DOCUMENT
+
+    def test_parsing_indexing(self):
+        index = Index.objects.create(label=TEST_INDEX_LABEL)
+
+        index.document_types.add(self.document_type)
+
+        root = index.template_root
+        index.node_templates.create(
+            parent=root, expression=TEST_PARSING_INDEX_NODE_TEMPLATE,
+            link_documents=True
+        )
+
+        self.document = self.upload_document()
+        self.document.submit_for_parsing()
+        print '@@@', list(self.document.latest_version.content())
+
+        self.assertTrue(
+            self.document in IndexInstanceNode.objects.get(
+                value='sample'
+            ).documents.all()
+        )