From 09de4122cf7caa91612867479e9a6b04a4dd67bf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fr=C3=A9d=C3=A9ric=20P=C3=A9ters?= <fpeters@entrouvert.com>
Date: Tue, 3 Dec 2013 11:03:22 +0100
Subject: [PATCH] add custom indexer to also include thesaurus terms (#3675)

---
 src/pfwbged/basecontent/indexers.py | 51 +++++++++++++++++++++++++++++
 1 file changed, 51 insertions(+)
 create mode 100644 src/pfwbged/basecontent/indexers.py

diff --git a/src/pfwbged/basecontent/indexers.py b/src/pfwbged/basecontent/indexers.py
new file mode 100644
index 0000000..5d64d97
--- /dev/null
+++ b/src/pfwbged/basecontent/indexers.py
@@ -0,0 +1,51 @@
+from ZODB.POSException import ConflictError
+from five import grok
+from Products.CMFCore.utils import getToolByName
+from plone.indexer import indexer
+
+from zope.component import getUtility
+from zope.schema.interfaces import IVocabularyFactory
+
+
+from .behaviors import IPfwbDocument
+
+@indexer(IPfwbDocument)
+def document_dynamic_searchable_text_indexer(obj):
+    indexed_elements = [obj.title]
+
+    # if there is no path to text/plain, do nothing
+    transforms = getToolByName(obj, 'portal_transforms')
+
+    had_version = False
+    for child in reversed(obj.values()):
+        if child.portal_type in ('dmsmainfile', 'dmsappendixfile'):
+            if not child.file or child.file.getSize() == 0:
+                continue
+
+            if not transforms._findPath(child.file.contentType, 'text/plain'):
+                continue
+
+            # only index the latest version
+            if child.portal_type == 'dmsmainfile' and had_version:
+                continue
+            had_version = True
+
+            # convert it to text/plain
+            try:
+                datastream = transforms.convertTo(
+                    'text/plain', child.file.data, mimetype=child.file.contentType,
+                    filename=child.file.filename)
+                indexed_elements.append(unicode(datastream.getData(), 'utf-8'))
+            except (ConflictError, KeyboardInterrupt):
+                raise
+
+    if hasattr(obj, 'keywords'):
+        factory = getUtility(IVocabularyFactory, 'dms.thesaurus.simple')
+        vocabulary = factory(obj)
+        for keyword in obj.keywords:
+            indexed_elements.append(vocabulary.getTerm(keyword).title)
+
+    return u' '.join(indexed_elements)
+
+grok.global_adapter(document_dynamic_searchable_text_indexer,
+                    name='SearchableText')