Haystack's reserved field names are now configurable.

Author: toastdriven

docs/searchindex_api.rst

@@ -75,6 +75,9 @@ search, et cetera).
     Haystack reserves the following field names for internal use: ``id``,
     ``django_ct``, ``django_id`` & ``content``. The ``name`` & ``type`` names
     used to be reserved but no longer are.
+    
+    You can override these field names using the ``HAYSTACK_ID_FIELD``,
+    ``HAYSTACK_DJANGO_CT_FIELD`` & ``HAYSTACK_DJANGO_ID_FIELD`` if needed.
 
 
 Significance Of ``document=True``

docs/settings.rst

@@ -299,3 +299,51 @@ An example::
     HAYSTACK_LIMIT_TO_REGISTERED_MODELS = False
 
 Default is ``True``.
+
+
+``HAYSTACK_ID_FIELD``
+=====================
+
+**Optional**
+
+This setting allows you to control what the unique field name used internally
+by Haystack is called. Rarely needed unless your field names collide with
+Haystack's defaults.
+
+An example::
+
+    HAYSTACK_ID_FIELD = 'my_id'
+
+Default is ``id``.
+
+
+``HAYSTACK_DJANGO_CT_FIELD``
+============================
+
+**Optional**
+
+This setting allows you to control what the content type field name used
+internally by Haystack is called. Rarely needed unless your field names
+collide with Haystack's defaults.
+
+An example::
+
+    HAYSTACK_DJANGO_CT_FIELD = 'my_django_ct'
+
+Default is ``django_ct``.
+
+
+``HAYSTACK_DJANGO_ID_FIELD``
+============================
+
+**Optional**
+
+This setting allows you to control what the primary key field name used
+internally by Haystack is called. Rarely needed unless your field names
+collide with Haystack's defaults.
+
+An example::
+
+    HAYSTACK_DJANGO_ID_FIELD = 'my_django_id'
+
+Default is ``django_id``.

haystack/backends/__init__.py

@@ -7,7 +7,7 @@
 from django.db.models.base import ModelBase
 from django.utils import tree
 from django.utils.encoding import force_unicode
-from haystack.constants import VALID_FILTERS, FILTER_SEPARATOR
+from haystack.constants import ID, DJANGO_CT, DJANGO_ID, VALID_FILTERS, FILTER_SEPARATOR
 from haystack.exceptions import SearchBackendError, MoreLikeThisError, FacetingError
 try:
     set
@@ -476,7 +476,7 @@ def build_query(self):
             query = self.matching_all_fragment()
 
         if len(self.models):
-            models = sorted(['django_ct:%s.%s' % (model._meta.app_label, model._meta.module_name) for model in self.models])
+            models = sorted(['%s:%s.%s' % (DJANGO_CT, model._meta.app_label, model._meta.module_name) for model in self.models])
             models_clause = ' OR '.join(models)
 
             if query != self.matching_all_fragment():

haystack/backends/solr_backend.py

@@ -4,6 +4,7 @@
 from django.core.exceptions import ImproperlyConfigured
 from django.db.models.loading import get_model
 from haystack.backends import BaseSearchBackend, BaseSearchQuery, log_query
+from haystack.constants import ID, DJANGO_CT, DJANGO_ID
 from haystack.exceptions import MissingDependency, MoreLikeThisError
 from haystack.models import SearchResult
 from haystack.utils import get_identifier
@@ -75,7 +76,11 @@ def remove(self, obj_or_string, commit=True):
         solr_id = get_identifier(obj_or_string)
 
         try:
-            self.conn.delete(id=solr_id, commit=commit)
+            kwargs = {
+                'commit': commit,
+                ID: solr_id
+            }
+            self.conn.delete(**kwargs)
         except (IOError, SolrError), e:
             self.log.error("Failed to remove document '%s' from Solr: %s", solr_id, e)
 
@@ -88,7 +93,7 @@ def clear(self, models=[], commit=True):
                 models_to_delete = []
 
                 for model in models:
-                    models_to_delete.append("django_ct:%s.%s" % (model._meta.app_label, model._meta.module_name))
+                    models_to_delete.append("%s:%s.%s" % (DJANGO_CT, model._meta.app_label, model._meta.module_name))
 
                 self.conn.delete(q=" OR ".join(models_to_delete), commit=commit)
 
@@ -175,7 +180,7 @@ def search(self, query_string, sort_by=None, start_offset=0, end_offset=None,
             registered_models = self.build_registered_models_list()
 
             if len(registered_models) > 0:
-                narrow_queries.add('django_ct:(%s)' % ' OR '.join(registered_models))
+                narrow_queries.add('%s:(%s)' % (DJANGO_CT, ' OR '.join(registered_models)))
 
         if narrow_queries is not None:
             kwargs['fq'] = list(narrow_queries)
@@ -223,15 +228,15 @@ def more_like_this(self, model_instance, additional_query_string=None,
             registered_models = self.build_registered_models_list()
 
             if len(registered_models) > 0:
-                narrow_queries.add('django_ct:(%s)' % ' OR '.join(registered_models))
+                narrow_queries.add('%s:(%s)' % (DJANGO_CT, ' OR '.join(registered_models)))
 
         if additional_query_string:
             narrow_queries.add(additional_query_string)
 
         if narrow_queries:
             params['fq'] = list(narrow_queries)
 
-        query = "id:%s" % get_identifier(model_instance)
+        query = "%s:%s" % (ID, get_identifier(model_instance))
 
         try:
             raw_results = self.conn.more_like_this(query, field_name, **params)
@@ -271,7 +276,7 @@ def _process_results(self, raw_results, highlight=False):
         indexed_models = site.get_indexed_models()
 
         for raw_result in raw_results.docs:
-            app_label, model_name = raw_result['django_ct'].split('.')
+            app_label, model_name = raw_result[DJANGO_CT].split('.')
             additional_fields = {}
             model = get_model(app_label, model_name)
 
@@ -285,14 +290,14 @@ def _process_results(self, raw_results, highlight=False):
                     else:
                         additional_fields[string_key] = self.conn._to_python(value)
 
-                del(additional_fields['django_ct'])
-                del(additional_fields['django_id'])
+                del(additional_fields[DJANGO_CT])
+                del(additional_fields[DJANGO_ID])
                 del(additional_fields['score'])
 
-                if raw_result['id'] in getattr(raw_results, 'highlighting', {}):
-                    additional_fields['highlighted'] = raw_results.highlighting[raw_result['id']]
+                if raw_result[ID] in getattr(raw_results, 'highlighting', {}):
+                    additional_fields['highlighted'] = raw_results.highlighting[raw_result[ID]]
 
-                result = SearchResult(app_label, model_name, raw_result['django_id'], raw_result['score'], **additional_fields)
+                result = SearchResult(app_label, model_name, raw_result[DJANGO_ID], raw_result['score'], **additional_fields)
                 results.append(result)
             else:
                 hits -= 1

haystack/backends/whoosh_backend.py

@@ -9,6 +9,7 @@
 from django.utils.datetime_safe import datetime
 from django.utils.encoding import force_unicode
 from haystack.backends import BaseSearchBackend, BaseSearchQuery, log_query
+from haystack.constants import ID, DJANGO_CT, DJANGO_ID
 from haystack.fields import DateField, DateTimeField, IntegerField, FloatField, BooleanField, MultiValueField
 from haystack.exceptions import MissingDependency, SearchBackendError
 from haystack.models import SearchResult
@@ -32,7 +33,8 @@
 
 # Bubble up the correct error.
 from whoosh.analysis import StemmingAnalyzer
-from whoosh.fields import Schema, ID, IDLIST, STORED, TEXT, KEYWORD, NUMERIC, BOOLEAN, DATETIME
+from whoosh.fields import Schema, IDLIST, STORED, TEXT, KEYWORD, NUMERIC, BOOLEAN, DATETIME
+from whoosh.fields import ID as WHOOSH_ID
 from whoosh import index
 from whoosh.qparser import QueryParser
 from whoosh.filedb.filestore import FileStorage, RamStorage
@@ -118,9 +120,9 @@ def setup(self):
 
     def build_schema(self, fields):
         schema_fields = {
-            'id': ID(stored=True, unique=True),
-            'django_ct': ID(stored=True),
-            'django_id': ID(stored=True),
+            ID: WHOOSH_ID(stored=True, unique=True),
+            DJANGO_CT: WHOOSH_ID(stored=True),
+            DJANGO_ID: WHOOSH_ID(stored=True),
         }
         # Grab the number of keys that are hard-coded into Haystack.
         # We'll use this to (possibly) fail slightly more gracefully later.
@@ -186,7 +188,7 @@ def remove(self, obj_or_string, commit=True):
 
         self.index = self.index.refresh()
         whoosh_id = get_identifier(obj_or_string)
-        self.index.delete_by_query(q=self.parser.parse(u'id:"%s"' % whoosh_id))
+        self.index.delete_by_query(q=self.parser.parse(u'%s:"%s"' % (ID, whoosh_id)))
 
     def clear(self, models=[], commit=True):
         if not self.setup_complete:
@@ -200,7 +202,7 @@ def clear(self, models=[], commit=True):
             models_to_delete = []
 
             for model in models:
-                models_to_delete.append(u"django_ct:%s.%s" % (model._meta.app_label, model._meta.module_name))
+                models_to_delete.append(u"%s:%s.%s" % (DJANGO_CT, model._meta.app_label, model._meta.module_name))
 
             self.index.delete_by_query(q=self.parser.parse(u" OR ".join(models_to_delete)))
 
@@ -301,7 +303,7 @@ def search(self, query_string, sort_by=None, start_offset=0, end_offset=None,
             registered_models = self.build_registered_models_list()
 
             if len(registered_models) > 0:
-                narrow_queries.add('django_ct:(%s)' % ' OR '.join(registered_models))
+                narrow_queries.add('%s:(%s)' % (DJANGO_CT, ' OR '.join(registered_models)))
 
         if narrow_queries is not None:
             # Potentially expensive? I don't see another way to do it in Whoosh...
@@ -404,7 +406,7 @@ def _process_results(self, raw_page, highlight=False, query_string='', spelling_
 
         for doc_offset, raw_result in enumerate(raw_page):
             score = raw_page.score(doc_offset) or 0
-            app_label, model_name = raw_result['django_ct'].split('.')
+            app_label, model_name = raw_result[DJANGO_CT].split('.')
             additional_fields = {}
             model = get_model(app_label, model_name)
 
@@ -425,8 +427,8 @@ def _process_results(self, raw_page, highlight=False, query_string='', spelling_
                     else:
                         additional_fields[string_key] = self._to_python(value)
 
-                del(additional_fields['django_ct'])
-                del(additional_fields['django_id'])
+                del(additional_fields[DJANGO_CT])
+                del(additional_fields[DJANGO_ID])
 
                 if highlight:
                     from whoosh import analysis
@@ -438,7 +440,7 @@ def _process_results(self, raw_page, highlight=False, query_string='', spelling_
                         self.content_field_name: [highlight(additional_fields.get(self.content_field_name), terms, sa, ContextFragmenter(terms), UppercaseFormatter())],
                     }
 
-                result = SearchResult(app_label, model_name, raw_result['django_id'], score, **additional_fields)
+                result = SearchResult(app_label, model_name, raw_result[DJANGO_ID], score, **additional_fields)
                 results.append(result)
             else:
                 hits -= 1

haystack/constants.py

@@ -1,5 +1,10 @@
 from django.conf import settings
 
+# Reserved field names
+ID = getattr(settings, 'HAYSTACK_ID_FIELD', 'id')
+DJANGO_CT = getattr(settings, 'HAYSTACK_DJANGO_CT_FIELD', 'django_ct')
+DJANGO_ID = getattr(settings, 'HAYSTACK_DJANGO_ID_FIELD', 'django_id')
+
 # Default operator. Valid options are AND/OR.
 DEFAULT_OPERATOR = getattr(settings, 'HAYSTACK_DEFAULT_OPERATOR', 'AND')
 

haystack/indexes.py

@@ -2,6 +2,7 @@
 import sys
 from django.db.models import signals
 from django.utils.encoding import force_unicode
+from haystack.constants import ID, DJANGO_CT, DJANGO_ID
 from haystack.fields import *
 from haystack.utils import get_identifier, get_facet_field_name
 
@@ -122,9 +123,9 @@ def prepare(self, obj):
         Fetches and adds/alters data before indexing.
         """
         self.prepared_data = {
-            'id': get_identifier(obj),
-            'django_ct': "%s.%s" % (obj._meta.app_label, obj._meta.module_name),
-            'django_id': force_unicode(obj.pk),
+            ID: get_identifier(obj),
+            DJANGO_CT: "%s.%s" % (obj._meta.app_label, obj._meta.module_name),
+            DJANGO_ID: force_unicode(obj.pk),
         }
 
         for field_name, field in self.fields.items():
@@ -302,7 +303,7 @@ class ModelSearchIndex(SearchIndex):
     """
     text = CharField(document=True, use_template=True)
     # list of reserved field names
-    fields_to_skip = ('id', 'django_ct', 'django_id', 'content', 'text')
+    fields_to_skip = (ID, DJANGO_CT, DJANGO_ID, 'content', 'text')
 
     def __init__(self, model, backend=None, extra_field_kwargs=None):
         self.model = model

haystack/utils/__init__.py

@@ -1,5 +1,6 @@
 import re
 from django.utils.html import strip_tags
+from haystack.constants import ID, DJANGO_CT, DJANGO_ID
 try:
     set
 except NameError:
@@ -26,7 +27,7 @@ def get_identifier(obj_or_string):
 
 
 def get_facet_field_name(fieldname):
-    if fieldname in ['id', 'django_id', 'django_ct']:
+    if fieldname in [ID, DJANGO_ID, DJANGO_CT]:
         return fieldname
 
     return "%s_exact" % fieldname