Architizer · manelore · Aug 24, 2011 · Aug 31, 2011 · Feb 25, 2013 · Mar 3, 2013
diff --git a/.travis.yml b/.travis.yml
@@ -1,9 +1,30 @@
 language: python
 python:
-    - "2.6"
-    - "2.7"
-    - "3.2"
+  - "2.7"
 
-branches:
-    only:
-        - travis-config
+before_install:
+  - sudo apt-get update
+  - sudo apt-get install wajig binutils gdal-bin libproj-dev libgeos-c1 libxapian22 python-xapian default-jdk
+  - curl https://raw.github.com/moliware/travis-solr/master/travis-solr.sh | SOLR_CONFS="tests/schema.xml tests/solrconfig.xml" SOLR_VERSION=4.4.0 bash
+
+install:
+  - pip install tox
+  - pip install -e .
+
+script:
+  - tox -e $TOX_ENV
+
+env:
+  - TOX_ENV=py26-django1.5
+  - TOX_ENV=py26-django1.6
+  - TOX_ENV=py27-django1.5
+  - TOX_ENV=py27-django1.6
+  - TOX_ENV=py33-django1.5
+  - TOX_ENV=py33-django1.6
+
+services:
+  - elasticsearch
+
+notifications:
+  #irc: "irc.freenode.org#haystack"
+  email: false
diff --git a/AUTHORS b/AUTHORS
@@ -4,7 +4,17 @@ Primary Authors:
     * Matt Croydon (some documentation, sanity checks and the sweet name)
     * Travis Cline (the original SQ implementation, improvements to ModelSearchIndex)
     * David Sauve (notanumber) for the Xapian backend, the simple backend and various patches.
-    * Jannis Leidel
+    * Jannis Leidel (jezdez)
+    * Chris Adams (acdha)
+    * Justin Caratzas (bigjust)
+    * Andrew Schoen (andrewschoen)
+    * Dan Watson (dcwatson)
+    * Matt Woodward (mpwoodward)
+    * Alex Vidal (avidal)
+    * Zach Smith (zmsmith)
+    * Stefan Wehrmeyer (stefanw)
+    * George Hickman (ghickman)
+    * Ben Spaulding (benspaulding)
 
 
 Thanks to
@@ -71,3 +81,12 @@ Thanks to
     * Stefan Wehrmeyer (stefanw) for a simple search filter fix
     * Dan Watson (dcwatson) for various patches.
     * Andrew Schoen (andrewschoen) for the addition of ``HAYSTACK_IDENTIFIER_METHOD``
+    * Pablo SEMINARIO (pabluk) for a docs fix, and a fix in the ElasticSearch backend.
+    * Eric Thurgood (ethurgood) for a import fix in the Elasticssearch backend.
+    * Revolution Systems & The Python Software Foundation for funding a significant portion of the port to Python 3!
+    * Artem Kostiuk (postatum) for patch allowing to search for slash character in ElasticSearch since Lucene 4.0.
+    * Luis Barrueco (luisbarrueco) for a simple fix regarding updating indexes using multiple backends.
+    * Szymon Teżewski (jasisz) for an update to the bounding-box calculation for spatial queries
+    * Chris Wilson (qris) and Orlando Fiol (overflow) for an update allowing the use of multiple order_by()
+      fields with Whoosh as long as they share a consistent sort direction
+    * Steven Skoczen (@skoczen) for an ElasticSearch bug fix
diff --git a/LICENSE b/LICENSE
@@ -1,4 +1,4 @@
-Copyright (c) 2009-2012, Daniel Lindsley.
+Copyright (c) 2009-2013, Daniel Lindsley.
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without modification,

diff --git a/README.rst b/README.rst
@@ -3,7 +3,7 @@ Haystack
 ========
 
 :author: Daniel Lindsley
-:date: 2013/04/28
+:date: 2013/07/28
 
 Haystack provides modular search for Django. It features a unified, familiar
 API that allows you to plug in different search backends (such as Solr_,
@@ -33,6 +33,7 @@ Documentation
 =============
 
 * Development version: http://docs.haystacksearch.org/
+* v2.1.X: http://django-haystack.readthedocs.org/en/v2.1.0/
 * v2.0.X: http://django-haystack.readthedocs.org/en/v2.0.0/
 * v1.2.X: http://django-haystack.readthedocs.org/en/v1.2.7/
 * v1.1.X: http://django-haystack.readthedocs.org/en/v1.1/
@@ -43,8 +44,8 @@ Requirements
 
 Haystack has a relatively easily-met set of requirements.
 
-* Python 2.5+
-* Django 1.3+
+* Python 2.6+ or Python 3.3+
+* Django 1.5+
 
 Additionally, each backend has its own requirements. You should refer to
 http://docs.haystacksearch.org/dev/installing_search_engines.html for more

diff --git a/docs/autocomplete.rst b/docs/autocomplete.rst
@@ -16,7 +16,7 @@ a small window passed over the string). Because this alters the way your
 data needs to be stored, the best approach is to add a new field to your
 ``SearchIndex`` that contains the text you want to autocomplete on.
 
-You have two choices: ``NgramField`` & ``EdgeNgramField``. Though very similar,
+You have two choices: ``NgramField`` and ``EdgeNgramField``. Though very similar,
 the choice of field is somewhat important.
 
 * If you're working with standard text, ``EdgeNgramField`` tokenizes on
@@ -54,14 +54,14 @@ Step 2. Performing The Query
 ============================
 
 Haystack ships with a convenience method to perform most autocomplete searches.
-You simply provide a field & the query you wish to search on to the
+You simply provide a field and the query you wish to search on to the
 ``SearchQuerySet.autocomplete`` method. Given the previous example, an example
 search would look like::
 
     from haystack.query import SearchQuerySet
 
     SearchQuerySet().autocomplete(content_auto='old')
-    # Result match things like 'goldfish', 'cuckold' & 'older'.
+    # Result match things like 'goldfish', 'cuckold' and 'older'.
 
 The results from the ``SearchQuerySet.autocomplete`` method are full search
 results, just like any regular filter.
@@ -82,14 +82,14 @@ Example Implementation
 
 The above is the low-level backend portion of how you implement autocomplete.
 To make it work in browser, you need both a view to run the autocomplete
-& some Javascript to fetch the results.
+and some Javascript to fetch the results.
 
 Since it comes up often, here is an example implementation of those things.
 
 .. warning::
 
     This code comes with no warranty. Don't ask for support on it. If you
-    copy-paste it & it burns down your server room, I'm not liable for any
+    copy-paste it and it burns down your server room, I'm not liable for any
     of it.
 
     It worked this one time on my machine in a simulated environment.
@@ -105,7 +105,7 @@ A stripped-down view might look like::
 
 
     def autocomplete(request):
-        sqs = SearchQuerySet().autocomplete(request.GET.get('q', ''))[:5]
+        sqs = SearchQuerySet().autocomplete(content_auto=request.GET.get('q', ''))[:5]
         suggestions = [result.title for result in sqs]
         # Make sure you return a JSON object, not a bare list.
         # Otherwise, you could be vulnerable to an XSS attack.
@@ -133,7 +133,7 @@ The template might look like::
       <script src="http://ajax.googleapis.com/ajax/libs/jquery/1.8.3/jquery.min.js"></script>
       <script type="text/javascript">
         // In a perfect world, this would be its own library file that got included
-        // on the page & only the ``$(document).ready(...)`` below would be present.
+        // on the page and only the ``$(document).ready(...)`` below would be present.
         // But this is an example.
         var Autocomplete = function(options) {
           this.form_selector = options.form_selector

diff --git a/docs/backend_support.rst b/docs/backend_support.rst
@@ -50,7 +50,7 @@ Elasticsearch
 * Stored (non-indexed) fields
 * Highlighting
 * Spatial search
-* Requires: pyelasticsearch 0.4+ & Elasticsearch 0.17.7+
+* Requires: elasticsearch-py 0.4.3+ & Elasticsearch 0.17.7+
 
 Whoosh
 ------
@@ -59,10 +59,11 @@ Whoosh
 
 * Full SearchQuerySet support
 * Automatic query building
+* "More Like This" functionality
 * Term Boosting
 * Stored (non-indexed) fields
 * Highlighting
-* Requires: whoosh (1.1.1+)
+* Requires: whoosh (2.0.0+)
 
 Xapian
 ------
@@ -89,7 +90,7 @@ Backend Support Matrix
 +----------------+------------------------+---------------------+----------------+------------+----------+---------------+--------------+---------+
 | Elasticsearch  | Yes                    | Yes                 | Yes            | Yes        | Yes      | Yes           | Yes          | Yes     |
 +----------------+------------------------+---------------------+----------------+------------+----------+---------------+--------------+---------+
-| Whoosh         | Yes                    | Yes                 | No             | Yes        | No       | Yes           | Yes          | No      |
+| Whoosh         | Yes                    | Yes                 | Yes            | Yes        | No       | Yes           | Yes          | No      |
 +----------------+------------------------+---------------------+----------------+------------+----------+---------------+--------------+---------+
 | Xapian         | Yes                    | Yes                 | Yes            | Yes        | Yes      | Yes           | Yes (plugin) | No      |
 +----------------+------------------------+---------------------+----------------+------------+----------+---------------+--------------+---------+

diff --git a/docs/conf.py b/docs/conf.py
@@ -38,16 +38,16 @@
 
 # General information about the project.
 project = u'Haystack'
-copyright = u'2009-2012, Daniel Lindsley'
+copyright = u'2009-2013, Daniel Lindsley'
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
 # built documents.
 #
 # The short X.Y version.
-version = '2.0.1'
+version = '2.1.1'
 # The full version, including alpha/beta/rc tags.
-release = '2.0.1-dev'
+release = '2.1.1-dev'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

diff --git a/docs/index.rst b/docs/index.rst
@@ -44,6 +44,7 @@ you up and running:
    debugging
 
    migration_from_1_to_2
+   python3
    contributing
 
 
@@ -109,24 +110,8 @@ Requirements
 
 Haystack has a relatively easily-met set of requirements.
 
-* Python 2.5+
-* Django 1.3+
+* Python 2.6+ or Python 3.3+
+* Django 1.5+
 
 Additionally, each backend has its own requirements. You should refer to
 :doc:`installing_search_engines` for more details.
-
-
-Commercial Support
-==================
-
-If you're using Haystack in a commercial environment, paid support is available
-from `Toast Driven`_. Services offered include:
-
-* Advice/help with setup
-* Implementation in your project
-* Bugfixes in Haystack itself
-* Features in Haystack itself
-
-If you're interested, please contact Daniel Lindsley (daniel@toastdriven.com).
-
-.. _`Toast Driven`: http://toastdriven.com/
diff --git a/docs/installing_search_engines.rst b/docs/installing_search_engines.rst
@@ -14,7 +14,7 @@ than the JRE and Jetty. It's very performant and has an advanced featureset.
 Haystack suggests using Solr 3.5+, though it's possible to get it working on
 Solr 1.4 with a little effort. Installation is relatively simple::
 
-    curl -O http://apache.mirrors.tds.net/lucene/solr/3.5.0/apache-solr-3.5.0.tgz
+    curl -O https://archive.apache.org/dist/lucene/solr/3.5.0/apache-solr-3.5.0.tgz
     tar xvzf apache-solr-3.5.0.tgz
     cd apache-solr-3.5.0
     cd example
@@ -152,17 +152,23 @@ locally. Modifications should be done in a YAML file, the stock one being
       logs: /usr/local/var/log
       data: /usr/local/var/data
 
-You'll also need an Elasticsearch binding: pyelasticsearch_ (**NOT**
-``pyes``). Place ``pyelasticsearch`` somewhere on your ``PYTHONPATH``
-(usually ``python setup.py install`` or ``pip install pyelasticsearch``).
+You'll also need an Elasticsearch binding: elasticsearch-py_ (**NOT**
+``pyes``). Place ``elasticsearch`` somewhere on your ``PYTHONPATH``
+(usually ``python setup.py install`` or ``pip install elasticsearch``).
 
-.. _pyelasticsearch: http://pypi.python.org/pypi/pyelasticsearch/
+.. _elasticsearch-py: http://pypi.python.org/pypi/elasticsearch/
 
 .. note::
+
+  Elasticsearch 1.0 is slightly backwards incompatible so you need to make sure
+  you have the proper version of `elasticsearch-py` installed - releases with
+  major version 1 (1.X.Y) are to be used with Elasticsearch 1.0 and later, 0.4
+  releases are meant to work with Elasticsearch 0.90.X.
 
-    ``pyelasticsearch`` has its own dependencies that aren't covered by
-    Haystack. You'll also need ``requests`` & ``simplejson`` for speedier
-    JSON construction/parsing.
+.. note::
+
+    ``elasticsearch`` has its own dependencies that aren't covered by
+    Haystack. You'll also need ``urllib3``.
 
 
 Whoosh

diff --git a/docs/other_apps.rst b/docs/other_apps.rst
@@ -28,6 +28,14 @@ Also provides a queue-based setup, this time centered around Celery. Useful
 for keeping the index fresh per model instance or with the included task
 to call the ``update_index`` management command instead.
 
+haystack-rqueue
+---------------
+
+https://github.com/mandx/haystack-rqueue (2.X compatible)
+
+Also provides a queue-based setup, this time centered around RQ. Useful
+for keeping the index fresh using ``./manage.py rqworker``.
+
 django-celery-haystack
 ----------------------
 

diff --git a/docs/python3.rst b/docs/python3.rst
@@ -0,0 +1,59 @@
+.. _ref-python3:
+
+================
+Python 3 Support
+================
+
+As of Haystack v2.1.0, it has been ported to support both Python 2 & Python 3
+within the same codebase. This builds on top of what `six`_ & `Django`_ provide.
+
+No changes are required for anyone running an existing Haystack
+installation. The API is completely backward-compatible, so you should be able
+to run your existing software without modification.
+
+Virtually all tests pass under both Python 2 & 3, with a small number of
+expected failures under Python (typically related to ordering, see below).
+
+.. _`six`: http://pythonhosted.org/six/
+.. _`Django`: https://docs.djangoproject.com/en/1.5/topics/python3/#str-and-unicode-methods
+
+
+Supported Backends
+==================
+
+The following backends are fully supported under Python 3. However, you may
+need to update these dependencies if you have a pre-existing setup.
+
+* Solr (pysolr>=3.1.0)
+* Elasticsearch
+
+
+Partially Supported Backends
+============================
+
+Most aspects of Whoosh work fine under Python 3. However, the built-in
+highlighting fails under Python 3.3.2. Using the Haystack-provided template tag
+highlighting is a good stopgap until
+https://bitbucket.org/mchaput/whoosh/324/highlighting-under-python-3-fails is
+resolved.
+
+* Whoosh (whoosh>=2.5.0)
+
+
+Notes
+=====
+
+Testing
+-------
+
+If you were testing things such as the query generated by a given
+``SearchQuerySet`` or how your forms would render, under Python 3.3.2+,
+`hash randomization`_ is in effect, which means that the ordering of
+dictionaries is no longer consistent, even on the same platform.
+
+Haystack took the approach of abandoning making assertions about the entire
+structure. Instead, we either simply assert that the new object contains the
+right things or make a call to ``sorted(...)`` around it to ensure order. It is
+recommended you take a similar approach.
+
+.. _`hash randomization`: http://docs.python.org/3/whatsnew/3.3.html#builtin-functions-and-types
diff --git a/docs/running_tests.rst b/docs/running_tests.rst
@@ -78,11 +78,7 @@ Configuring Solr
 ----------------
 
 Haystack assumes that you have a Solr server running on port ``9001`` which uses the schema and
-configuration provided in the ``tests/`` directory. Currently, these steps will result in a working
-test server:
-
-#. Download the current Solr release from http://lucene.apache.org/solr/
-#. Copy ``tests/solrconfig.xml`` to ``example/solr/conf/solrconfig.xml``
-#. Copy ``tests/solr_test_schema.xml`` to ``example/solr/conf/schema.xml``
-#. Change into the ``example`` directory
-#. Start Solr: ``java -Djetty.port=9001 -jar start.jar``
+configuration provided in the ``tests/`` directory. For convenience, a script is provided which
+will download, configure and start a test Solr server::
+
+    tests/start-solr-test-server.sh