shrishty / djangosearch Goto Github PK

trunk patch: https://gist.github.com/9397c1bd0e56dbd8df50
branch patch: https://gist.github.com/3916123cd0328fe47907

I'd like to be able to reindex just a single model at times, such as when I
make an update to that model's index template.  I've included a patch to
the reindex management command to make this possible.

djangosearch does not accept non-latin characters as query

from djangosearch import search
search("ต้น")

The problem cause since Word does not accept non-latin character.

indexer.get_indexers currently takes a model argument, which isn't used and
doesn't seem valid based on what get_indexers is supposed to do.

What steps will reproduce the problem?
1. The test suite is broken when using solr

I am attaching a diff of a reorganized  test suite.
>>> Article.index.get_indexed_fields(a)
[('title', u'test'), ('date', u'2007-10-31 00:00:00')]

As of now the expected date is  datetime.datetime(2007, 10, 31, 0, 0) but
atleast the solr backend does not return it in this format.



What version of the product are you using? On what operating system?
recent svn checkout

Please provide any additional information below.

At Pycon, we discussed adding [GrassyKnoll http://grassyknoll.googlecode.com/] 
support to Django 
Search.   We're tracking this in our bug tracker: 
http://code.google.com/p/grassyknoll/issues/detail?id=97

Please post comments there - just want to create a stub.

When constructing search terms, djangosearch restricts itself, rather 
arbitrarily, to ASCII 
alphanumerics, plus what pyparsing calls alphas8bit, which gives a few European 
accented 
characters.

It would be nicer if it let itself use all unicode alphanumerics. The attached 
patch does this (and 
also adds unicode symbols, which might or might not useful, it is for me)

There is no documentation, please add some

Given a badly constructed query (eg "(((") djangosearch fails by throwing an 
internal exception from 
pyparsing. it would be rather nicer if it threw an exception of its own, which 
I could catch from the 
public API.

The attached patch (against soc-new-backends) does this.

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?
There is not clear documentation.

What version of the product are you using? On what operating system?
using djagosearch on ubuntu

Please provide any additional information below.
when i used the same example of given in readme.txt and wrote
import djangosearch
from djangosearch import ModelIndex
class Event1(models.Model):
       title = models.CharField(max_length=255)
       date = models.DateField()
       is_outdoors = models.BooleanField()

       index = djangosearch.ModelIndex(fields=['title', 'date'])

it raise an error:
 raise ImproperlyConfigured("You must define the SEARCH_ENGINE setting
before using the search framework.")
django.core.exceptions.ImproperlyConfigured: You must define the
SEARCH_ENGINE setting before using the search framework

i do not know how may i configure SEARCH_ENGINE. i am using mysql.

During large imports of data running through models, having per-object save
update the search index is painfully slow.  Ideally I'd like to disable the
updates that save() and delete() call in favor of running a batch reindex
sometime after I've imported data.

I've attached a patch that allows for this functionality.  From the model
class, a user can set the indexer's 'do_updates' property to False, which
will cause update_object/remove_object to not run.  I'm not wild about the
'do_updates' name, so feel free to change it to something more clever.

What steps will reproduce the problem?
1. Adding   index = djangosearch.ModelIndex(text=['field1'], 
additional=['field2', 'field3']) to 
any model
2. Running the development server or syncdb

What is the expected output? What do you see instead?
I'm following the instructions at: 
http://code.google.com/p/djangosearch/source/browse/branches/soc-new-
backends/doc/README.txt?r=18.  I would expect it to at least run without 
errors, but I get the 
same error every time: TypeError: __init__() got an unexpected keyword argument 
'text'.  If I 
remove both 'text=' and 'additional=' I get the following error: 
"ImproperlyConfigured: You must 
define the SEARCH_ENGINE setting before using the search framework." However, I 
just want to 
use MySQL for searching, and not one of the extra backends, so I'm not sure 
what to put here.  
The documentation is not very clear.


What version of the product are you using? On what operating system?
I'm using r19 on leopard using the django development server

Please provide any additional information below.
I'm using MySQL as my db, and I don't need any of the extra search backends.  

change in backends/solr.py to:

def _result_callback(self, result):
        app_label, model_name = result['django_ct_s'].split('.')
        return (app_label, model_name, result['django_id_s'], result['score'])

currently it reads:

def _result_callback(self, result):
        app_label, model_name = result['django_ct_s'].split('.')
        return (app_label, model_name, result['django_id_s'], None)

When looking at a result in a Django template, hitting result._relevance throws 
a 
TempalteSyntaxError: Variables and attributes may not begin with underscores: 
'result._relevance'

Is there any way around this?

What steps will reproduce the problem?
1. from djangosearch import search
2. search(query="query_string timestamp:[2009-01-08T20:16:36.1Z TO
2009-01-08T20:16:36.500Z]").query


What is the expected output? 
query_string AND timestamp:[2009-01-08T20:16:36.1Z TO 2009-01-08T20:16:36.500Z]

What do you see instead?
query_string

What version of the product are you using? On what operating system?
trunk

Please provide any additional information below.
I am suggesting the addition of a parameter in search called
"query_abstraction_layer". The default will be True that means that search
will take advantage of "convert_query" and if set to False it will use the
original query string.

This approach will enable the use of all the specificities of the backend.
--yml