Login

Haystack whoosh backend with custom analyzer (allow using any lang, word processing, etc)

Author:
sakkada
Posted:
February 4, 2016
Language:
Python
Version:
1.7
Score:
0 (after 0 ratings)

It is a haystack custom whoosh backend which provides analyzer customisation (required get_analyzer method definition). That means it is possible to define any complex analyzers (see whoosh docs and source).

Sample code shows how to use it, builtin LanguageAnalyzer instance configured for working with russian language used as custom analyzer.

Base idea taken from this snippet: https://djangosnippets.org/snippets/3025/

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
# encoding: utf-8
from django.utils.datetime_safe import datetime

from haystack.constants import DJANGO_CT, DJANGO_ID, ID
from haystack.exceptions import MissingDependency, SearchBackendError, SkipDocument
from haystack.models import SearchResult
from haystack.utils.app_loading import haystack_get_model
from haystack.backends.whoosh_backend import (WhooshHtmlFormatter,
                                              WhooshSearchBackend,
                                              WhooshEngine) # fixed

from whoosh.fields import ID as WHOOSH_ID
from whoosh.fields import BOOLEAN, DATETIME, IDLIST, KEYWORD, NGRAM, NGRAMWORDS, NUMERIC, Schema, TEXT
from whoosh.highlight import highlight as whoosh_highlight
from whoosh.highlight import ContextFragmenter, HtmlFormatter
from whoosh.analysis.analyzers import LanguageAnalyzer # fixed


# fixes: allow to customize analyzer for whoosh indexer
#        code of build_schema and _process_results methods completely taken
#        from haystack except lines marked "# fixed" comment
#        get_analyzer method created for extending and should return
#        any custom analyzer

class CustomAnalyzerWhooshSearchBackend(WhooshSearchBackend):
    def get_analyzer(self):
        # note: extend it to set custom analyzer
        raise NotImplementedError('No analyzer defined')

    def build_schema(self, fields):
        schema_fields = {
            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.
        initial_key_count = len(schema_fields)
        content_field_name = ''

        for field_name, field_class in fields.items():
            if field_class.is_multivalued:
                if field_class.indexed is False:
                    schema_fields[field_class.index_fieldname] = IDLIST(stored=True, field_boost=field_class.boost)
                else:
                    schema_fields[field_class.index_fieldname] = KEYWORD(stored=True, commas=True, scorable=True, field_boost=field_class.boost)
            elif field_class.field_type in ['date', 'datetime']:
                schema_fields[field_class.index_fieldname] = DATETIME(stored=field_class.stored, sortable=True)
            elif field_class.field_type == 'integer':
                schema_fields[field_class.index_fieldname] = NUMERIC(stored=field_class.stored, numtype=int, field_boost=field_class.boost)
            elif field_class.field_type == 'float':
                schema_fields[field_class.index_fieldname] = NUMERIC(stored=field_class.stored, numtype=float, field_boost=field_class.boost)
            elif field_class.field_type == 'boolean':
                # Field boost isn't supported on BOOLEAN as of 1.8.2.
                schema_fields[field_class.index_fieldname] = BOOLEAN(stored=field_class.stored)
            elif field_class.field_type == 'ngram':
                schema_fields[field_class.index_fieldname] = NGRAM(minsize=3, maxsize=15, stored=field_class.stored, field_boost=field_class.boost)
            elif field_class.field_type == 'edge_ngram':
                schema_fields[field_class.index_fieldname] = NGRAMWORDS(minsize=2, maxsize=15, at='start', stored=field_class.stored, field_boost=field_class.boost)
            else:
                schema_fields[field_class.index_fieldname] = TEXT(stored=True, analyzer=self.get_analyzer(), field_boost=field_class.boost, sortable=True) # fixed

            if field_class.document is True:
                content_field_name = field_class.index_fieldname
                schema_fields[field_class.index_fieldname].spelling = True

        # Fail more gracefully than relying on the backend to die if no fields
        # are found.
        if len(schema_fields) <= initial_key_count:
            raise SearchBackendError("No fields were found in any search_indexes. Please correct this before attempting to search.")

        return (content_field_name, Schema(**schema_fields))

    def _process_results(self, raw_page, highlight=False, query_string='', spelling_query=None, result_class=None):
        from haystack import connections
        results = []

        # It's important to grab the hits first before slicing. Otherwise, this
        # can cause pagination failures.
        hits = len(raw_page)

        if result_class is None:
            result_class = SearchResult

        facets = {}
        spelling_suggestion = None
        unified_index = connections[self.connection_alias].get_unified_index()
        indexed_models = unified_index.get_indexed_models()

        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('.')
            additional_fields = {}
            model = haystack_get_model(app_label, model_name)

            if model and model in indexed_models:
                for key, value in raw_result.items():
                    index = unified_index.get_index(model)
                    string_key = str(key)

                    if string_key in index.fields and hasattr(index.fields[string_key], 'convert'):
                        # Special-cased due to the nature of KEYWORD fields.
                        if index.fields[string_key].is_multivalued:
                            if value is None or len(value) is 0:
                                additional_fields[string_key] = []
                            else:
                                additional_fields[string_key] = value.split(',')
                        else:
                            additional_fields[string_key] = index.fields[string_key].convert(value)
                    else:
                        additional_fields[string_key] = self._to_python(value)

                del(additional_fields[DJANGO_CT])
                del(additional_fields[DJANGO_ID])

                if highlight:
                    sa = self.get_analyzer() # fixed
                    formatter = WhooshHtmlFormatter('em')
                    terms = [token.text for token in sa(query_string)]

                    whoosh_result = whoosh_highlight(
                        additional_fields.get(self.content_field_name),
                        terms,
                        sa,
                        ContextFragmenter(),
                        formatter
                    )
                    additional_fields['highlighted'] = {
                        self.content_field_name: [whoosh_result],
                    }

                result = result_class(app_label, model_name, raw_result[DJANGO_ID], score, **additional_fields)
                results.append(result)
            else:
                hits -= 1

        if self.include_spelling:
            if spelling_query:
                spelling_suggestion = self.create_spelling_suggestion(spelling_query)
            else:
                spelling_suggestion = self.create_spelling_suggestion(query_string)

        return {
            'results': results,
            'hits': hits,
            'facets': facets,
            'spelling_suggestion': spelling_suggestion,
        }


# sample usage
class RussianAnalyzerWhooshSearchBackend(CustomAnalyzerWhooshSearchBackend):
    def get_analyzer(self):
        return LanguageAnalyzer('ru')


class CustomAnalyzerWhooshEngine(WhooshEngine):
    backend = RussianAnalyzerWhooshSearchBackend

More like this

  1. Month / Year SelectDateWidget based on django SelectDateWidget by pierreben 2 months, 4 weeks ago
  2. Python Django CRUD Example Tutorial by tuts_station 3 months, 2 weeks ago
  3. Browser-native date input field by kytta 4 months, 4 weeks ago
  4. Generate and render HTML Table by LLyaudet 5 months, 1 week ago
  5. My firs Snippets by GutemaG 5 months, 1 week ago

Comments

Please login first before commenting.