Merge pull request #599 from liangliangyy/dev

feat: 新增 elasticsearch 搜索 suggest_search 拼写纠正功能

Author: liangliangyy

blog/views.py

@@ -11,10 +11,11 @@
 from django.views.decorators.csrf import csrf_exempt
 from django.views.generic.detail import DetailView
 from django.views.generic.list import ListView
+from haystack.views import SearchView
 
-from blog.models import Article, Category, Tag, Links, LinkShowType
+from blog.models import Article, Category, LinkShowType, Links, Tag
 from comments.forms import CommentForm
-from djangoblog.utils import cache, get_sha256, get_blog_setting
+from djangoblog.utils import cache, get_blog_setting, get_sha256
 
 logger = logging.getLogger(__name__)
 
@@ -267,6 +268,23 @@ def get_queryset(self):
         return Links.objects.filter(is_enable=True)
 
 
+class EsSearchView(SearchView):
+    def get_context(self):
+        paginator, page = self.build_page()
+        context = {
+            "query": self.query,
+            "form": self.form,
+            "page": page,
+            "paginator": paginator,
+            "suggestion": None,
+        }
+        if hasattr(self.results, "query") and self.results.query.backend.include_spelling:
+            context["suggestion"] = self.results.query.get_spelling_suggestion()
+        context.update(self.extra_context())
+
+        return context
+
+
 @csrf_exempt
 def fileupload(request):
     """

djangoblog/elasticsearch_backend.py

@@ -1,6 +1,7 @@
 from django.utils.encoding import force_str
 from elasticsearch_dsl import Q
 from haystack.backends import BaseEngine, BaseSearchBackend, BaseSearchQuery, log_query
+from haystack.forms import ModelSearchForm
 from haystack.models import SearchResult
 from haystack.utils import log as logging
 
@@ -18,6 +19,7 @@ def __init__(self, connection_alias, **connection_options):
             connection_alias,
             **connection_options)
         self.manager = ArticleDocumentManager()
+        self.include_spelling = True
 
     def _get_models(self, iterable):
         models = iterable if iterable and iterable[0] else Article.objects.all()
@@ -51,15 +53,40 @@ def remove(self, obj_or_string):
     def clear(self, models=None, commit=True):
         self.remove(None)
 
+    @staticmethod
+    def get_suggestion(query: str) -> str:
+        """获取推荐词, 如果没有找到添加原搜索词"""
+
+        search = ArticleDocument.search() \
+            .query("match", body=query) \
+            .suggest('suggest_search', query, term={'field': 'body'}) \
+            .execute()
+
+        keywords = []
+        for suggest in search.suggest.suggest_search:
+            if suggest["options"]:
+                keywords.append(suggest["options"][0]["text"])
+            else:
+                keywords.append(suggest["text"])
+
+        return ' '.join(keywords)
+
     @log_query
     def search(self, query_string, **kwargs):
         logger.info('search query_string:' + query_string)
 
         start_offset = kwargs.get('start_offset')
         end_offset = kwargs.get('end_offset')
 
-        q = Q('bool', should=[Q('match', body=query_string), Q(
-            'match', title=query_string)], minimum_should_match="70%")
+        # 推荐词搜索
+        if getattr(self, "is_suggest", None):
+            suggestion = self.get_suggestion(query_string)
+        else:
+            suggestion = query_string
+
+        q = Q('bool',
+              should=[Q('match', body=suggestion), Q('match', title=suggestion)],
+              minimum_should_match="70%")
 
         search = ArticleDocument.search() \
                      .query('bool', filter=[q]) \
@@ -85,7 +112,7 @@ def search(self, query_string, **kwargs):
                 **additional_fields)
             raw_results.append(result)
         facets = {}
-        spelling_suggestion = None
+        spelling_suggestion = None if query_string == suggestion else suggestion
 
         return {
             'results': raw_results,
@@ -134,6 +161,22 @@ def get_count(self):
         results = self.get_results()
         return len(results) if results else 0
 
+    def get_spelling_suggestion(self, preferred_query=None):
+        return self._spelling_suggestion
+
+    def build_params(self, spelling_query=None):
+        kwargs = super(ElasticSearchQuery, self).build_params(spelling_query=spelling_query)
+        return kwargs
+
+
+class ElasticSearchModelSearchForm(ModelSearchForm):
+
+    def search(self):
+        # 是否建议搜索
+        self.searchqueryset.query.backend.is_suggest = self.data.get("is_suggest") != "no"
+        sqs = super().search()
+        return sqs
+
 
 class ElasticSearchEngine(BaseEngine):
     backend = ElasticSearchBackend

djangoblog/urls.py

@@ -14,14 +14,17 @@
     2. Add a URL to urlpatterns:  url(r'^blog/', include('blog.urls'))
 """
 from django.conf import settings
-from django.urls import include, re_path
 from django.conf.urls.static import static
 from django.contrib.sitemaps.views import sitemap
 from django.urls import include
+from django.urls import re_path
+from haystack.views import search_view_factory
 
+from blog.views import EsSearchView
 from djangoblog.admin_site import admin_site
+from djangoblog.elasticsearch_backend import ElasticSearchModelSearchForm
 from djangoblog.feeds import DjangoBlogFeed
-from djangoblog.sitemap import StaticViewSitemap, ArticleSiteMap, CategorySiteMap, TagSiteMap, UserSiteMap
+from djangoblog.sitemap import ArticleSiteMap, CategorySiteMap, StaticViewSitemap, TagSiteMap, UserSiteMap
 
 sitemaps = {
 
@@ -43,10 +46,11 @@
                   re_path(r'', include('accounts.urls', namespace='account')),
                   re_path(r'', include('oauth.urls', namespace='oauth')),
                   re_path(r'^sitemap\.xml$', sitemap, {'sitemaps': sitemaps},
-                      name='django.contrib.sitemaps.views.sitemap'),
+                          name='django.contrib.sitemaps.views.sitemap'),
                   re_path(r'^feed/$', DjangoBlogFeed()),
                   re_path(r'^rss/$', DjangoBlogFeed()),
-                  re_path(r'^search', include('haystack.urls'), name='search'),
+                  re_path('^search', search_view_factory(view_class=EsSearchView, form_class=ElasticSearchModelSearchForm),
+                          name='search'),
                   re_path(r'', include('servermanager.urls', namespace='servermanager')),
                   re_path(r'', include('owntracks.urls', namespace='owntracks'))
               ] + static(settings.STATIC_URL, document_root=settings.STATIC_ROOT)

requirements.txt

@@ -1,7 +1,7 @@
 coverage==6.4
 bleach==5.0.0
-Django==4.0.7
-django-compressor==4.0
+Django==4.1
+django-compressor==4.1
 django-haystack==3.2.1
 django-ipware==4.0.2
 django-mdeditor==0.1.20

templates/search/search.html

@@ -15,8 +15,16 @@
         <div id="content" role="main">
             {% if query %}
                 <header class="archive-header">
-
-                    <h2 class="archive-title"> 搜索：<span style="color: red">{{ query }}</span></h2>
+                    {% if suggestion %}
+                        <h2 class="archive-title">
+                            已显示<span style="color: red"> “{{ suggestion }}” </span>的搜索结果。&nbsp;&nbsp;
+                            仍然搜索：<a style="text-transform: none;" href="/search/?q={{ query }}&is_suggest=no">{{ query }}</a> <br>
+                        </h2>
+                    {% else %}
+                        <h2 class="archive-title">
+                            搜索：<span style="color: red">{{ query }} </span> &nbsp;&nbsp;
+                        </h2>
+                    {% endif %}
                 </header><!-- .archive-header -->
             {% endif %}
             {% if query and page.object_list %}