jekyll_search 0.0.3 → 0.0.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: e903bb117234165daca55baee25cc4a352e5cd91
-  data.tar.gz: db9eb1cf7534ec9929057194af1e0248b52dc243
+  metadata.gz: 93f638b21ff6bac656f0b96a2fb8976fec06b956
+  data.tar.gz: 83ed6b3e6770ad95b9525e73832c17be3098e650
 SHA512:
-  metadata.gz: 7e6f6263cb8391612c44c4f81b1cb24285f9bdf68b910ae3f2bbe6f2d71eccede16a95cf61648b3e576fc3368766caaf7fe70706a75277e2d2ca2b72486a200d
-  data.tar.gz: 84529eaef4b0b389bb13bfb1d11ed6ad3bd22e997bef30c59eef4ff18c4706d593ae04f28934939cc568d5c22e6716b6d9abe6244db9b2049f795560187b0d31
+  metadata.gz: 8e78345248b71e96def1cf31dd344fba3f70b62c9d14ccc3e45244bd36e04abd25dc03ff6195aee41e3bb55674c6a185259704013c8cd39b5b715be48a3f5f39
+  data.tar.gz: 10d31a2a55029a583ce597dc4ba30c4b53d0bb3d91dced2db670173d10b32d796936af3979bef62ba2a110921679f59fb7d199eff0ea663ab47cf6173e8a8bb8

data/README.md

@@ -42,10 +42,59 @@ search:
 ```
 
 Now run `jekyll index` to iterate over all pages and index them with Elasticsearch. With `jekyll search my query`
-you can throw some test searches against your freshly created search index.
+you can throw some test searches against your freshly created search index. When you plan to integrate a AJAX
+based search into your Jekyll page, then the following `curl` example should help to get started (see also [here][elasticsearch-searchapi]):
+
+```text
+curl -XPOST localhost:9200/myindex/section/_search?pretty -d '{
+  "query": {
+    "match_phrase_prefix": {
+      "content": {
+        "query": "I SEARCH FOR SOMETHING",
+        "slop": 10
+      }
+    }
+  },
+  "highlight": {
+    "fields": {
+      "content": {}
+    }
+  }
+}'
+```
+
+```json
+{
+  // ...
+  "hits": {
+    "total": 7,
+    "max_score": 0.61569,
+    "hits": [
+      {
+        // ...
+        "_score" : 0.61569,
+        "_source": {
+          "url": "/link-to-page.html#headline-id",
+          "title": "The title",
+          "content": "The content"
+        },
+        "highlight": {
+          "content": [
+            "Some <em>highlighted</em> stuff"
+          ]
+        }
+      }
+      // ...
+    ]
+  }
+  // ...
+}
+```
+
+## Customize search index
 
 If you want to customize how Elasticsearch creates the search index, then provide an additional `index.settings`
-property in your `_config.yml` (see [here][elasticsearch-createindex]:
+property in your `_config.yml` (see also [here][elasticsearch-createindex]):
 
 ```yaml
 # Search index settings
@@ -66,6 +115,17 @@ search:
             content:
               type: string
               analyzer: english
+        section:
+          properties:
+            url:
+              type: string
+              analyzer: keyword
+            title:
+              type: string
+              analyzer: english
+            content:
+              type: string
+              analyzer: english
 ```
 
 ## Contributing
@@ -77,4 +137,5 @@ search:
 5. Create a new Pull Request
 
 [elasticsearch]: http://www.elasticsearch.org/
+[elasticsearch-searchapi]: http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/search-search.html
 [elasticsearch-createindex]: http://www.rubydoc.info/gems/elasticsearch-api/Elasticsearch/API/Indices/Actions#create-instance_method

data/lib/jekyll_search/version.rb

@@ -1,3 +1,3 @@
 module JekyllSearch
-  VERSION = "0.0.3"
+  VERSION = "0.0.4"
 end

data/lib/jekyll_search.rb

@@ -31,23 +31,22 @@ module Jekyll
           client = Elasticsearch::Client.new host: settings['host'], log: false
           create_index(client, settings)
 
-          pages = site.pages.
-              select { |p| p.url =~ /\.html$/ }.
-              select { |p| p.data['searchable'].nil? or p.data['searchable'] != false }
+          pages = collect_content(site, site.pages)
+          posts = collect_content(site, site.posts)
 
-          for page in pages
+          for page in pages + posts
             page_body = {
-              url: site.baseurl + page.url,
-              title: page.data['title'],
-              content: JekyllSearch::HtmlProcessor.strip_html(page.content)
+              url: page[:url],
+              title: page[:title],
+              content: JekyllSearch::HtmlProcessor.strip_html(page[:content])
             }
 
             client.index index: settings['index']['name'], type: 'page', body: page_body
 
-            for section in JekyllSearch::HtmlProcessor.detect_sections(page.content)
+            for section in JekyllSearch::HtmlProcessor.detect_sections(page[:content])
               section_body = {
-                url: if section[:id] != nil then site.baseurl + page.url + '#' + section[:id] else site.baseurl + page.url end,
-                title: if section[:title] != nil then section[:title] else page.data['title'] end,
+                url: if section[:id] != nil then site.baseurl + page[:url] + '#' + section[:id] else site.baseurl + page[:url] end,
+                title: if section[:title] != nil then section[:title] else page[:title] end,
                 content: JekyllSearch::HtmlProcessor.strip_html(section[:content])
               }
 
@@ -63,6 +62,19 @@ module Jekyll
 
           client.indices.create index: settings['index']['name'], body: (settings['index']['settings'] or {})
         end
+
+        def collect_content(site, elements)
+          elements.
+            select { |p| p.url =~ /\.html$/ }.
+            select { |p| p.data['searchable'].nil? or p.data['searchable'] != false }.
+            map do |p|
+              {
+                :url => site.baseurl + p.url,
+                :title => p.data['title'],
+                :content => p.content
+              }
+            end
+        end
       end
     end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: jekyll_search
 version: !ruby/object:Gem::Version
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors:
 - Christian Hoffmeister