middleman-search 0.8.0 → 0.9.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: e7801c0f17626f09057c99a551a7cd9c1e462dc1
-  data.tar.gz: b7b1e6fb3098d533c073d25923916aa3d42181d9
+  metadata.gz: 4fc0d3fb638aed4295e5f062bee990b4dd5b17fa
+  data.tar.gz: b9bce741409fcfba231c0de4db86c85f4bad9077
 SHA512:
-  metadata.gz: 254aeb5de6fb9687de60ebb47e7a01f6d49b6f01dbea6cc5b8fe4b191d52ee86bf83d07cc3cc105b2ea95b7ad4a12eb3e48c7aa6199a2422adf51aed8a11c7de
-  data.tar.gz: 6c33807e8c13e227b71f4a4bf76c7330800b3676af08812f2eb241a52136a96089a5dbbf1a75bc481cea969df4b98fd24bc0400c45274a238b9834c0c3418833
+  metadata.gz: 6e07b8c657510273d73c1a815e1bd34cb50398ed173cfe9b1be48edac9b31c48ab6985241a9df50eb781d0bc0c5788f4af94eadfcede20eeeecff926c70e68e1
+  data.tar.gz: 39511db8a4ad2151f58277796e0c545c40e5baf211b53aec91378ddac34d71ab55e3e5390cced788f471d745b43c62c5baf223be70b67572e894ed28c44e0cb9

data/README.md

@@ -26,6 +26,10 @@ activate :search do |search|
   search.resources = ['blog/', 'index.html', 'contactus/index.html']
 
   search.index_path = 'search/lunr-index.json' # defaults to `search.json`
+  
+  search.lunr_dirs = ['source/vendor/lunr-custom/'] # optional alternate paths where to look for lunr js files
+
+  search.language = 'es' # defaults to 'en'
 
   search.fields = {
     title:   {boost: 100, store: true, required: true},
@@ -49,6 +53,12 @@ All fields values are retrieved from the resource `data` (i.e. its frontmatter),
 - `url` which is the actual resource url
 - `content` the text extracted from the rendered resource, without including its layout
 
+### i18n
+
+This gem includes assets for alternate languages as provided by [MihaiValentin/lunr-languages](https://github.com/MihaiValentin/lunr-languages). Please refer to that repository for a list of the languages available.
+
+If you want to work with a language that is not included, set up a `lunr.yourlang.js` file in a folder in your project, and add that folder to `lunr_dirs` so the gem knows where to look for it.
+
 ### Manual index manipulation
 
 You can fully customise the content to be indexed and stored per resource by defining a `before_index` callback:
@@ -146,4 +156,5 @@ A big thank you to:
 - [Octo-Labs](https://github.com/Octo-Labs)'s [jagthedrummer](https://github.com/jagthedrummer) for his [`middleman-alias`](https://github.com/Octo-Labs/middleman-alias) extension, in which we based for developing this one.
 - [jnovos](https://github.com/jnovos) and [256dpi](https://github.com/256dpi), for their [`middleman-lunrjs`](https://github.com/jnovos/middleman-lunrjs) and [`middleman-lunr`](https://github.com/256dpi/middleman-lunr) extensions, which served as inspirations for making this one.
 - [olivernn](https://github.com/olivernn) and all [`lunr.js`](http://lunrjs.com/) [contributors](https://github.com/olivernn/lunr.js/graphs/contributors)
+- [MihaiValentin](https://github.com/MihaiValentin) for the support for 10+ languages in [lunr-languages](https://github.com/MihaiValentin/lunr-languages).
 - [The Middleman](https://middlemanapp.com/) [team](https://github.com/orgs/middleman/people) and [contributors](https://github.com/middleman/middleman/graphs/contributors)

data/lib/middleman-search/extension.rb

@@ -9,6 +9,8 @@ module Middleman
     option :index_path, 'search.json', 'Index file path'
     option :pipeline, {}, 'Javascript pipeline functions to use in lunr index'
     option :cache, false, 'Avoid the cache to be rebuilt on every request in development mode'
+    option :language, 'en', 'Language code ("es", "fr") to use when indexing site\'s content'
+    option :lunr_dirs, [], 'Directories in which to look for custom lunr.js files'
 
     def manipulate_resource_list(resources)
       resources.push Middleman::Sitemap::SearchIndexResource.new(@app.sitemap, @options[:index_path], @options)

data/lib/middleman-search/search-index-resource.rb

@@ -9,6 +9,8 @@ module Middleman
         @callback = options[:before_index]
         @pipeline = options[:pipeline]
         @cache_index = options[:cache]
+        @language = options[:language]
+        @lunr_dirs = options[:lunr_dirs] + [File.expand_path("../../../vendor/assets/javascripts/", __FILE__)]
         super(store, path)
       end
 
@@ -31,7 +33,14 @@ module Middleman
       def build_index
         # Build js context
         context = V8::Context.new
-        context.load(File.expand_path('../../../vendor/assets/javascripts/lunr.min.js', __FILE__))
+        context.load(lunr_resource('lunr.js'))
+
+        if @language != 'en' # English is the default
+          context.load(lunr_resource("lunr.stemmer.support.js"))
+          context.load(lunr_resource("lunr.#{@language}.js"))
+          lunr_lang = context.eval("lunr.#{@language}")
+        end
+
         context.eval('lunr.Index.prototype.indexJson = function () {return JSON.stringify(this.toJSON());}')
 
         # Register pipeline functions
@@ -54,6 +63,7 @@ module Middleman
           end
 
           # Define fields with boost
+          this.use(lunr_lang) if @language
           @fields.each do |field, opts|
             next if opts[:index] == false
             this.field(field, {:boost => opts[:boost]})
@@ -118,6 +128,20 @@ module Middleman
           value ? Array(value).compact.join(" ") : nil
         end
       end
+
+      private
+
+      def minified_path(resource_name)
+        return resource_name if resource_name.end_with? '.min.js'
+        return resource_name unless resource_name.end_with? '.js'
+        resource_name.sub(/(.*)\.js$/,'\1.min.js')
+      end
+
+      def lunr_resource(resource_name)
+        @lunr_dirs.flat_map do |dir|
+          [File.join(dir, minified_path(resource_name)), File.join(dir, resource_name)]
+        end.detect { |file| File.exists? file } or raise "Couldn't find #{resource_name} nor #{minified_path(resource_name)} in #{@lunr_dirs.map {|dir| File.absolute_path dir }.join File::PATH_SEPARATOR}"
+      end
     end
   end
 end

data/lib/middleman-search/version.rb

@@ -1,3 +1,3 @@
 module MiddlemanSearch
-  VERSION = "0.8.0"
+  VERSION = "0.9.0"
 end

data/vendor/assets/javascripts/lunr.da.js

@@ -0,0 +1,273 @@
+/*!
+ * Lunr languages, `Danish` language
+ * https://github.com/MihaiValentin/lunr-languages
+ *
+ * Copyright 2014, Mihai Valentin
+ * http://www.mozilla.org/MPL/
+ */
+/*!
+ * based on
+ * Snowball JavaScript Library v0.3
+ * http://code.google.com/p/urim/
+ * http://snowball.tartarus.org/
+ *
+ * Copyright 2010, Oleg Mazko
+ * http://www.mozilla.org/MPL/
+ */
+
+/**
+ * export the module via AMD, CommonJS or as a browser global
+ * Export code from https://github.com/umdjs/umd/blob/master/returnExports.js
+ */
+;
+(function(root, factory) {
+  if (typeof define === 'function' && define.amd) {
+    // AMD. Register as an anonymous module.
+    define(factory)
+  } else if (typeof exports === 'object') {
+    /**
+     * Node. Does not work with strict CommonJS, but
+     * only CommonJS-like environments that support module.exports,
+     * like Node.
+     */
+    module.exports = factory()
+  } else {
+    // Browser globals (root is window)
+    factory()(root.lunr);
+  }
+}(this, function() {
+  /**
+   * Just return a value to define the module export.
+   * This example returns an object, but the module
+   * can return a function as the exported value.
+   */
+  return function(lunr) {
+    /* throw error if lunr is not yet included */
+    if ('undefined' === typeof lunr) {
+      throw new Error('Lunr is not present. Please include / require Lunr before this script.');
+    }
+
+    /* throw error if lunr stemmer support is not yet included */
+    if ('undefined' === typeof lunr.stemmerSupport) {
+      throw new Error('Lunr stemmer support is not present. Please include / require Lunr stemmer support before this script.');
+    }
+
+    /* register specific locale function */
+    lunr.da = function() {
+      this.pipeline.reset();
+      this.pipeline.add(
+        lunr.da.stopWordFilter,
+        lunr.da.stemmer
+      );
+    };
+
+    /* lunr stemmer function */
+    lunr.da.stemmer = (function() {
+      /* create the wrapped stemmer object */
+      var Among = lunr.stemmerSupport.Among,
+        SnowballProgram = lunr.stemmerSupport.SnowballProgram,
+        st = new function DanishStemmer() {
+          var a_0 = [new Among("hed", -1, 1), new Among("ethed", 0, 1),
+              new Among("ered", -1, 1), new Among("e", -1, 1),
+              new Among("erede", 3, 1), new Among("ende", 3, 1),
+              new Among("erende", 5, 1), new Among("ene", 3, 1),
+              new Among("erne", 3, 1), new Among("ere", 3, 1),
+              new Among("en", -1, 1), new Among("heden", 10, 1),
+              new Among("eren", 10, 1), new Among("er", -1, 1),
+              new Among("heder", 13, 1), new Among("erer", 13, 1),
+              new Among("s", -1, 2), new Among("heds", 16, 1),
+              new Among("es", 16, 1), new Among("endes", 18, 1),
+              new Among("erendes", 19, 1), new Among("enes", 18, 1),
+              new Among("ernes", 18, 1), new Among("eres", 18, 1),
+              new Among("ens", 16, 1), new Among("hedens", 24, 1),
+              new Among("erens", 24, 1), new Among("ers", 16, 1),
+              new Among("ets", 16, 1), new Among("erets", 28, 1),
+              new Among("et", -1, 1), new Among("eret", 30, 1)
+            ],
+            a_1 = [
+              new Among("gd", -1, -1), new Among("dt", -1, -1),
+              new Among("gt", -1, -1), new Among("kt", -1, -1)
+            ],
+            a_2 = [
+              new Among("ig", -1, 1), new Among("lig", 0, 1),
+              new Among("elig", 1, 1), new Among("els", -1, 1),
+              new Among("l\u00F8st", -1, 2)
+            ],
+            g_v = [17, 65, 16, 1, 0, 0, 0, 0,
+              0, 0, 0, 0, 0, 0, 0, 0, 48, 0, 128
+            ],
+            g_s_ending = [239, 254, 42, 3,
+              0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 16
+            ],
+            I_x, I_p1, S_ch, sbp = new SnowballProgram();
+          this.setCurrent = function(word) {
+            sbp.setCurrent(word);
+          };
+          this.getCurrent = function() {
+            return sbp.getCurrent();
+          };
+
+          function r_mark_regions() {
+            var v_1, c = sbp.cursor + 3;
+            I_p1 = sbp.limit;
+            if (0 <= c && c <= sbp.limit) {
+              I_x = c;
+              while (true) {
+                v_1 = sbp.cursor;
+                if (sbp.in_grouping(g_v, 97, 248)) {
+                  sbp.cursor = v_1;
+                  break;
+                }
+                sbp.cursor = v_1;
+                if (v_1 >= sbp.limit)
+                  return;
+                sbp.cursor++;
+              }
+              while (!sbp.out_grouping(g_v, 97, 248)) {
+                if (sbp.cursor >= sbp.limit)
+                  return;
+                sbp.cursor++;
+              }
+              I_p1 = sbp.cursor;
+              if (I_p1 < I_x)
+                I_p1 = I_x;
+            }
+          }
+
+          function r_main_suffix() {
+            var among_var, v_1;
+            if (sbp.cursor >= I_p1) {
+              v_1 = sbp.limit_backward;
+              sbp.limit_backward = I_p1;
+              sbp.ket = sbp.cursor;
+              among_var = sbp.find_among_b(a_0, 32);
+              sbp.limit_backward = v_1;
+              if (among_var) {
+                sbp.bra = sbp.cursor;
+                switch (among_var) {
+                  case 1:
+                    sbp.slice_del();
+                    break;
+                  case 2:
+                    if (sbp.in_grouping_b(g_s_ending, 97, 229))
+                      sbp.slice_del();
+                    break;
+                }
+              }
+            }
+          }
+
+          function r_consonant_pair() {
+            var v_1 = sbp.limit - sbp.cursor,
+              v_2;
+            if (sbp.cursor >= I_p1) {
+              v_2 = sbp.limit_backward;
+              sbp.limit_backward = I_p1;
+              sbp.ket = sbp.cursor;
+              if (sbp.find_among_b(a_1, 4)) {
+                sbp.bra = sbp.cursor;
+                sbp.limit_backward = v_2;
+                sbp.cursor = sbp.limit - v_1;
+                if (sbp.cursor > sbp.limit_backward) {
+                  sbp.cursor--;
+                  sbp.bra = sbp.cursor;
+                  sbp.slice_del();
+                }
+              } else
+                sbp.limit_backward = v_2;
+            }
+          }
+
+          function r_other_suffix() {
+            var among_var, v_1 = sbp.limit - sbp.cursor,
+              v_2, v_3;
+            sbp.ket = sbp.cursor;
+            if (sbp.eq_s_b(2, "st")) {
+              sbp.bra = sbp.cursor;
+              if (sbp.eq_s_b(2, "ig"))
+                sbp.slice_del();
+            }
+            sbp.cursor = sbp.limit - v_1;
+            if (sbp.cursor >= I_p1) {
+              v_2 = sbp.limit_backward;
+              sbp.limit_backward = I_p1;
+              sbp.ket = sbp.cursor;
+              among_var = sbp.find_among_b(a_2, 5);
+              sbp.limit_backward = v_2;
+              if (among_var) {
+                sbp.bra = sbp.cursor;
+                switch (among_var) {
+                  case 1:
+                    sbp.slice_del();
+                    v_3 = sbp.limit - sbp.cursor;
+                    r_consonant_pair();
+                    sbp.cursor = sbp.limit - v_3;
+                    break;
+                  case 2:
+                    sbp.slice_from("l\u00F8s");
+                    break;
+                }
+              }
+            }
+          }
+
+          function r_undouble() {
+            var v_1;
+            if (sbp.cursor >= I_p1) {
+              v_1 = sbp.limit_backward;
+              sbp.limit_backward = I_p1;
+              sbp.ket = sbp.cursor;
+              if (sbp.out_grouping_b(g_v, 97, 248)) {
+                sbp.bra = sbp.cursor;
+                S_ch = sbp.slice_to(S_ch);
+                sbp.limit_backward = v_1;
+                if (sbp.eq_v_b(S_ch))
+                  sbp.slice_del();
+              } else
+                sbp.limit_backward = v_1;
+            }
+          }
+          this.stem = function() {
+            var v_1 = sbp.cursor;
+            r_mark_regions();
+            sbp.limit_backward = v_1;
+            sbp.cursor = sbp.limit;
+            r_main_suffix();
+            sbp.cursor = sbp.limit;
+            r_consonant_pair();
+            sbp.cursor = sbp.limit;
+            r_other_suffix();
+            sbp.cursor = sbp.limit;
+            r_undouble();
+            return true;
+          }
+        };
+
+      /* and return a function that stems a word for the current locale */
+      return function(word) {
+        st.setCurrent(word);
+        st.stem();
+        return st.getCurrent();
+      }
+    })();
+
+    lunr.Pipeline.registerFunction(lunr.da.stemmer, 'stemmer-da');
+
+    /* stop word filter function */
+    lunr.da.stopWordFilter = function(token) {
+      if (lunr.da.stopWordFilter.stopWords.indexOf(token) === -1) {
+        return token;
+      }
+    };
+
+    lunr.da.stopWordFilter.stopWords = new lunr.SortedSet();
+    lunr.da.stopWordFilter.stopWords.length = 95;
+
+    // The space at the beginning is crucial: It marks the empty string
+    // as a stop word. lunr.js crashes during search when documents
+    // processed by the pipeline still contain the empty string.
+    lunr.da.stopWordFilter.stopWords.elements = ' ad af alle alt anden at blev blive bliver da de dem den denne der deres det dette dig din disse dog du efter eller en end er et for fra ham han hans har havde have hende hendes her hos hun hvad hvis hvor i ikke ind jeg jer jo kunne man mange med meget men mig min mine mit mod ned noget nogle nu når og også om op os over på selv sig sin sine sit skal skulle som sådan thi til ud under var vi vil ville vor være været'.split(' ');
+
+    lunr.Pipeline.registerFunction(lunr.da.stopWordFilter, 'stopWordFilter-da');
+  };
+}))