jekyll_ranked_search 0.0.4 → 0.0.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: aa336a48a9eb94c3a0e7e395dee1a9c4a7938339cfa800e528db49b3c2876953
-  data.tar.gz: ced1e0e9cd3919a5a5fc28848c80523a81987698c11b0be0b4ea487f5f3ca3a7
+  metadata.gz: b7a50962fa0617b6acc80b702c06695a27f2ad056030e80e4c099230f958d39b
+  data.tar.gz: c26e47d1f49e7d50edd91c1263dde8a6188b0a2c491bc99d22223f98dc60b9a6
 SHA512:
-  metadata.gz: f87c796fa346ac49d87bb939ee85ed37cfbe4083954a2963e2fb04e95963b87e0b95d2b5b03356364a1ce84f59d366fc74477e65fd532f1808ed272b810cda22
-  data.tar.gz: 811a1bebdcbef011282022bfe7083d2284d0c690132aad7a79431fd01fd0548144540f8499c61e04458e6f93ab1b29f4a61a3d05c220e5fede490ce7c05d5861
+  metadata.gz: d6f1fa610a8afa056eb0451ea234294044beae405e4375a46e495c83cf6ab697a57f978e394082705f0692057ffb316e53a99d6d80099cfa6cebfdcd15a7df9e
+  data.tar.gz: 4557557cbbc06268b6fbe725de5d7cbe31ab72ffaffa3393b46ec9e9c4ba9eee68cff3ba39ce655d3e1c97f420eb152f86246b7eb88eb8333fab32d3ec029e66

data/lib/jekyll_ranked_search.rb

@@ -122,15 +122,26 @@ class TfidfConverter < Jekyll::Generator
     site.data['tfidf'] = tfidf.to_json
   end
 
+  # Tokenize document by removing special characters and splitting
+  # the document into tokens.
+  # @param [String] doc The document to tokenize
+  # @return [Array<String>] individual tokens/words
   def tokenize_words(doc)
     # Remove stopwords from document
     @stopwords ||= self.load_stopwords
 
+    # TODO: Remove Liquid tags via regex
+
     # Split document into tokens
     splitted_doc = doc.strip.downcase.split
 
     # Remove stopwords in place
-    splitted_doc.delete_if { |word| @stopwords.include?(word) }
+    splitted_doc.delete_if do |word|
+      if @stopwords.include?(word)
+        Jekyll.logger.debug "Removing stopword:", word
+      end
+      @stopwords.include?(word)
+    end
 
     # Remove special characters (only at beginning and end)
     splitted_doc.map! { |word| word.gsub(/[^a-z0-9_\/\-\s]/i, '') }
@@ -138,11 +149,12 @@ class TfidfConverter < Jekyll::Generator
     splitted_doc
   end
 
-  # Load stopwords from file
+  # Load english stopwords from file
+  # @return [Array<String>] the stopwords
   def load_stopwords
     Jekyll.logger.info "Loading stopwords"
     stopwords = Set.new
-    File.open(File.join(File.dirname(__FILE__), "stopwords.txt"), "r") do |f|
+    File.open(File.join(File.dirname(__FILE__), "stopwords/en.txt"), "r") do |f|
       f.each_line do |line|
         stopwords.add line.strip
       end

data/lib/search.js

@@ -54,7 +54,7 @@ class SearchBox extends LitElement {
       box-sizing: border-box;
       width: 100%;
       // margin: 0 auto;
-      padding: .4em;
+      padding: .4em 30px;
       border: 1px solid #ccc;
       font-size: 1.2em;
       border-radius: 4px;
@@ -62,6 +62,20 @@ class SearchBox extends LitElement {
       z-index: 11;
     }
 
+    label {
+      position: relative;
+    }
+
+    label::before {
+      content: "";
+      position: absolute;
+      left: 10px;
+      top: 0;
+      bottom: 0;
+      width: 20px;
+      background: url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' fill='none' viewBox='0 0 24 24' stroke-width='1.5' stroke='currentColor' class=''%3E%3Cpath stroke-linecap='round' stroke-linejoin='round' d='M21 21l-5.197-5.197m0 0A7.5 7.5 0 105.196 5.196a7.5 7.5 0 0010.607 10.607z' /%3E%3C/svg%3E") no-repeat center center;
+    }
+
     #results {
       position: absolute;
       width: 100%;
@@ -245,7 +259,9 @@ class SearchBox extends LitElement {
 
   render() {
     return html`<div>
+      <label>
       <input id="q" type="text" placeholder="${this._placeholder}" @keyup="${this.search}" @click=${this.openIfResults} @focus=${this.focus}>
+      </label>
       ${this._open ? html`
         <div id="results">
           ${this._results.map((result) => html`

data/lib/{stopwords.txt → stopwords/en.txt}

@@ -2,8 +2,6 @@
 'tis
 'twas
 've
-10
-39
 a
 a's
 able

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: jekyll_ranked_search
 version: !ruby/object:Gem::Version
-  version: 0.0.4
+  version: 0.0.6
 platform: ruby
 authors:
 - Friedrich Ewald
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-07-05 00:00:00.000000000 Z
+date: 2024-01-31 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: redcarpet
@@ -24,7 +24,10 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '3.6'
-description: Offline search plugin for Jekyll posts using TF-IDF
+description: |
+  A webcomponent based search box that provides search functionality for your Jekyll blog.
+
+  If you have any feedback or suggestions for improvement, please open an issue on Github.
 email: freddiemailster@gmail.com
 executables: []
 extensions: []
@@ -33,7 +36,7 @@ files:
 - lib/jekyll_ranked_search.rb
 - lib/search.js
 - lib/search.json
-- lib/stopwords.txt
+- lib/stopwords/en.txt
 homepage: https://github.com/f-ewald/jekyll_ranked_search
 licenses:
 - MIT
@@ -53,7 +56,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.13
+rubygems_version: 3.4.15
 signing_key:
 specification_version: 4
 summary: TF-IDF offline search for Jekyll posts