From 139314c2133fb74a9e94ffbdc92f3551de279bea Mon Sep 17 00:00:00 2001
From: Sam <sam.saffron@gmail.com>
Date: Mon, 10 Aug 2015 17:41:14 +1000
Subject: [PATCH] FIX: search for 1.5 etc was broken FEATURE: search by exact
 phrase

Use PG tokenizer to split up search phrase instead of relying on Ruby code
This produces much more accurate tokenization
---
 lib/search.rb                  | 22 ++++++++++++++--------
 spec/components/search_spec.rb |  8 ++++++++
 2 files changed, 22 insertions(+), 8 deletions(-)

diff --git a/lib/search.rb b/lib/search.rb
index 0a07dcf9f79..2dd6b100392 100644
--- a/lib/search.rb
+++ b/lib/search.rb
@@ -385,6 +385,10 @@ class Search
           posts = posts.where("posts.raw  || ' ' || u.username || ' ' || u.name ilike ?", "%#{@term}%")
         else
           posts = posts.where("post_search_data.search_data @@ #{ts_query}")
+          exact_terms = @term.scan(/"([^"]+)"/).flatten
+          exact_terms.each do |exact|
+            posts = posts.where("posts.raw ilike ?", "%#{exact}%")
+          end
         end
       end
 
@@ -457,20 +461,22 @@ class Search
     end
 
     def self.ts_query(term, locale = nil, joiner = "&")
+
+
+      data = Post.exec_sql("SELECT to_tsvector(:locale, :term)",
+                            locale: locale || long_locale,
+                            term: term
+                          ).values[0][0]
+
       locale = Post.sanitize(locale) if locale
-      all_terms = term.gsub(/[\p{P}\p{S}]+/, ' ').squish.split
+      all_terms = data.scan(/'([^']+)'\:\d+/).flatten
       query = Post.sanitize(all_terms.map {|t| "#{PG::Connection.escape_string(t)}:*"}.join(" #{joiner} "))
       "TO_TSQUERY(#{locale || query_locale}, #{query})"
     end
 
     def ts_query(locale=nil)
-      if !locale
-        @ts_query ||= begin
-          Search.ts_query(@term, locale)
-        end
-      else
-        Search.ts_query(@term, locale)
-      end
+      @ts_query_cache ||= {}
+      @ts_query_cache[(locale || query_locale) + " " + @term] ||= Search.ts_query(@term, locale)
     end
 
     def aggregate_search(opts = {})
diff --git a/spec/components/search_spec.rb b/spec/components/search_spec.rb
index 1ded92fbc77..f2b15a86d40 100644
--- a/spec/components/search_spec.rb
+++ b/spec/components/search_spec.rb
@@ -389,6 +389,14 @@ describe Search do
 
     end
 
+    it 'can search numbers correctly, and match exact phrases' do
+      topic = Fabricate(:topic, created_at: 3.months.ago)
+      Fabricate(:post, raw: '3.0 eta is in 2 days horrah', topic: topic)
+
+      expect(Search.execute('3.0 eta').posts.length).to eq(1)
+      expect(Search.execute('"3.0, eta is"').posts.length).to eq(0)
+    end
+
     it 'can find by status' do
       post = Fabricate(:post, raw: 'hi this is a test 123 123')
       topic = post.topic