twitter-text 1.0.4 → 1.1.1

data/README.rdoc

@@ -57,10 +57,15 @@ Special care has been taken to be sure that auto-linking and extraction work
 in Tweets of all languages. This means that languages without spaces between
 words should work equally well.
 
+=== Hit Highlighting
+
+Use to provide emphasis around the "hits" returned from the Search API, built
+to work against text that has been auto-linked already.
+
 === Conformance
 
 To run the Conformance suite, you'll need to add that project as a git submodule.  From the root twitter-text-rb directory, run:
 
-git submodule add git@github.com:mzsanford/twitter-text-conformance.git test/twitter-text-conformance/
-git submodule init
-git submodule update
+  git submodule add git@github.com:mzsanford/twitter-text-conformance.git test/twitter-text-conformance/
+  git submodule init
+  git submodule update

data/Rakefile

@@ -9,9 +9,9 @@ require 'digest'
 
 spec = Gem::Specification.new do |s|
   s.name = "twitter-text"
-  s.version = "1.0.4"
-  s.authors = ["Matt Sanford", "Patrick Ewing"]
-  s.email = ["matt@twitter.com", "patrick.henry.ewing@gmail.com"]
+  s.version = "1.1.1"
+  s.authors = ["Matt Sanford", "Patrick Ewing", "Ben Cherry", "Britt Selvitelle"]
+  s.email = ["matt@twitter.com", "patrick.henry.ewing@gmail.com", "bcherry@gmail.com", "bs@brittspace.com"]
   s.homepage = "http://twitter.com"
   s.description = s.summary = "A gem that provides text handling for Twitter"
 

data/TODO

@@ -1,3 +1,4 @@
 TODO:
 
- * More tests
+ * @mentions preceded by a dash should work. "I am great -@greatguy"
+ * HitHighlighter tests should be moved to the conformance suite

data/lib/autolink.rb

@@ -38,7 +38,7 @@ module Twitter
         options),
       options)
     end
-
+    
     # Add <tt><a></a></tt> tags around the usernames and lists in the provided <tt>text</tt>. The
     # <tt><a></tt> tags can be controlled with the following entries in the <tt>options</tt>
     # hash:

data/lib/hithighlighter.rb

@@ -0,0 +1,81 @@
+
+module Twitter
+  # Module for doing "hit highlighting" on tweets that have been auto-linked already.
+  # Useful with the results returned from the Search API.
+  module HitHighlighter
+    # Default Tag used for hit highlighting
+    DEFAULT_HIGHLIGHT_TAG = "em"
+
+    # Add <tt><em></em></tt> tags around the <tt>hits</tt> provided in the <tt>text</tt>. The
+    # <tt>hits</tt> should be an array of (start, end) index pairs, relative to the original
+    # text, before auto-linking (but the <tt>text</tt> may already be auto-linked if desired)
+    #
+    # The <tt><em></em></tt> tags can be overridden using the <tt>:tag</tt> option. For example:
+    #
+    #  irb> hit_highlight("test hit here", [[5, 8]], :tag => 'strong')
+    #  => "test <strong>hit</strong> here"
+    def hit_highlight(text, hits = [], options = {})
+      if hits.empty?
+        return text
+      end
+
+      tag_name = options[:tag] || DEFAULT_HIGHLIGHT_TAG
+      tags = ["<" + tag_name + ">", "</" + tag_name + ">"]
+
+      chunks = text.split("<").map do |item|
+        item.blank? ? item : item.split(">")
+      end.flatten
+
+      result = ""
+      chunk_index, chunk = 0, chunks[0]
+      chunk_chars = chunk.respond_to?("mb_chars") ? chunk.mb_chars : chunk.respond_to?("chars") && chunk.chars.respond_to?("[]") ? chunk.chars : chunk
+      prev_chunks_len = 0
+      chunk_cursor = 0
+      start_in_chunk = false
+      for hit, index in hits.flatten.each_with_index do
+        tag = tags[index % 2]
+
+        placed = false
+        until chunk.nil? || hit < prev_chunks_len + chunk.length do
+          result << chunk_chars[chunk_cursor..-1]
+          if start_in_chunk && hit == prev_chunks_len + chunk_chars.length
+            result << tag
+            placed = true
+          end
+
+          # correctly handle highlights that end on the final character.
+          if tag_text = chunks[chunk_index+1]
+            result << "<#{tag_text}>"
+          end
+
+          prev_chunks_len += chunk_chars.length
+          chunk_cursor = 0
+          chunk_index += 2
+          chunk = chunks[chunk_index]
+          chunk_chars = chunk.respond_to?("mb_chars") ? chunk.mb_chars : chunk.respond_to?("chars") && chunk.chars.respond_to?("[]") ? chunk.chars : chunk
+          start_in_chunk = false
+        end
+
+        if !placed && !chunk.nil?
+          hit_spot = hit - prev_chunks_len
+          result << chunk_chars[chunk_cursor...hit_spot].to_s + tag
+          chunk_cursor = hit_spot
+          if index % 2 == 0
+            start_in_chunk = true
+          end
+        end
+      end
+
+      if chunk
+        if chunk_cursor < chunk_chars.length
+          result << chunk_chars[chunk_cursor..-1]
+        end
+        (chunk_index+1).upto(chunks.length-1).each do |index|
+          result << (index.even? ? chunks[index] : "<#{chunks[index]}>")
+        end
+      end
+
+      result
+    end
+  end
+end

data/lib/regex.rb

@@ -44,7 +44,7 @@ module Twitter
 
     # URL related hash regex collection
     REGEXEN[:valid_preceding_chars] = /(?:[^\/"':!=]|^|\:)/
-    REGEXEN[:valid_domain] = /(?:[^[:punct:]\s][\.-][^[:punct:]\s]|[^[:punct:]\s])+\.[a-z]{2,}(?::[0-9]+)?/i
+    REGEXEN[:valid_domain] = /(?:[^[:punct:]\s][\.-](?=[^[:punct:]\s])|[^[:punct:]\s]){1,}\.[a-z]{2,}(?::[0-9]+)?/i
     REGEXEN[:valid_url_path_chars] = /[\.\,]?[a-z0-9!\*'\(\);:=\+\$\/%#\[\]\-_,~@]/i
     # Valid end-of-path chracters (so /foo. does not gobble the period).
     #   1. Allow ) for Wikipedia URLs.
@@ -67,7 +67,7 @@ module Twitter
     REGEXEN.each_pair{|k,v| v.freeze }
 
     # Return the regular expression for a given <tt>key</tt>. If the <tt>key</tt>
-    # is not a known symbol a <tt>nil</tt> will be returned. 
+    # is not a known symbol a <tt>nil</tt> will be returned.
     def self.[](key)
       REGEXEN[key]
     end

data/lib/twitter-text.rb

@@ -10,3 +10,4 @@ require File.join(File.dirname(__FILE__), 'autolink')
 require File.join(File.dirname(__FILE__), 'extractor')
 require File.join(File.dirname(__FILE__), 'unicode')
 require File.join(File.dirname(__FILE__), 'validation')
+require File.join(File.dirname(__FILE__), 'hithighlighter')

data/spec/autolinking_spec.rb

@@ -430,6 +430,7 @@ describe Twitter::Autolink do
       end
 
     end
+    
   end
 
 end

data/spec/hithighlighter_spec.rb

@@ -0,0 +1,83 @@
+require File.dirname(__FILE__) + '/spec_helper'
+
+class TestHitHighlighter
+  include Twitter::HitHighlighter
+end
+
+describe Twitter::HitHighlighter do
+  describe "highlight" do
+    before do
+      @highlighter = TestHitHighlighter.new
+    end
+
+    context "with options" do
+      before do
+        @original = "Testing this hit highliter"
+        @hits = [[13,16]]
+      end
+
+      it "should default to <em> tags" do
+        @highlighter.hit_highlight(@original, @hits).should == "Testing this <em>hit</em> highliter"
+      end
+
+      it "should allow tag override" do
+        @highlighter.hit_highlight(@original, @hits, :tag => 'b').should == "Testing this <b>hit</b> highliter"
+      end
+    end
+
+    context "without links" do
+      before do
+        @original = "Hey! this is a test tweet"
+      end
+
+      it "should return original when no hits are provided" do
+        @highlighter.hit_highlight(@original).should == @original
+      end
+
+      it "should highlight one hit" do
+        @highlighter.hit_highlight(@original, hits = [[5, 9]]).should == "Hey! <em>this</em> is a test tweet"
+      end
+
+      it "should highlight two hits" do
+        @highlighter.hit_highlight(@original, hits = [[5, 9], [15, 19]]).should == "Hey! <em>this</em> is a <em>test</em> tweet"
+      end
+
+      it "should correctly highlight first-word hits" do
+        @highlighter.hit_highlight(@original, hits = [[0, 3]]).should == "<em>Hey</em>! this is a test tweet"
+      end
+
+      it "should correctly highlight last-word hits" do
+        @highlighter.hit_highlight(@original, hits = [[20, 25]]).should == "Hey! this is a test <em>tweet</em>"
+      end
+    end
+
+    context "with links" do
+      it "should highlight with a single link" do
+        @highlighter.hit_highlight("@<a>bcherry</a> this was a test tweet", [[9, 13]]).should == "@<a>bcherry</a> <em>this</em> was a test tweet"
+      end
+
+      it "should highlight with link at the end" do
+        @highlighter.hit_highlight("test test <a>test</a>", [[5, 9]]).should == "test <em>test</em> <a>test</a>"
+      end
+
+      it "should highlight with a link at the beginning" do
+        @highlighter.hit_highlight("<a>test</a> test test", [[5, 9]]).should == "<a>test</a> <em>test</em> test"
+      end
+
+      it "should highlight an entire link" do
+        @highlighter.hit_highlight("test <a>test</a> test", [[5, 9]]).should == "test <a><em>test</em></a> test"
+      end
+
+      it "should highlight within a link" do
+        @highlighter.hit_highlight("test <a>test</a> test", [[6, 8]]).should == "test <a>t<em>es</em>t</a> test"
+      end
+
+      it "should highlight around a link" do
+        @highlighter.hit_highlight("test <a>test</a> test", [[3, 11]]).should == "tes<em>t <a>test</a> t</em>est"
+      end
+
+    end
+
+  end
+
+end

data/spec/regex_spec.rb

@@ -16,7 +16,7 @@ describe "Twitter::Regex regular expressions" do
 
   describe "invalid URLS" do
     it "does not link urls with invalid characters" do
-      TestUrls::INVALID.each {|url| url.should_not have_autolinked_url(url)}
+      TestUrls::INVALID.each {|url| url.should_not match_autolink_expression}
     end
   end
 

data/spec/test_urls.rb

@@ -7,7 +7,7 @@ module TestUrls
     "http://twitter.com/#search?q=iphone%20-filter%3Alinks",
     "http://www.boingboing.net/2007/02/14/katamari_damacy_phon.html",
     "http://somehost.com:3000",
-    "http://x.com/~matthew+%-x",
+    "http://xo.com/~matthew+%-x",
     "http://en.wikipedia.org/wiki/Primer_(film)",
     "http://www.ams.org/bookstore-getitem/item=mbk-59",
     "http://chilp.it/?77e8fd",
@@ -16,14 +16,15 @@ module TestUrls
     "http://tell.me/why",
     "http://longtlds.info",
     "http://✪df.ws/ejp",
-    "http://日本.com"
+    "http://日本.com",
+    "http://search.twitter.com/search?q=avro&lang=en",
+    "http://mrs.domain-dash.biz",
+    "http://x.com/has/one/char/domain",
   ]
 
   INVALID = [
     "http://no-tld",
     "http://tld-too-short.x",
-    "http://x.com/,,,/.../@@@/;;;/:::/---/%%%x",
-    "http://domain-dash.com",
     "http://-doman_dash.com"
   ]
 

metadata

@@ -1,32 +1,43 @@
 --- !ruby/object:Gem::Specification 
 name: twitter-text
 version: !ruby/object:Gem::Version 
-  version: 1.0.4
+  prerelease: false
+  segments: 
+  - 1
+  - 1
+  - 1
+  version: 1.1.1
 platform: ruby
 authors: 
 - Matt Sanford
 - Patrick Ewing
+- Ben Cherry
+- Britt Selvitelle
 autorequire: ""
 bindir: bin
 cert_chain: []
 
-date: 2010-03-11 00:00:00 -08:00
+date: 2010-04-28 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: actionpack
-  type: :runtime
-  version_requirement: 
-  version_requirements: !ruby/object:Gem::Requirement 
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
+        segments: 
+        - 0
         version: "0"
-    version: 
+  type: :runtime
+  version_requirements: *id001
 description: A gem that provides text handling for Twitter
 email: 
 - matt@twitter.com
 - patrick.henry.ewing@gmail.com
+- bcherry@gmail.com
+- bs@brittspace.com
 executables: []
 
 extensions: []
@@ -40,12 +51,14 @@ files:
 - TODO
 - lib/autolink.rb
 - lib/extractor.rb
+- lib/hithighlighter.rb
 - lib/regex.rb
 - lib/twitter-text.rb
 - lib/unicode.rb
 - lib/validation.rb
 - spec/autolinking_spec.rb
 - spec/extractor_spec.rb
+- spec/hithighlighter_spec.rb
 - spec/regex_spec.rb
 - spec/spec_helper.rb
 - spec/test_urls.rb
@@ -64,18 +77,20 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      segments: 
+      - 0
       version: "0"
-  version: 
 required_rubygems_version: !ruby/object:Gem::Requirement 
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      segments: 
+      - 0
       version: "0"
-  version: 
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.3.5
+rubygems_version: 1.3.6
 signing_key: 
 specification_version: 3
 summary: Twitter text handling library