using pageDown Markdown library, fixing autolinking. created...

using pageDown Markdown library, fixing autolinking.  created app.helpers.textFormatter, which takes care of text formatting; functions can be called individually throughout the app

config/assets.yml

@@ -9,10 +9,9 @@ javascripts:
   main:
     - public/javascripts/vendor/underscore.js
     - public/javascripts/vendor/backbone.js
-
-    - public/javascripts/vendor/markdown.js
-
+    - public/javascripts/vendor/markdown/*
     - public/javascripts/app/app.js
+    - public/javascripts/app/helpers/*
     - public/javascripts/app/router.js
     - public/javascripts/app/views.js
     - public/javascripts/app/models/post.js

public/javascripts/app/app.js

 var app = {
   collections: {},
   models: {},
+  helpers: {},
   views: {},
 
   user: function(user) {

public/javascripts/app/helpers/text_formatter.js

+(function(){
+  var textFormatter = function textFormatter(model) {
+    var text = model.get("text");
+    var mentions = model.get("mentioned_people");
+
+    return textFormatter.mentionify(
+      textFormatter.hashtagify(
+        textFormatter.markdownify(text)
+        ), mentions
+      )
+  };
+
+  textFormatter.markdownify = function markdownify(text){
+    var converter = Markdown.getSanitizingConverter();
+    return converter.makeHtml(text)
+  };
+
+  textFormatter.hashtagify = function hashtagify(text){
+    var utf8WordCharcters =/(\s|^|>)#([\u0080-\uFFFF|\w|-]+|<3)/g
+    return text.replace(utf8WordCharcters, function(hashtag, preceeder, tagText) {
+      return preceeder + "<a href='/tags/" + tagText + "' class='tag'>#" + tagText + "</a>"
+    })
+  };
+
+  textFormatter.mentionify = function mentionify(text, mentions) {
+    var mentionRegex = /@\{([^;]+); ([^\}]+)\}/g
+    return text.replace(mentionRegex, function(mentionText, fullName, diasporaId) {
+      var personId = _.find(mentions, function(person){
+        return person.diaspora_id == diasporaId
+      }).id
+
+      return "<a href='/people/" + personId + "' class='mention'>" + fullName + "</a>"
+    })
+  }
+
+  app.helpers.textFormatter = textFormatter;
+})();
+

public/javascripts/app/views/content_view.js

@@ -2,54 +2,10 @@ app.views.Content = app.views.StreamObject.extend({
   presenter : function(){
     var model = this.model
     return _.extend(this.defaultPresenter(), {
-      text : metafyText(model.get("text")),
+      text : app.helpers.textFormatter(model),
       o_embed_html : embedHTML(model)
     })
 
-    function metafyText(text) {
-      //we want it to return at least a <p> from markdown
-      text = text || ""
-      return urlify(
-        mentionify(
-          hashtagify(
-            markdownify(text)
-          )
-        )
-      )
-    }
-
-    function markdownify(text){
-      //markdown returns falsy when it performs no substitutions, apparently...
-      return markdown.toHTML(text) || text
-    }
-
-    function hashtagify(text){
-      var utf8WordCharcters =/(\s|^|>)#([\u0080-\uFFFF|\w|-]+|&lt;3)/g
-      return text.replace(utf8WordCharcters, function(hashtag, preceeder, tagText) {
-        return preceeder + "<a href='/tags/" + tagText + "' class='tag'>#" + tagText + "</a>"
-      })
-    }
-
-    function mentionify(text) {
-      var mentionRegex = /@\{([^;]+); ([^\}]+)\}/g
-      return text.replace(mentionRegex, function(mentionText, fullName, diasporaId) {
-        var personId = _.find(model.get("mentioned_people"), function(person){
-          return person.diaspora_id == diasporaId
-        }).id
-
-        return "<a href='/people/" + personId + "' class='mention'>" + fullName + "</a>"
-      })
-    }
-
-    function urlify(text) {
-      var urlRegex = /(=\s?'|=\s?")?[-a-zA-Z0-9@:%_\+.~#?&//=]{2,256}\.[a-z]{2,4}\b(\/[-a-zA-Z0-9@:%_\+.~#?(#!)&//=;]*)?/gi
-      return text.replace(urlRegex, function(url, preceeder, bang) {
-        if(preceeder) return url
-        var protocol = (url.search(/:\/\//) == -1 ? "http://" : "")
-        return "<a href='" + protocol + url + "' target=_blank>" + url + "</a>"
-      })
-    }
-
     function embedHTML(model){
       if(!model.get("o_embed_cache")) { return ""; }
       return model.get("o_embed_cache").data.html
@@ -57,7 +13,6 @@ app.views.Content = app.views.StreamObject.extend({
   }
 })
 
-
 app.views.StatusMessage = app.views.Content.extend({
   template_name : "#status-message-template"
 });

public/javascripts/vendor/markdown/Markdown.Sanitizer.js

+(function () {
+    var output, Converter;
+    if (typeof exports === "object" && typeof require === "function") { // we're in a CommonJS (e.g. Node.js) module
+        output = exports;
+        Converter = require("./Markdown.Converter").Converter;
+    } else {
+        output = window.Markdown;
+        Converter = output.Converter;
+    }
+        
+    output.getSanitizingConverter = function () {
+        var converter = new Converter();
+        converter.hooks.chain("postConversion", sanitizeHtml);
+        converter.hooks.chain("postConversion", balanceTags);
+        return converter;
+    }
+
+    function sanitizeHtml(html) {
+        return html.replace(/<[^>]*>?/gi, sanitizeTag);
+    }
+
+    // (tags that can be opened/closed) | (tags that stand alone)
+    var basic_tag_whitelist = /^(<\/?(b|blockquote|code|del|dd|dl|dt|em|h1|h2|h3|i|kbd|li|ol|p|pre|s|sup|sub|strong|strike|ul)>|<(br|hr)\s?\/?>)$/i;
+    // <a href="url..." optional title>|</a>
+    var a_white = /^(<a\shref="((https?|ftp):\/\/|\/)[-A-Za-z0-9+&@#\/%?=~_|!:,.;\(\)]+"(\stitle="[^"<>]+")?\s?>|<\/a>)$/i;
+
+    // <img src="url..." optional width  optional height  optional alt  optional title
+    var img_white = /^(<img\ssrc="(https?:\/\/|\/)[-A-Za-z0-9+&@#\/%?=~_|!:,.;\(\)]+"(\swidth="\d{1,3}")?(\sheight="\d{1,3}")?(\salt="[^"<>]*")?(\stitle="[^"<>]*")?\s?\/?>)$/i;
+
+    function sanitizeTag(tag) {
+        if (tag.match(basic_tag_whitelist) || tag.match(a_white) || tag.match(img_white))
+            return tag;
+        else
+            return "";
+    }
+
+    /// <summary>
+    /// attempt to balance HTML tags in the html string
+    /// by removing any unmatched opening or closing tags
+    /// IMPORTANT: we *assume* HTML has *already* been 
+    /// sanitized and is safe/sane before balancing!
+    /// 
+    /// adapted from CODESNIPPET: A8591DBA-D1D3-11DE-947C-BA5556D89593
+    /// </summary>
+    function balanceTags(html) {
+
+        if (html == "")
+            return "";
+
+        var re = /<\/?\w+[^>]*(\s|$|>)/g;
+        // convert everything to lower case; this makes
+        // our case insensitive comparisons easier
+        var tags = html.toLowerCase().match(re);
+
+        // no HTML tags present? nothing to do; exit now
+        var tagcount = (tags || []).length;
+        if (tagcount == 0)
+            return html;
+
+        var tagname, tag;
+        var ignoredtags = "<p><img><br><li><hr>";
+        var match;
+        var tagpaired = [];
+        var tagremove = [];
+        var needsRemoval = false;
+
+        // loop through matched tags in forward order
+        for (var ctag = 0; ctag < tagcount; ctag++) {
+            tagname = tags[ctag].replace(/<\/?(\w+).*/, "$1");
+            // skip any already paired tags
+            // and skip tags in our ignore list; assume they're self-closed
+            if (tagpaired[ctag] || ignoredtags.search("<" + tagname + ">") > -1)
+                continue;
+
+            tag = tags[ctag];
+            match = -1;
+
+            if (!/^<\//.test(tag)) {
+                // this is an opening tag
+                // search forwards (next tags), look for closing tags
+                for (var ntag = ctag + 1; ntag < tagcount; ntag++) {
+                    if (!tagpaired[ntag] && tags[ntag] == "</" + tagname + ">") {
+                        match = ntag;
+                        break;
+                    }
+                }
+            }
+
+            if (match == -1)
+                needsRemoval = tagremove[ctag] = true; // mark for removal
+            else
+                tagpaired[match] = true; // mark paired
+        }
+
+        if (!needsRemoval)
+            return html;
+
+        // delete all orphaned tags from the string
+
+        var ctag = 0;
+        html = html.replace(re, function (match) {
+            var res = tagremove[ctag] ? "" : match;
+            ctag++;
+            return res;
+        });
+        return html;
+    }
+})();

spec/javascripts/app/helpers/text_formatter_spec.js

+describe("app.helpers.textFormatter", function(){
+
+  beforeEach(function(){
+    this.statusMessage = factory.post();
+    this.formatter = app.helpers.textFormatter;
+  })
+
+  describe("main", function(){
+    it("calls mentionify, hashtagify, and markdownify", function(){
+      spyOn(app.helpers.textFormatter, "mentionify")
+      spyOn(app.helpers.textFormatter, "hashtagify")
+      spyOn(app.helpers.textFormatter, "markdownify")
+
+      app.helpers.textFormatter(this.statusMessage)
+      expect(app.helpers.textFormatter.mentionify).toHaveBeenCalled()
+      expect(app.helpers.textFormatter.hashtagify).toHaveBeenCalled()
+      expect(app.helpers.textFormatter.markdownify).toHaveBeenCalled()
+    })
+
+    // A couple of complex (intergration) test cases here would be rad.
+  })
+
+  describe(".markdownify", function(){
+    // NOTE: for some strange reason, links separated by just a whitespace character
+    // will not be autolinked; thus we join our URLS here with (" and ").
+    // This test will fail if our join is just (" ") -- an edge case that should be addressed.
+
+    it("autolinks", function(){
+      var links = ["http://google.com",
+        "https://joindiaspora.com",
+        "http://www.yahooligans.com",
+        "http://obama.com",
+        "http://japan.co.jp"]
+
+      // The join that would make this particular test fail:
+      //
+      // var formattedText = this.formatter.markdownify(links.join(" "))
+
+      var formattedText = this.formatter.markdownify(links.join(" and "))
+      var wrapper = $("<div>").html(formattedText);
+
+      _.each(links, function(link) {
+        expect(wrapper.find("a[href='" + link + "']").text()).toContain(link)
+      })
+    })
+  })
+
+  describe(".hashtagify", function(){
+    context("changes hashtags to links", function(){
+      it("creates links to hashtags", function(){
+        var formattedText = this.formatter.hashtagify("I love #parties and #rockstars and #unicorns")
+        var wrapper = $("<div>").html(formattedText);
+
+        _.each(["parties", "rockstars", "unicorns"], function(tagName){
+          expect(wrapper.find("a[href='/tags/" + tagName + "']").text()).toContain(tagName)
+        })
+      })
+
+      it("requires hashtags to be preceeded with a space", function(){
+        var formattedText = this.formatter.hashtagify("I love the#parties")
+        expect(formattedText).not.toContain('/tags/parties')
+      })
+
+      // NOTE THIS DIVERGES FROM GRUBER'S ORIGINAL DIALECT OF MARKDOWN.
+      // We had to edit Markdown.Converter.js line 747
+      //
+      //    text = text.replace(/^(\#{1,6})[ \t]+(.+?)[ \t]*\#*\n+/gm,
+      //    [ \t]* changed to [ \t]+
+      //
+      it("doesn't create a header tag if the first word is a hashtag", function(){
+        var formattedText = this.formatter.hashtagify("#parties, I love")
+        var wrapper = $("<div>").html(formattedText);
+
+        expect(wrapper.find("h1").length).toBe(0)
+        expect(wrapper.find("a[href='/tags/parties']").text()).toContain("#parties")
+      })
+    })
+  })
+
+  describe(".mentionify", function(){
+    context("changes mention markup to links", function(){
+      beforeEach(function(){
+        this.alice = factory.author({
+          name : "Alice Smith",
+          diaspora_id : "alice@example.com",
+          id : "555"
+        })
+
+        this.bob = factory.author({
+          name : "Bob Grimm",
+          diaspora_id : "bob@example.com",
+          id : "666"
+        })
+
+        this.statusMessage.set({text: "hey there @{Alice Smith; alice@example.com} and @{Bob Grimm; bob@example.com}"})
+        this.statusMessage.set({mentioned_people : [this.alice, this.bob]})
+      })
+
+      it("matches mentions", function(){
+        var formattedText = this.formatter.mentionify(this.statusMessage.get("text"), this.statusMessage.get("mentioned_people"))
+        var wrapper = $("<div>").html(formattedText);
+
+        _.each([this.alice, this.bob], function(person) {
+          expect(wrapper.find("a[href='/people/" + person.id + "']").text()).toContain(person.name)
+        })
+      })
+    })
+  })
+})

spec/javascripts/app/views/post_view_spec.js

@@ -40,132 +40,6 @@ describe("app.views.Post", function(){
       expect(view.$(".post_initial_info").html()).not.toContain("0 Reshares")
     })
 
-    it("should markdownify the post's text", function(){
-      this.statusMessage.set({text: "I have three Belly Buttons"})
-      spyOn(window.markdown, "toHTML")
-      new app.views.Post({model : this.statusMessage}).render();
-      expect(window.markdown.toHTML).toHaveBeenCalledWith("I have three Belly Buttons")
-    })
-
-    context("changes hashtags to links", function(){
-      it("links to a hashtag to the tag page", function(){
-        this.statusMessage.set({text: "I love #parties"})
-        var view = new app.views.Post({model : this.statusMessage}).render();
-        expect(view.$("a:contains('#parties')").attr('href')).toBe('/tags/parties')
-      })
-
-      it("changes all hashtags", function(){
-        this.statusMessage.set({text: "I love #parties and #rockstars and #unicorns"})
-        var view = new app.views.Post({model : this.statusMessage}).render();
-        expect(view.$("a.tag").length).toBe(3)
-        expect(view.$("a:contains('#parties')")).toExist();
-        expect(view.$("a:contains('#rockstars')")).toExist();
-        expect(view.$("a:contains('#unicorns')")).toExist();
-      })
-
-      it("requires hashtags to be preceeded with a space", function(){
-        this.statusMessage.set({text: "I love the#parties"})
-        var view = new app.views.Post({model : this.statusMessage}).render();
-        expect(view.$(".tag").length).toBe(0)
-      })
-
-      // NOTE THIS DIVERGES FROM GRUBER'S ORIGINAL DIALECT OF MARKDOWN.
-      // We had to edit markdown.js line 291 - good people would have made a new dialect.
-      //
-      //    original : var m = block.match( /^(#{1,6})\s*(.*?)\s*#*\s*(?:\n|$)/ );
-      //    \s* changed to \s+
-      //
-      it("doesn't create a header tag if the first word is a hashtag", function(){
-        this.statusMessage.set({text: "#parties, I love"})
-        var view = new app.views.Post({model : this.statusMessage}).render();
-        expect(view.$("h1:contains(parties)")).not.toExist();
-        expect(view.$("a:contains('#parties')")).toExist();
-      })
-
-      it("works on reshares", function(){
-        this.statusMessage.set({text: "I love #parties"})
-        var reshare = new app.models.Reshare(factory.post({
-          text : this.statusMessage.get("text"),
-          root : this.statusMessage
-        }))
-
-        var view = new app.views.Post({model : reshare}).render();
-        expect(view.$("a:contains('#parties')").attr('href')).toBe('/tags/parties')
-      })
-    })
-
-    context("changes mention markup to links", function(){
-      beforeEach(function(){
-        this.alice = factory.author({
-          name : "Alice Smith",
-          diaspora_id : "alice@example.com",
-          id : "555"
-        })
-
-        this.bob = factory.author({
-          name : "Bob Grimm",
-          diaspora_id : "bob@example.com",
-          id : "666"
-        })
-
-        this.statusMessage.set({mentioned_people : [this.alice, this.bob]})
-        this.statusMessage.set({text: "hey there @{Alice Smith; alice@example.com} and @{Bob Grimm; bob@example.com}"})
-      })
-
-      it("links to the mentioned person's page", function(){
-        var view = new app.views.Post({model : this.statusMessage}).render();
-        expect(view.$("a:contains('Alice Smith')").attr('href')).toBe('/people/555')
-      })
-
-      it("matches all mentions", function(){
-        var view = new app.views.Post({model : this.statusMessage}).render();
-        expect(view.$("a.mention").length).toBe(2)
-      })
-
-      it("works on reshares", function(){
-        var reshare = new app.models.Reshare(factory.post({
-          text : this.statusMessage.get("text"),
-          mentioned_people :  this.statusMessage.get("mentioned_people"),
-          root : this.statusMessage
-        }))
-
-        var view = new app.views.Post({model : reshare}).render();
-        expect(view.$("a.mention").length).toBe(2)
-      })
-    })
-
-    context("generates urls from plaintext", function(){
-      it("works", function(){
-        links = ["http://google.com",
-                 "https://joindiaspora.com",
-                 "http://www.yahooligans.com",
-                 "http://obama.com",
-                 "http://japan.co.jp"]
-
-        this.statusMessage.set({text : links.join(" ")})
-        var view = new app.views.Post({model : this.statusMessage}).render();
-
-        _.each(links, function(link) {
-          expect(view.$("a[href='" + link + "']").text()).toContain(link)
-        })
-      })
-
-      it("works with urls that use #! syntax (i'm looking at you, twitter)')", function(){
-        link = "http://twitter.com/#!/hashbangs?gross=true"
-        this.statusMessage.set({text : link})
-        var view = new app.views.Post({model : this.statusMessage}).render();
-
-        expect(view.$("a[href='" + link + "']").text()).toContain(link)
-      })
-
-      it("doesn't create link tags for links that are already in <a/> or <img/> tags", function(){
-        link = "http://google.com"
-
-        this.statusMessage.set({text : "![cats](http://google.com/cats)"})
-        var view = new app.views.Content({model : this.statusMessage})
-        expect(view.presenter().text).toNotContain('</a>')
-      })
-    })
 
     context("embed_html", function(){
       it("provides oembed html from the model response", function(){

spec/javascripts/support/jasmine.yml

@@ -22,7 +22,7 @@ src_files:
   - public/javascripts/vendor/jquery.charcount.js
   - public/javascripts/vendor/timeago.js
   - public/javascripts/vendor/facebox.js
-  - public/javascripts/vendor/markdown.js
+  - public/javascripts/vendor/markdown/*
   - public/javascripts/jquery.infieldlabel-custom.js
   - public/javascripts/vendor/underscore.js
   - public/javascripts/vendor/backbone.js
@@ -36,6 +36,7 @@ src_files:
   - public/javascripts/widgets/*
 
   - public/javascripts/app/app.js
+  - public/javascripts/app/helpers/*
   - public/javascripts/app/router.js
   - public/javascripts/app/views.js
   - public/javascripts/app/models/post.js