Merge branch 'fuzzy-anchoring-3' into develop

Conflicts:
	h/assets.py
	h/js/inject/host.coffee
	h/js/services.coffee

h/js/inject/host.coffee

@@ -34,20 +34,24 @@ class Annotator.Host extends Annotator
       if not @plugins[name]
         this.addPlugin(name, opts)
 
+    # Scan the document text with the DOM Text libraries
+    this.scanDocument "Annotator initialized"
+
   _setupXDM: ->
     # Set up the bridge plugin, which bridges the main annotation methods
     # between the host page and the panel widget.
+    whitelist = ['diffHTML', 'quote', 'ranges', 'target']
     this.addPlugin 'Bridge',
       origin: '*'
       window: @frame[0].contentWindow
       formatter: (annotation) =>
         formatted = {}
-        for k, v of annotation when k in ['quote', 'ranges']
+        for k, v of annotation when k in whitelist
           formatted[k] = v
         formatted
       parser: (annotation) =>
         parsed = {}
-        for k, v of annotation when k in ['quote', 'ranges']
+        for k, v of annotation when k in whitelist
           parsed[k] = v
         parsed
 
@@ -104,16 +108,7 @@ class Annotator.Host extends Annotator
               $(this).removeClass('annotator-hl-active')
         )
 
-        .bind('getHref', =>
-          uri = decodeURIComponent document.location.href
-          if document.location.hash
-            uri = uri.slice 0, (-1 * location.hash.length)
-          $('meta[property^="og:url"]').each ->
-            uri = decodeURIComponent this.content
-          $('link[rel^="canonical"]').each ->
-            uri = decodeURIComponent this.href
-          return uri
-        )
+        .bind('getHref', => this.getHref())
 
         .bind('getMaxBottom', =>
           sel = '*' + (":not(.annotator-#{x})" for x in [
@@ -140,12 +135,24 @@ class Annotator.Host extends Annotator
           $('html, body').stop().animate {scrollTop: y}, 600
         )
 
+  scanDocument: (reason = "something happened") =>
+    try
+      console.log "Analyzing host frame, because " + reason + "..."
+      r = @domMatcher.scan()
+      scanTime = r.time
+      console.log "Traversal+scan took " + scanTime + " ms."
+    catch e
+      console.log e.message
+      console.log e.stack
+
   _setupWrapper: ->
     @wrapper = @element
     .on 'mouseup', =>
       if not @ignoreMouseup
         setTimeout =>
           unless @selectedRanges?.length then @panel?.notify method: 'back'
+    this._setupMatching()
+    @domMatcher.setRootNode @wrapper[0]
     this
 
   _setupDocumentEvents: ->

h/js/services.coffee

@@ -61,17 +61,20 @@ class Hypothesis extends Annotator
     $window = @element.injector().get '$window'
     drafts = @element.injector().get 'drafts'
 
+    # Set up the bridge plugin, which bridges the main annotation methods
+    # between the host page and the panel widget.
+    whitelist = ['diffHTML', 'quote', 'ranges', 'target']
     this.addPlugin 'Bridge',
       origin: $location.search().xdm
       window: $window.parent
       formatter: (annotation) =>
         formatted = {}
-        for k, v of annotation when k in ['quote', 'ranges']
+        for k, v of annotation when k in whitelist
           formatted[k] = v
         formatted
       parser: (annotation) =>
         parsed = {}
-        for k, v of annotation when k in ['quote', 'ranges']
+        for k, v of annotation when k in whitelist
           parsed[k] = v
         parsed
 
@@ -203,12 +206,15 @@ class Hypothesis extends Annotator
   _setupViewer: -> this
   _setupEditor: -> this
 
+  # (Optionally) put some HTML formatting around a quote
+  getHtmlQuote: (quote) -> quote
+
   setupAnnotation: (annotation) ->
-    # This is needed until Annotator stops assuming ranges and highlights
-    # are always added.
-    unless annotation.ranges?
+    # Delagate to Annotator implementation after we give it a valid array of
+    # targets. This is needed until Annotator stops assuming targets need to be
+    unless annotation.target?
       annotation.highlights = []
-      annotation.ranges = []
+      annotation.target = []
     @plugins.Threading.thread annotation
     annotation
 

h/lib/annotator.auth.js

 /*
-** Annotator 1.2.6-dev-c4fcdfe
+** Annotator 1.2.5-dev-5a5e03a
 ** https://github.com/okfn/annotator/
 **
 ** Copyright 2012 Aron Carroll, Rufus Pollock, and Nick Stenning.
 ** Dual licensed under the MIT and GPLv3 licenses.
 ** https://github.com/okfn/annotator/blob/master/LICENSE
 **
-** Built at: 2013-01-29 11:20:24Z
+** Built at: 2013-02-25 17:23:26Z
 */
 
 (function() {

h/lib/annotator.permissions.js

 /*
-** Annotator 1.2.6-dev-c4fcdfe
+** Annotator 1.2.5-dev-5a5e03a
 ** https://github.com/okfn/annotator/
 **
 ** Copyright 2012 Aron Carroll, Rufus Pollock, and Nick Stenning.
 ** Dual licensed under the MIT and GPLv3 licenses.
 ** https://github.com/okfn/annotator/blob/master/LICENSE
 **
-** Built at: 2013-01-29 11:20:26Z
+** Built at: 2013-02-25 17:23:22Z
 */
 
 (function() {

h/lib/annotator.store.js

 /*
-** Annotator 1.2.6-dev-c4fcdfe
+** Annotator 1.2.5-dev-859d4e3
 ** https://github.com/okfn/annotator/
 **
 ** Copyright 2012 Aron Carroll, Rufus Pollock, and Nick Stenning.
 ** Dual licensed under the MIT and GPLv3 licenses.
 ** https://github.com/okfn/annotator/blob/master/LICENSE
 **
-** Built at: 2013-01-29 11:20:27Z
+** Built at: 2013-03-13 14:59:29Z
 */
 
 (function() {
@@ -65,10 +65,17 @@
     };
 
     Store.prototype.annotationCreated = function(annotation) {
-      var _this = this;
+      var quote, quoteHTML,
+        _this = this;
       if (__indexOf.call(this.annotations, annotation) < 0) {
         this.registerAnnotation(annotation);
+        quote = annotation.quote;
+        quoteHTML = annotation.quoteHTML;
+        delete annotation.quote;
+        delete annotation.quoteHTML;
         return this._apiRequest('create', annotation, function(data) {
+          annotation.quote = quote;
+          annotation.quoteHTML = quoteHTML;
           if (!(data.id != null)) {
             console.warn(Annotator._t("Warning: No ID returned from server for annotation "), annotation);
           }
@@ -80,9 +87,16 @@
     };
 
     Store.prototype.annotationUpdated = function(annotation) {
-      var _this = this;
+      var quote, quoteHTML,
+        _this = this;
       if (__indexOf.call(this.annotations, annotation) >= 0) {
+        quote = annotation.quote;
+        quoteHTML = annotation.quoteHTML;
+        delete annotation.quote;
+        delete annotation.quoteHTML;
         return this._apiRequest('update', annotation, (function(data) {
+          annotation.quote = quote;
+          annotation.quoteHTML = quoteHTML;
           return _this.updateAnnotation(annotation, data);
         }));
       }
@@ -180,7 +194,7 @@
     Store.prototype._urlFor = function(action, id) {
       var replaceWith, url;
       replaceWith = id != null ? '/' + id : '';
-      url = this.options.prefix != null ? this.options.prefix : '';
+      url = this.options.prefix || '/';
       url += this.options.urls[action];
       url = url.replace(/\/:id/, replaceWith);
       return url;

h/lib/text_match_engines.coffee

+# Naive text matcher 
+class window.DTM_ExactMatcher
+  constructor: ->
+    @distinct = true
+    @caseSensitive = false  
+ 
+  setDistinct: (value) -> @distinct = value
+
+  setCaseSensitive: (value) -> @caseSensitive = value
+        
+  search: (text, pattern) ->
+#    console.log "Searching for '" + pattern + "' in '" + text + "'."
+    pLen = pattern.length
+    results = []
+    index = 0
+    unless @caseSensitive
+      text = text.toLowerCase()
+      pattern = pattern.toLowerCase()
+    while (i = text.indexOf pattern) > -1
+      do =>
+#        console.log "Found '" + pattern + "' @ " + i +
+#            " (=" + (index + i) + ")"
+        results.push
+          start: index + i
+          end: index + i + pLen
+        if @distinct
+          text = text.substr i + pLen
+          index += i + pLen
+        else
+          text = text.substr i + 1
+          index += i + 1
+
+    results
+
+class window.DTM_RegexMatcher
+  constructor: ->
+    @caseSensitive = false  
+ 
+  setCaseSensitive: (value) -> @caseSensitive = value
+
+  search: (text, pattern) ->
+    re = new RegExp pattern, if @caseSensitive then "g" else "gi"
+    { start: m.index, end: m.index + m[0].length } while m = re.exec text
+                
+# diff-match-patch - based text matcher 
+class window.DTM_DMPMatcher
+  constructor: ->
+     @dmp = new diff_match_patch
+     @dmp.Diff_Timeout = 0
+     @caseSensitive = false
+
+  _reverse: (text) -> text.split("").reverse().join ""
+
+  # Use this to get the max allowed pattern length.
+  # Trying to use a longer pattern will give an error.
+  getMaxPatternLength: -> @dmp.Match_MaxBits
+
+  # The following example is a classic dilemma.
+  # There are two potential matches, one is close to the expected location
+  # but contains a one character error, the other is far from the expected
+  # location but is exactly the pattern sought after:
+  # 
+  # match_main("abc12345678901234567890abbc", "abc", 26)
+  # 
+  # Which result is returned (0 or 24) is determined by the
+  # MatchDistance property.
+  # 
+  # An exact letter match which is 'distance' characters away
+  # from the fuzzy location would score as a complete mismatch.
+  # For example, a distance of '0' requires the match be at the exact
+  # location specified, whereas a threshold of '1000' would require
+  # a perfect match to be within 800 characters of the expected location
+  # to be found using a 0.8 threshold (see below).
+  #
+  # The larger MatchDistance is, the slower search may take to compute.
+  # 
+  # This variable defaults to 1000.
+  setMatchDistance: (distance) -> @dmp.Match_Distance = distance
+  getMatchDistance: -> @dmp.Match_Distance
+
+  # MatchThreshold determines the cut-off value for a valid match.
+  #  
+  # If Match_Threshold is closer to 0, the requirements for accuracy
+  # increase. If Match_Threshold is closer to 1 then it is more likely
+  # that a match will be found. The larger Match_Threshold is, the slower
+  # search may take to compute.
+  # 
+  # This variable defaults to 0.5.
+  setMatchThreshold: (threshold) -> @dmp.Match_Threshold = threshold
+  getMatchThreshold: -> @dmp.Match_Threshold
+
+  getCaseSensitive: -> caseSensitive
+  setCaseSensitive: (value) -> @caseSensitive = value
+
+  # Given a text to search, a pattern to search for and an
+  # expected location in the text near which to find the pattern,
+  # return the location which matches closest.
+  # 
+  # The function will search for the best match based on both the number
+  # of character errors between the pattern and the potential match,
+  # as well as the distance between the expected location and the
+  # potential match.
+  #
+  # If no match is found, the function returns null.
+  search: (text, pattern, expectedStartLoc = 0, options = {}) ->
+#    console.log "In dtm search. text: '" + text + "', pattern: '" + pattern +
+#       "', expectedStartLoc: " + expectedStartLoc + ", options:"
+#    console.log options
+    if expectedStartLoc < 0 then throw new Error "Can't search at negative indices!"
+
+    unless @caseSensitive
+      text = text.toLowerCase()
+      pattern = pattern.toLowerCase()
+
+    pLen = pattern.length
+    maxLen = @getMaxPatternLength()
+
+    if pLen <= maxLen
+      result = @searchForSlice text, pattern, expectedStartLoc
+    else
+      startSlice = pattern.substr 0, maxLen
+      startPos = @searchForSlice text, startSlice, expectedStartLoc
+      if startPos?
+        startLen = startPos.end - startPos.start
+        endSlice = pattern.substr pLen - maxLen, maxLen
+        endLoc = startPos.start + pLen - maxLen
+        endPos = @searchForSlice text, endSlice, endLoc
+        if endPos?
+          endLen = endPos.end - endPos.start
+          matchLen = endPos.end - startPos.start
+          startIndex = startPos.start
+          endIndex = endPos.end
+        
+          if pLen*0.5 <= matchLen <= pLen*1.5
+            result =
+              start: startIndex
+              end: endPos.end
+#              data: 
+#                startError: startPos.data.error
+#                endError: endPos.data.error
+#                uncheckedMidSection: Math.max 0, matchLen - startLen - endLen
+#                lengthError: matchLen - pLen
+#          else
+#            console.log "Sorry, matchLen (" + matchLen + ") is not between " +
+#                0.5*pLen + " and " + 1.5*pLen
+#        else
+#          console.log "endSlice ('" + endSlice + "') not found"
+#      else
+#        console.log "startSlice ('" + startSlice + "') not found"
+
+    unless result? then return []
+
+    if options.withLevenhstein or options.withDiff
+      found = text.substr result.start, result.end - result.start
+      result.diff = @dmp.diff_main pattern, found
+      if options.withLevenshstein
+        result.lev = @dmp.diff_levenshtein result.diff
+      if options.withDiff
+        @dmp.diff_cleanupSemantic result.diff
+        result.diffHTML = @dmp.diff_prettyHtml result.diff
+
+    [result]
+
+  # Compare two string slices, get Levenhstein and visual diff
+  compare: (text1, text2) ->
+    unless (text1? and text2?)
+      throw new Error "Can not compare non-existing strings!"
+    result = {}
+    result.diff = @dmp.diff_main text1, text2
+    result.lev = @dmp.diff_levenshtein result.diff
+    result.errorLevel = result.lev / text1.length
+    @dmp.diff_cleanupSemantic result.diff
+    result.diffHTML = @dmp.diff_prettyHtml result.diff
+    result
+
+
+  # ============= Private part ==========================================
+  # You don't need to call the functions below this point manually
+
+  searchForSlice: (text, slice, expectedStartLoc) ->
+#    console.log "searchForSlice: '" + text + "', '" + slice + "', " +
+#        expectedStartLoc
+
+    r1 = @dmp.match_main text, slice, expectedStartLoc
+    startIndex = r1.index
+    if startIndex is -1 then return null
+        
+    txet = @_reverse text
+    nrettap = @_reverse slice
+    expectedEndLoc = startIndex + slice.length
+    expectedDneLoc = text.length - expectedEndLoc
+    r2 = @dmp.match_main txet, nrettap, expectedDneLoc
+    dneIndex = r2.index
+    endIndex = text.length - dneIndex
+
+    result =
+      start: startIndex
+      end: endIndex