Reader was reimplemented.

After experiments with loading the whole file at once, and
with decoding and parsing in separate thread, lazy reader
turned to be the fastest/least memory intensive solution.
Characters are now decoded in small batches.
This improved parsing speed by ~20%.

No global state anymore. Anchors are now zero terminated strings
and TagDirectives are a simple array. Event structure was changed
to prevent size increase.
Minor fixes and improvements.

doc/html/api/dyaml.dumper.html

@@ -112,57 +112,57 @@
 <dd><p>Set stream name. Used in debugging messages.</p>
 
 </dd>
-<dt class="d_decl">void <a name="resolver"></a><span class="ddoc_psymbol">resolver</span>(Resolver <a name="resolver"></a><span class="ddoc_psymbol">resolver</span>);
+<dt class="d_decl">@property void <a name="resolver"></a><span class="ddoc_psymbol">resolver</span>(Resolver <a name="resolver"></a><span class="ddoc_psymbol">resolver</span>);
 </dt>
 <dd><p>Specify custom Resolver to use.</p>
 
 </dd>
-<dt class="d_decl">void <a name="representer"></a><span class="ddoc_psymbol">representer</span>(Representer <a name="representer"></a><span class="ddoc_psymbol">representer</span>);
+<dt class="d_decl">@property void <a name="representer"></a><span class="ddoc_psymbol">representer</span>(Representer <a name="representer"></a><span class="ddoc_psymbol">representer</span>);
 </dt>
 <dd><p>Specify custom Representer to use.</p>
 
 </dd>
-<dt class="d_decl">void <a name="canonical"></a><span class="ddoc_psymbol">canonical</span>(in bool <a name="canonical"></a><span class="ddoc_psymbol">canonical</span>);
+<dt class="d_decl">@property void <a name="canonical"></a><span class="ddoc_psymbol">canonical</span>(bool <a name="canonical"></a><span class="ddoc_psymbol">canonical</span>);
 </dt>
 <dd><p>Write scalars in canonical form?</p>
 
 </dd>
-<dt class="d_decl">void <a name="indent"></a><span class="ddoc_psymbol">indent</span>(in uint <a name="indent"></a><span class="ddoc_psymbol">indent</span>);
+<dt class="d_decl">@property void <a name="indent"></a><span class="ddoc_psymbol">indent</span>(uint <a name="indent"></a><span class="ddoc_psymbol">indent</span>);
 </dt>
 <dd><p>Set indentation width. 2 by default. Must not be zero.</p>
 
 </dd>
-<dt class="d_decl">void <a name="textWidth"></a><span class="ddoc_psymbol">textWidth</span>(in uint <b>width</b>);
+<dt class="d_decl">@property void <a name="textWidth"></a><span class="ddoc_psymbol">textWidth</span>(uint <b>width</b>);
 </dt>
 <dd><p>Set preferred text width.</p>
 
 </dd>
-<dt class="d_decl">void <a name="lineBreak"></a><span class="ddoc_psymbol">lineBreak</span>(in LineBreak <a name="lineBreak"></a><span class="ddoc_psymbol">lineBreak</span>);
+<dt class="d_decl">@property void <a name="lineBreak"></a><span class="ddoc_psymbol">lineBreak</span>(LineBreak <a name="lineBreak"></a><span class="ddoc_psymbol">lineBreak</span>);
 </dt>
 <dd><p>Set line break to use. Unix by default.</p>
 
 </dd>
-<dt class="d_decl">void <a name="encoding"></a><span class="ddoc_psymbol">encoding</span>(in Encoding <a name="encoding"></a><span class="ddoc_psymbol">encoding</span>);
+<dt class="d_decl">@property void <a name="encoding"></a><span class="ddoc_psymbol">encoding</span>(Encoding <a name="encoding"></a><span class="ddoc_psymbol">encoding</span>);
 </dt>
 <dd><p>Set character encoding to use. UTF-8 by default.</p>
 
 </dd>
-<dt class="d_decl">void <a name="explicitStart"></a><span class="ddoc_psymbol">explicitStart</span>(in bool <b>explicit</b>);
+<dt class="d_decl">@property void <a name="explicitStart"></a><span class="ddoc_psymbol">explicitStart</span>(bool <b>explicit</b>);
 </dt>
 <dd><p>Always explicitly write document start?</p>
 
 </dd>
-<dt class="d_decl">void <a name="explicitEnd"></a><span class="ddoc_psymbol">explicitEnd</span>(in bool <b>explicit</b>);
+<dt class="d_decl">@property void <a name="explicitEnd"></a><span class="ddoc_psymbol">explicitEnd</span>(bool <b>explicit</b>);
 </dt>
 <dd><p>Always explicitly write document end?</p>
 
 </dd>
-<dt class="d_decl">void <a name="YAMLVersion"></a><span class="ddoc_psymbol">YAMLVersion</span>(in string <a name="YAMLVersion"></a><span class="ddoc_psymbol">YAMLVersion</span>);
+<dt class="d_decl">@property void <a name="YAMLVersion"></a><span class="ddoc_psymbol">YAMLVersion</span>(string <a name="YAMLVersion"></a><span class="ddoc_psymbol">YAMLVersion</span>);
 </dt>
 <dd><p>Specify YAML version string. "1.1" by default.</p>
 
 </dd>
-<dt class="d_decl">void <a name="tagDirectives"></a><span class="ddoc_psymbol">tagDirectives</span>(string[string] <b>tags</b>);
+<dt class="d_decl">@property void <a name="tagDirectives"></a><span class="ddoc_psymbol">tagDirectives</span>(string[string] <b>tags</b>);
 </dt>
 <dd><p>Specify tag directives.
 </p>

doc/html/api/dyaml.loader.html

@@ -98,11 +98,11 @@
 </pre>
 </div>
 
-<dl><dt class="d_decl">this(in const(immutable(char)[]) <b>filename</b>);
+<dl><dt class="d_decl">this(string <b>filename</b>);
 </dt>
 <dd><p>Construct a Loader to load YAML from a file.
 </p>
-<b>Parameters:</b><div class="pbr"><table class=parms><tr><td valign=top>const(immutable(char)[]) <b>filename</b></td>
+<b>Parameters:</b><div class="pbr"><table class=parms><tr><td valign=top>string <b>filename</b></td>
 <td valign=top>Name of the file to load from.</td></tr>
 </table></div>
 <b>Throws:</b><div class="pbr">YAMLException if the file could not be opened or read.</div>
@@ -138,6 +138,9 @@
 <dd><p>Load single YAML document.
 </p>
 <p>If none or more than one YAML document is found, this throws a YAMLException.
+<br>
+
+ This can only be called once; this is enforced by contract.
 
 </p>
 <b>Returns:</b><div class="pbr">Root node of the document.
@@ -155,6 +158,9 @@
  them all at once. Calling <a name="loadAll"></a><span class="ddoc_psymbol">loadAll</span> after iterating over the node or
  vice versa will not return any documents, as they have all been parsed
  already.
+<br>
+
+ This can only be called once; this is enforced by contract.
 
 </p>
 <b>Returns:</b><div class="pbr">Array of root nodes of all documents in the file/stream.
@@ -168,6 +174,9 @@
 <dd><p>Foreach over YAML documents.
 </p>
 <p>Parses documents lazily, when they are needed.
+<br>
+
+ Foreach over a Loader can only be used once; this is enforced by contract.
 
 </p>
 <b>Throws:</b><div class="pbr">YAMLException on a parsing error.</div>

doc/html/api/dyaml.representer.html

@@ -182,7 +182,7 @@
 </div>
 
 </dd>
-<dt class="d_decl">Node <a name="representScalar"></a><span class="ddoc_psymbol">representScalar</span>(in string <b>tag</b>, string <b>scalar</b>, ScalarStyle <b>style</b> = (ScalarStyle).Invalid);
+<dt class="d_decl">Node <a name="representScalar"></a><span class="ddoc_psymbol">representScalar</span>(string <b>tag</b>, string <b>scalar</b>, ScalarStyle <b>style</b> = (ScalarStyle).Invalid);
 </dt>
 <dd><p>Represent a scalar with specified tag.
 </p>
@@ -217,7 +217,7 @@
 </p>
 
 </dd>
-<dt class="d_decl">Node <a name="representSequence"></a><span class="ddoc_psymbol">representSequence</span>(in string <b>tag</b>, Node[] <b>sequence</b>, CollectionStyle <b>style</b> = (CollectionStyle).Invalid);
+<dt class="d_decl">Node <a name="representSequence"></a><span class="ddoc_psymbol">representSequence</span>(string <b>tag</b>, Node[] <b>sequence</b>, CollectionStyle <b>style</b> = (CollectionStyle).Invalid);
 </dt>
 <dd><p>Represent a sequence with specified tag, representing children first.
 </p>
@@ -256,7 +256,7 @@
 </p>
 
 </dd>
-<dt class="d_decl">Node <a name="representMapping"></a><span class="ddoc_psymbol">representMapping</span>(in string <b>tag</b>, Pair[] <b>pairs</b>, CollectionStyle <b>style</b> = (CollectionStyle).Invalid);
+<dt class="d_decl">Node <a name="representMapping"></a><span class="ddoc_psymbol">representMapping</span>(string <b>tag</b>, Pair[] <b>pairs</b>, CollectionStyle <b>style</b> = (CollectionStyle).Invalid);
 </dt>
 <dd><p>Represent a mapping with specified tag, representing children first.
 </p>

doc/html/articles/spec_differences.html

@@ -138,7 +138,7 @@ struct appears in Phobos.</p>
     </div>
     <div class="footer">
         &copy; Copyright 2011, Ferdinand Majerech. Based on PyYAML http://www.pyyaml.org by Kirill Simonov.
-      Last updated on Oct 30, 2011.
+      Last updated on Nov 15, 2011.
       Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.7.
     </div>
   </body>

doc/html/index.html

@@ -104,7 +104,7 @@
     </div>
     <div class="footer">
         &copy; Copyright 2011, Ferdinand Majerech. Based on PyYAML http://www.pyyaml.org by Kirill Simonov.
-      Last updated on Oct 30, 2011.
+      Last updated on Nov 15, 2011.
       Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.7.
     </div>
   </body>

doc/html/search.html

@@ -87,7 +87,7 @@
     </div>
     <div class="footer">
         &copy; Copyright 2011, Ferdinand Majerech. Based on PyYAML http://www.pyyaml.org by Kirill Simonov.
-      Last updated on Oct 30, 2011.
+      Last updated on Nov 15, 2011.
       Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.7.
     </div>
   </body>

doc/html/tutorials/custom_types.html

@@ -368,7 +368,7 @@ directory of the D:YAML package.</p>
     </div>
     <div class="footer">
         &copy; Copyright 2011, Ferdinand Majerech. Based on PyYAML http://www.pyyaml.org by Kirill Simonov.
-      Last updated on Oct 30, 2011.
+      Last updated on Nov 15, 2011.
       Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.7.
     </div>
   </body>

doc/html/tutorials/getting_started.html

@@ -237,7 +237,7 @@ example in the <tt class="docutils literal"><span class="pre">example/getting_st
     </div>
     <div class="footer">
         &copy; Copyright 2011, Ferdinand Majerech. Based on PyYAML http://www.pyyaml.org by Kirill Simonov.
-      Last updated on Oct 30, 2011.
+      Last updated on Nov 15, 2011.
       Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.7.
     </div>
   </body>

doc/html/tutorials/yaml_syntax.html

@@ -330,7 +330,7 @@ Some of these might change in the future (especially !!map and !!set).</p>
     </div>
     <div class="footer">
         &copy; Copyright 2011, Ferdinand Majerech. Based on PyYAML http://www.pyyaml.org by Kirill Simonov.
-      Last updated on Oct 30, 2011.
+      Last updated on Nov 15, 2011.
       Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.0.7.
     </div>
   </body>

dyaml/anchor.d

@@ -7,24 +7,7 @@
 ///YAML anchor.
 module dyaml.anchor;
 
-import dyaml.sharedobject;
+import dyaml.zerostring;
-
 
 ///YAML anchor (reference) struct. Encapsulates an anchor to save memory.
-struct Anchor
+alias ZeroString!"Anchor" Anchor;
-{
-    public:
-        mixin SharedObject!(string, Anchor);
-
-        ///Construct an anchor from a string representation.
-        this(string anchor)
-        {
-            if(anchor is null || anchor == "")
-            {
-                index_ = uint.max;
-                return;
-            }
-
-            add(anchor);
-        }
-}

dyaml/dumper.d

@@ -25,7 +25,7 @@ import dyaml.node;
 import dyaml.representer;
 import dyaml.resolver;
 import dyaml.serializer;
-import dyaml.tagdirectives;
+import dyaml.tagdirective;
 
 
 /**
@@ -146,7 +146,7 @@ struct Dumper
         ///YAML version string.
         string YAMLVersion_ = "1.1";
         ///Tag directives to use.
-        TagDirectives tags_ = TagDirectives();
+        TagDirective[] tags_ = null;
         ///Always write document start?
         bool explicitStart_ = false;
         ///Always write document end?
@@ -167,7 +167,7 @@ struct Dumper
          *
          * Throws: YAMLException if the file can not be dumped to (e.g. cannot be opened).
          */
-        this(in string filename)
+        this(string filename)
         {
             name_ = filename;
             try{this(new File(filename, FileMode.OutNew));}
@@ -184,20 +184,16 @@ struct Dumper
             resolver_ = defaultResolver_;
             representer_ = defaultRepresenter_;
             stream_ = stream;
-            Anchor.addReference();
-            TagDirectives.addReference();
         }
 
         ///Destroy the Dumper.
         ~this()
         {
-            Anchor.removeReference();
-            TagDirectives.removeReference();
             YAMLVersion_ = null;
         }
 
         ///Set stream _name. Used in debugging messages.
-        @property void name(in string name)
+        @property void name(string name)
         {
             name_ = name;
         }
@@ -217,13 +213,13 @@ struct Dumper
         }
 
         ///Write scalars in _canonical form?
-        @property void canonical(in bool canonical)
+        @property void canonical(bool canonical)
         {
             canonical_ = canonical;
         }
 
         ///Set indentation width. 2 by default. Must not be zero.
-        @property void indent(in uint indent)
+        @property void indent(uint indent)
         in
         {   
             assert(indent != 0, "Can't use zero YAML indent width");
@@ -234,37 +230,37 @@ struct Dumper
         }
 
         ///Set preferred text _width.
-        @property void textWidth(in uint width)
+        @property void textWidth(uint width)
         {
             textWidth_ = width;
         }
 
         ///Set line break to use. Unix by default.
-        @property void lineBreak(in LineBreak lineBreak)
+        @property void lineBreak(LineBreak lineBreak)
         {
             lineBreak_ = lineBreak;
         }
 
         ///Set character _encoding to use. UTF-8 by default.
-        @property void encoding(in Encoding encoding)
+        @property void encoding(Encoding encoding)
         {
             encoding_ = encoding;
         }    
 
         ///Always explicitly write document start?
-        @property void explicitStart(in bool explicit)
+        @property void explicitStart(bool explicit)
         {
             explicitStart_ = explicit;
         }
 
         ///Always explicitly write document end?
-        @property void explicitEnd(in bool explicit)
+        @property void explicitEnd(bool explicit)
         {
             explicitEnd_ = explicit;
         }
 
         ///Specify YAML version string. "1.1" by default.
-        @property void YAMLVersion(in string YAMLVersion)
+        @property void YAMLVersion(string YAMLVersion)
         {
             YAMLVersion_ = YAMLVersion;
         }
@@ -301,16 +297,16 @@ struct Dumper
          */
         @property void tagDirectives(string[string] tags)
         {
-            tagDirective[] t;
+            TagDirective[] t;
             foreach(handle, prefix; tags)
             {
                 assert(handle.length >= 1 && handle[0] == '!' && handle[$ - 1] == '!',
                        "A tag handle is empty or does not start and end with a "
                        "'!' character : " ~ handle);
                 assert(prefix.length >= 1, "A tag prefix is empty");
-                t ~= tagDirective(handle, prefix);
+                t ~= TagDirective(handle, prefix);
             }
-            tags_ = TagDirectives(t);
+            tags_ = t;
         }
 
         /**
@@ -352,7 +348,7 @@ struct Dumper
          *
          * Throws:  YAMLException if unable to emit.
          */
-        void emit(in Event[] events)
+        void emit(Event[] events)
         {
             try
             {

dyaml/emitter.d

@@ -71,11 +71,11 @@ private mixin FastCharSearch!"\n\u0085\u2028\u2029"d newlineSearch_;
 struct Emitter 
 {
     private:
-        alias dyaml.tagdirectives.tagDirective tagDirective;
+        alias dyaml.tagdirective.TagDirective TagDirective;
 
         ///Default tag handle shortcuts and replacements.
-        static tagDirective[] defaultTagDirectives_ = 
+        static TagDirective[] defaultTagDirectives_ = 
-            [tagDirective("!", "!"), tagDirective("!!", "tag:yaml.org,2002:")];
+            [TagDirective("!", "!"), TagDirective("!!", "tag:yaml.org,2002:")];
 
         ///Stream to write to.
         Stream stream_;
@@ -135,7 +135,7 @@ struct Emitter
         LineBreak bestLineBreak_;
 
         ///Tag directive handle - prefix pairs.
-        tagDirective[] tagDirectives_;
+        TagDirective[] tagDirectives_;
 
         ///Anchor/alias to process.
         string preparedAnchor_ = null;
@@ -193,7 +193,7 @@ struct Emitter
         }
 
         ///Emit an event. Throws EmitterException on error.
-        void emit(immutable Event event)
+        void emit(Event event)
         {
             events_.push(event);
             while(!needMoreEvents())
@@ -254,7 +254,7 @@ struct Emitter
         {
             if(events_.length == 0){return true;}
 
-            immutable event = events_.peek();
+            immutable event = cast(immutable Event)events_.peek();
             if(event.id == EventID.DocumentStart){return needEvents(1);}
             if(event.id == EventID.SequenceStart){return needEvents(2);}
             if(event.id == EventID.MappingStart) {return needEvents(3);}
@@ -274,7 +274,7 @@ struct Emitter
             events_.next();
             while(!events_.iterationOver())
             {
-                immutable event = events_.next();
+                immutable event = cast(immutable Event)events_.next();
                 static starts = [EventID.DocumentStart, EventID.SequenceStart, EventID.MappingStart];
                 static ends   = [EventID.DocumentEnd, EventID.SequenceEnd, EventID.MappingEnd];
                 if(starts.canFind(event.id))   {++level;}
@@ -347,8 +347,8 @@ struct Emitter
             if(event_.id == EventID.DocumentStart)
             {
                 const YAMLVersion = event_.value;
-                const tagDirectives = event_.tagDirectives;
+                auto tagDirectives = event_.tagDirectives;
-                if(openEnded_ && (YAMLVersion !is null || !tagDirectives.isNull()))
+                if(openEnded_ && (YAMLVersion !is null || tagDirectives !is null))
                 {
                     writeIndicator("...", true);
                     writeIndent();
@@ -359,10 +359,10 @@ struct Emitter
                     writeVersionDirective(prepareVersion(YAMLVersion));
                 }
 
-                if(!tagDirectives.isNull())
+                if(tagDirectives !is null)
                 {
-                    tagDirectives_ = tagDirectives.get;
+                    tagDirectives_ = tagDirectives;
-                    sort!"icmp(a[0], b[0]) < 0"(tagDirectives_);
+                    sort!"icmp(a.handle, b.handle) < 0"(tagDirectives_);
 
                     foreach(ref pair; tagDirectives_)
                     {
@@ -371,7 +371,7 @@ struct Emitter
                     }
                 }
 
-                bool eq(ref tagDirective a, ref tagDirective b){return a.handle == b.handle;}
+                bool eq(ref TagDirective a, ref TagDirective b){return a.handle == b.handle;}
                 //Add any default tag directives that have not been overriden.
                 foreach(ref def; defaultTagDirectives_) 
                 {
@@ -382,7 +382,7 @@ struct Emitter
                 }
 
                 const implicit = first && !event_.explicitDocument && !canonical_ &&
-                                 YAMLVersion is null && tagDirectives.isNull() && 
+                                 YAMLVersion is null && tagDirectives is null && 
                                  !checkEmptyDocument();
                 if(!implicit)
                 {
@@ -684,7 +684,7 @@ struct Emitter
                 return false;
             }
 
-            immutable event = events_.peek();
+            immutable event = cast(immutable Event)events_.peek();
             const emptyScalar = event.id == EventID.Scalar && event.anchor.isNull() &&
                                 event.tag.isNull() && event.implicit && event.value == "";
             return emptyScalar;
@@ -933,14 +933,14 @@ struct Emitter
             string suffix = tagString;
 
             //Sort lexicographically by prefix.
-            sort!"icmp(a[1], b[1]) < 0"(tagDirectives_);
+            sort!"icmp(a.prefix, b.prefix) < 0"(tagDirectives_);
             foreach(ref pair; tagDirectives_)
             {
-                auto prefix = pair[1];
+                auto prefix = pair.prefix;
                 if(tagString.startsWith(prefix) && 
                    (prefix != "!" || prefix.length < tagString.length))
                 {
-                    handle = pair[0];
+                    handle = pair.handle;
                     suffix = tagString[prefix.length .. $];
                 }
             }

dyaml/event.d

@@ -19,7 +19,7 @@ import dyaml.encoding;
 import dyaml.exception;
 import dyaml.reader;
 import dyaml.tag;
-import dyaml.tagdirectives;
+import dyaml.tagdirective;
 import dyaml.style;
 
 
@@ -55,30 +55,40 @@ struct Event
     Mark startMark;
     ///End position of the event in file/stream.
     Mark endMark;
-    ///Anchor of the event, if any.
+    union
-    Anchor anchor;
+    {
-    ///Tag of the event, if any.
+        struct
-    Tag tag;
+        {
+            ///Anchor of the event, if any.
+            Anchor anchor;
+            ///Tag of the event, if any.
+            Tag tag;
+        }
+        ///Tag directives, if this is a DocumentStart.
+        //TagDirectives tagDirectives;
+        TagDirective[] tagDirectives;
+    }
     ///Event type.
     EventID id = EventID.Invalid;
     ///Style of scalar event, if this is a scalar event.
-    ScalarStyle scalarStyle;
+    ScalarStyle scalarStyle = ScalarStyle.Invalid;
-    ///Should the tag be implicitly resolved? 
+    union
-    bool implicit;
+    {
+        ///Should the tag be implicitly resolved? 
+        bool implicit;
+        /**
+         * Is this document event explicit?
+         *
+         * Used if this is a DocumentStart or DocumentEnd.
+         */
+        bool explicitDocument;
+    }
     ///TODO figure this out - Unknown, used by PyYAML with Scalar events.
     bool implicit_2;
-    /**
-     * Is this document event explicit?
-     *
-     * Used if this is a DocumentStart or DocumentEnd.
-     */
-    alias implicit explicitDocument;
-    ///Tag directives, if this is a DocumentStart.
-    TagDirectives tagDirectives;
     ///Encoding of the stream, if this is a StreamStart.
     Encoding encoding;
     ///Collection style, if this is a SequenceStart or MappingStart.
-    CollectionStyle collectionStyle;
+    CollectionStyle collectionStyle = CollectionStyle.Invalid;
 
     ///Is this a null (uninitialized) event?
     @property bool isNull() const {return id == EventID.Invalid;}
@@ -96,7 +106,12 @@ struct Event
  */
 Event event(EventID id)(in Mark start, in Mark end, in Anchor anchor = Anchor()) pure
 {
-    return Event(null, start, end, anchor, Tag(), id);
+    Event result;
+    result.startMark = start;
+    result.endMark = end;
+    result.anchor = anchor;
+    result.id = id;
+    return result;
 }
 
 /**
@@ -114,8 +129,15 @@ Event collectionStartEvent(EventID id)(in Mark start, in Mark end, in Anchor anc
 {
     static assert(id == EventID.SequenceStart || id == EventID.SequenceEnd ||
                   id == EventID.MappingStart || id == EventID.MappingEnd);
-    return Event(null, start, end, anchor, tag, id, ScalarStyle.Invalid, implicit,
+    Event result;
-                 false, TagDirectives(), Encoding.UTF_8, style);
+    result.startMark = start;
+    result.endMark = end;
+    result.anchor = anchor;
+    result.tag = tag;
+    result.id = id;
+    result.implicit = implicit;
+    result.collectionStyle = style;
+    return result;
 }
 
 /**
@@ -127,8 +149,12 @@ Event collectionStartEvent(EventID id)(in Mark start, in Mark end, in Anchor anc
  */
 Event streamStartEvent(in Mark start, in Mark end, in Encoding encoding) pure
 {
-    return Event(null, start, end, Anchor(), Tag(), EventID.StreamStart, 
+    Event result;
-                 ScalarStyle.Invalid, false, false, TagDirectives(), encoding);
+    result.startMark = start;
+    result.endMark = end;
+    result.id = EventID.StreamStart;
+    result.encoding = encoding;
+    return result;
 }
 
 ///Aliases for simple events.
@@ -151,10 +177,16 @@ alias collectionStartEvent!(EventID.MappingStart) mappingStartEvent;
  *          tagDirectives = Tag directives of the document.
  */
 Event documentStartEvent(in Mark start, in Mark end, bool explicit, string YAMLVersion,
-                         in TagDirectives tagDirectives) pure
+                         TagDirective[] tagDirectives) pure
 {
-    return Event(YAMLVersion, start, end, Anchor(), Tag(), EventID.DocumentStart, 
+    Event result;
-                 ScalarStyle.Invalid, explicit, false, tagDirectives);
+    result.value = YAMLVersion;
+    result.startMark = start;
+    result.endMark = end;
+    result.id = EventID.DocumentStart;
+    result.explicitDocument = explicit;
+    result.tagDirectives = tagDirectives;
+    return result;
 }
 
 /**
@@ -166,8 +198,12 @@ Event documentStartEvent(in Mark start, in Mark end, bool explicit, string YAMLV
  */
 Event documentEndEvent(in Mark start, in Mark end, bool explicit) pure
 {
-    return Event(null, start, end, Anchor(), Tag(), EventID.DocumentEnd,
+    Event result;
-                 ScalarStyle.Invalid, explicit);
+    result.startMark = start;
+    result.endMark = end;
+    result.id = EventID.DocumentEnd;
+    result.explicitDocument = explicit;
+    return result;
 }
 
 /**
@@ -185,6 +221,15 @@ Event scalarEvent(in Mark start, in Mark end, in Anchor anchor, in Tag tag,
                   in Tuple!(bool, bool) implicit, in string value, 
                   in ScalarStyle style = ScalarStyle.Invalid) pure
 {
-    return Event(value, start, end, anchor, tag, EventID.Scalar, style, implicit[0],
+    Event result;
-                 implicit[1]);
+    result.value = value;
+    result.startMark = start;
+    result.endMark = end;
+    result.anchor = anchor;
+    result.tag = tag;
+    result.id = EventID.Scalar;
+    result.scalarStyle = style;
+    result.implicit = implicit[0];
+    result.implicit_2 = implicit[1];
+    return result;
 }

dyaml/loader.d

@@ -13,7 +13,6 @@ module dyaml.loader;
 import std.exception;
 import std.stream;
 
-import dyaml.anchor;
 import dyaml.composer;
 import dyaml.constructor;
 import dyaml.event;
@@ -23,7 +22,6 @@ import dyaml.parser;
 import dyaml.reader;
 import dyaml.resolver;
 import dyaml.scanner;
-import dyaml.tagdirectives;
 import dyaml.token;
 
 
@@ -114,6 +112,8 @@ struct Loader
         Constructor constructor_;
         ///Name of the input file or stream, used in error messages.
         string name_ = "<unknown>";
+        ///Are we done loading?
+        bool done_ = false;
 
     public:
         @disable this();
@@ -127,7 +127,7 @@ struct Loader
          *
          * Throws:  YAMLException if the file could not be opened or read.
          */
-        this(in string filename)
+        this(string filename)
         {
             name_ = filename;
             try{this(new File(filename));}
@@ -154,8 +154,6 @@ struct Loader
                 parser_      = new Parser(scanner_);
                 resolver_    = defaultResolver_;
                 constructor_ = defaultConstructor_;
-                Anchor.addReference();
-                TagDirectives.addReference();
             }
             catch(YAMLException e)
             {
@@ -167,8 +165,6 @@ struct Loader
         ///Destroy the Loader.
         ~this()
         {
-            Anchor.removeReference();
-            TagDirectives.removeReference();
             clear(reader_);
             clear(scanner_);
             clear(parser_);
@@ -196,6 +192,8 @@ struct Loader
          * Load single YAML document.
          *
          * If none or more than one YAML document is found, this throws a YAMLException.
+         *
+         * This can only be called once; this is enforced by contract.
          *                  
          * Returns: Root node of the document.
          *
@@ -203,9 +201,15 @@ struct Loader
          *          or on a YAML parsing error.
          */
         Node load()
+        in
+        {
+            assert(!done_, "Loader: Trying to load YAML twice");
+        }
+        body
         {
             try
             {
+                scope(exit){done_ = true;}
                 auto composer = new Composer(parser_, resolver_, constructor_);
                 enforce(composer.checkNode(), new YAMLException("No YAML document to load"));
                 return composer.getSingleNode();
@@ -224,6 +228,8 @@ struct Loader
          * them all at once. Calling loadAll after iterating over the node or
          * vice versa will not return any documents, as they have all been parsed
          * already.
+         *
+         * This can only be called once; this is enforced by contract.
          *                  
          * Returns: Array of root nodes of all documents in the file/stream.
          *
@@ -241,10 +247,18 @@ struct Loader
          *
          * Parses documents lazily, when they are needed.
          *
+         * Foreach over a Loader can only be used once; this is enforced by contract.
+         *
          * Throws: YAMLException on a parsing error.
          */
         int opApply(int delegate(ref Node) dg)
+        in
         {
+            assert(!done_, "Loader: Trying to load YAML twice");
+        }
+        body
+        {
+            scope(exit){done_ = true;}
             try
             {
                 auto composer = new Composer(parser_, resolver_, constructor_);
@@ -284,11 +298,11 @@ struct Loader
         }
 
         //Parse and return all events. Used for debugging.
-        Event[] parse()
+        immutable(Event)[] parse()
         {
             try
             {
-                Event[] result;
+                immutable(Event)[] result;
                 while(parser_.checkEvent()){result ~= parser_.getEvent();}
                 return result;
             }

dyaml/parser.d

@@ -24,7 +24,7 @@ import dyaml.scanner;
 import dyaml.style;
 import dyaml.token;
 import dyaml.tag;
-import dyaml.tagdirectives;
+import dyaml.tagdirective;
 
 
 package:
@@ -108,10 +108,10 @@ final class Parser
 {
     private:
         ///Default tag handle shortcuts and replacements.
-        static tagDirective[] defaultTagDirectives_;
+        static TagDirective[] defaultTagDirectives_;
         static this()
         {
-            defaultTagDirectives_ = [tagDirective("!", "!"), tagDirective("!!", "tag:yaml.org,2002:")];
+            defaultTagDirectives_ = [TagDirective("!", "!"), TagDirective("!!", "tag:yaml.org,2002:")];
         }
 
         ///Scanner providing YAML tokens.
@@ -123,7 +123,7 @@ final class Parser
         ///YAML version string.
         string YAMLVersion_ = null;
         ///Tag handle shortcuts and replacements.
-        tagDirective[] tagDirectives_;
+        TagDirective[] tagDirectives_;
 
         ///Stack of states.
         Array!(Event delegate()) states_;
@@ -193,13 +193,13 @@ final class Parser
          *
          * Must not be called if there are no events left.
          */
-        Event peekEvent()
+        immutable(Event) peekEvent()
         {
             if(currentEvent_.isNull && state_ !is null)
             {
                 currentEvent_ = state_();
             }
-            if(!currentEvent_.isNull){return currentEvent_;}
+            if(!currentEvent_.isNull){return cast(immutable Event)currentEvent_;}
             assert(false, "No event left to peek");
         }
 
@@ -208,7 +208,7 @@ final class Parser
          *
          * Must not be called if there are no events left.
          */
-        Event getEvent()
+        immutable(Event) getEvent()
         {
             //Get the next event and proceed further.
             if(currentEvent_.isNull && state_ !is null)
@@ -218,7 +218,7 @@ final class Parser
 
             if(!currentEvent_.isNull)
             {
-                immutable Event result = currentEvent_;
+                immutable Event result = cast(immutable Event)currentEvent_;
                 currentEvent_.id = EventID.Invalid;
                 return result;
             }
@@ -273,7 +273,7 @@ final class Parser
                 states_ ~= &parseDocumentEnd;
                 state_ = &parseBlockNode;
                 
-                return documentStartEvent(token.startMark, token.endMark, false, null, TagDirectives());
+                return documentStartEvent(token.startMark, token.endMark, false, null, null);
             }
             return parseDocumentStart();
         }
@@ -336,7 +336,7 @@ final class Parser
         }
 
         ///Process directives at the beginning of a document.
-        TagDirectives processDirectives()
+        TagDirective[] processDirectives()
         {
             //Destroy version and tag handles from previous document.
             YAMLVersion_ = null;
@@ -367,21 +367,21 @@ final class Parser
                     foreach(ref pair; tagDirectives_)
                     {
                         //handle
-                        const h = pair[0];
+                        const h = pair.handle;
                         enforce(h != handle, new Error("Duplicate tag handle: " ~ handle,
                                                        token.startMark));
                     }
-                    tagDirectives_ ~= tagDirective(handle, parts[2]);
+                    tagDirectives_ ~= TagDirective(handle, parts[2]);
                 }
             }
 
-            TagDirectives value = tagDirectives_.length == 0 ? TagDirectives() : TagDirectives(tagDirectives_);
+            TagDirective[] value = tagDirectives_;
 
             //Add any default tag handles that haven't been overridden.
             foreach(ref defaultPair; defaultTagDirectives_)
             {
                 bool found = false;
-                foreach(ref pair; tagDirectives_) if(defaultPair[0] == pair[0])
+                foreach(ref pair; tagDirectives_) if(defaultPair.handle == pair.handle)
                 {
                     found = true;
                     break;
@@ -540,10 +540,9 @@ final class Parser
                 string replacement = null;
                 foreach(ref pair; tagDirectives_)
                 {
-                    //pair[0] is handle, pair[1] replacement.
+                    if(pair.handle == handle)
-                    if(pair[0] == handle)
                     {
-                        replacement = pair[1];
+                        replacement = pair.prefix;
                         break;
                     }
                 }

dyaml/queue.d

@@ -37,7 +37,7 @@ struct Queue(T)
         ///Linked list node containing one element and pointer to the next node.
         struct Node
         {
-            T payload_ = T.init;
+            T payload_;
             Node* next_ = null;
         }
 
@@ -90,7 +90,7 @@ struct Queue(T)
         }
 
         ///Push new item to the queue.
-        void push(in T item)
+        void push(T item)
         {
             Node* newLast = allocate!Node(item, cast(Node*)null);
             if(last_ !is null){last_.next_ = newLast;}
@@ -100,7 +100,7 @@ struct Queue(T)
         }
 
         ///Insert a new item putting it to specified index in the linked list.
-        void insert(in T item, in size_t idx)
+        void insert(T item, in size_t idx)
         in
         {
             assert(idx <= length_);
@@ -155,7 +155,7 @@ struct Queue(T)
         }
 
         ///Return the next element in the queue.
-        ref const(T) peek() const
+        ref inout(T) peek() inout
         in
         {
             assert(!empty, "Trying to peek at an element in an empty queue");

dyaml/reader.d

@@ -7,7 +7,9 @@
 module dyaml.reader;
 
 
+import core.stdc.stdlib;
 import core.stdc.string;
+import core.thread;
 
 import std.algorithm;
 import std.conv;
@@ -34,47 +36,30 @@ class ReaderException : YAMLException
     }
 }
 
-///Reads data from a stream and converts it to UTF-32 (dchar) data.
+///Lazily reads and decodes data from stream, only storing as much as needed at any moment.
 final class Reader
 {
     private:
-        ///Input stream.
+        //Input stream.
         EndianStream stream_;
-        ///Allocated space for buffer_.
+        //Allocated space for buffer_.
-        dchar[] bufferAllocated_;
+        dchar[] bufferAllocated_ = null;
-        ///Buffer of currently loaded characters.
+        //Buffer of currently loaded characters.
-        dchar[] buffer_;
+        dchar[] buffer_ = null;
-        ///Current position within buffer. Only data after this position can be read.
+        //Current position within buffer. Only data after this position can be read.
         uint bufferOffset_ = 0;
-        ///Index of the current character in the stream.
+        //Index of the current character in the stream.
         size_t charIndex_ = 0;
-        ///Encoding of the input stream.
+        //Current line in file.
-        Encoding encoding_;
-        ///Current line in file.
         uint line_;
-        ///Current column in file.
+        //Current column in file.
         uint column_;
-        ///Number of bytes still available (not read) in the stream.
+        //Decoder reading data from file and decoding it to UTF-32.
-        size_t available_;
+        UTFFastDecoder decoder_;
-
-        ///Capacity of raw buffers.
-        static immutable bufferLength8_ = 8;
-        ///Capacity of raw buffers.
-        static immutable bufferLength16_ = bufferLength8_ / 2;
-
-        union
-        {
-            ///Buffer to hold UTF-8 data before decoding.
-            char[bufferLength8_ + 1] rawBuffer8_;
-            ///Buffer to hold UTF-16 data before decoding.
-            wchar[bufferLength16_ + 1] rawBuffer16_;
-        }
-        ///Number of elements held in the used raw buffer.
-        uint rawUsed_ = 0;
 
     public:
-        /**
+        /*
-         * Construct a Reader.
+         * Construct an AbstractReader.
          *
          * Params:  stream = Input stream. Must be readable and seekable.
          *
@@ -89,51 +74,14 @@ final class Reader
         body
         {
             stream_ = new EndianStream(stream);
-            available_ = stream_.available;
+            decoder_ = UTFFastDecoder(stream_);
-
-            //handle files short enough not to have a BOM
-            if(available_ < 2)
-            {
-                encoding_ = Encoding.UTF_8;
-                return;
-            }
-
-            //readBOM will determine and set stream endianness
-            switch(stream_.readBOM(2))
-            {
-                case -1: 
-                    //readBOM() eats two more bytes in this case so get them back
-                    const wchar bytes = stream_.getcw();
-                    rawBuffer8_[0] = cast(char)(bytes % 256);
-                    rawBuffer8_[1] = cast(char)(bytes / 256);
-                    rawUsed_ = 2;
-                    goto case 0;
-                case 0:  encoding_ = Encoding.UTF_8; break;
-                case 1, 2: 
-                    //readBOM() eats two more bytes in this case so get them back
-                    encoding_ = Encoding.UTF_16; 
-                    rawBuffer16_[0] = stream_.getcw();
-                    rawUsed_ = 1;
-                    enforce(available_ % 2 == 0, 
-                            new ReaderException("Odd byte count in an UTF-16 stream"));
-                    break;
-                case 3, 4: 
-                    enforce(available_ % 4 == 0, 
-                            new ReaderException("Byte count in an UTF-32 stream not divisible by 4"));
-                    encoding_ = Encoding.UTF_32;
-                    break;
-                default: assert(false, "Unknown UTF BOM");
-            }
-            available_ = stream_.available;
-
-            auto ptr = cast(dchar*)core.stdc.stdlib.malloc(dchar.sizeof * 256);
-            bufferAllocated_ = ptr[0 .. 256];
         }
 
-        ///Destroy the Reader.
         ~this()
         {
-            core.stdc.stdlib.free(bufferAllocated_.ptr);
+            //Delete the buffer, if allocated.
+            if(bufferAllocated_ is null){return;}
+            free(bufferAllocated_.ptr);
             buffer_ = bufferAllocated_ = null;
         }
 
@@ -148,12 +96,13 @@ final class Reader
          * Throws:  ReaderException if trying to read past the end of the stream
          *          or if invalid data is read.
          */
-        dchar peek(in size_t index = 0)
+        dchar peek(size_t index = 0)
         {
-            if(buffer_.length <= bufferOffset_ + index + 1)
+            if(buffer_.length < bufferOffset_ + index + 1)
             {
                 updateBuffer(index + 1);
             }
+
             if(buffer_.length <= bufferOffset_ + index)
             {
                 throw new ReaderException("Trying to read past the end of the stream");
@@ -172,7 +121,7 @@ final class Reader
          *
          * Returns: Characters starting at current position or an empty slice if out of bounds.
          */
-        const(dstring) prefix(in size_t length)
+        const(dstring) prefix(size_t length)
         {
             return slice(0, length);
         }
@@ -194,12 +143,12 @@ final class Reader
             {
                 updateBuffer(end);
             }
+
             end += bufferOffset_;
             start += bufferOffset_;
             end = min(buffer_.length, end);
-            if(end <= start){return "";}
 
-            return cast(dstring)buffer_[start .. end];
+            return end > start ? cast(dstring)buffer_[start .. end] : "";
         }
 
         /**
@@ -227,7 +176,7 @@ final class Reader
          * Throws:  ReaderException if trying to read past the end of the stream
          *          or if invalid data is read.
          */
-        dstring get(in size_t length)
+        dstring get(size_t length)
         {
             auto result = prefix(length).dup;
             forward(length);
@@ -244,13 +193,13 @@ final class Reader
          */
         void forward(size_t length = 1)
         {
-            mixin FastCharSearch!"\n\u0085\u2028\u2029"d search;
-
             if(buffer_.length <= bufferOffset_ + length + 1)
             {
                 updateBuffer(length + 1);
             }
 
+            mixin FastCharSearch!"\n\u0085\u2028\u2029"d search;
+
             while(length > 0)
             {
                 const c = buffer_[bufferOffset_];
@@ -268,19 +217,19 @@ final class Reader
         }
 
         ///Get a string describing current stream position, used for error messages.
-        @property Mark mark() const {return Mark(line_, column_);}
+        @property final Mark mark() const {return Mark(line_, column_);}
 
         ///Get current line number.
-        @property uint line() const {return line_;}
+        @property final uint line() const {return line_;}
 
-        ///Get current line number.
+        ///Get current column number.
-        @property uint column() const {return column_;}
+        @property final uint column() const {return column_;}
 
         ///Get index of the current character in the stream.
-        @property size_t charIndex() const {return charIndex_;}
+        @property final size_t charIndex() const {return charIndex_;}
 
         ///Get encoding of the input stream.
-        @property Encoding encoding() const {return encoding_;}
+        @property final Encoding encoding() const {return decoder_.encoding;}
 
     private:
         /**
@@ -296,7 +245,7 @@ final class Reader
          */
         void updateBuffer(in size_t length)
         {
-            //get rid of unneeded data in the buffer
+            //Get rid of unneeded data in the buffer.
             if(bufferOffset_ > 0)
             {
                 size_t bufferLength = buffer_.length - bufferOffset_;
@@ -306,12 +255,12 @@ final class Reader
                 bufferOffset_ = 0;
             }
 
-            ////Load chars in batches of at most 1024 bytes (256 chars)
+            //Load chars in batches of at most 1024 bytes (256 chars)
             while(buffer_.length <= bufferOffset_ + length)
             {
-                loadChars(256);
+                loadChars(512);
 
-                if(done)
+                if(decoder_.done)
                 {
                     if(buffer_.length == 0 || buffer_[$ - 1] != '\0')
                     {
@@ -325,9 +274,11 @@ final class Reader
         }
 
         /**
-         * Load at most specified number of characters.
+         * Load more characters to the buffer.
          *
-         * Params:  chars = Maximum number of characters to load.
+         * Params:  chars = Recommended number of characters to load. 
+         *                  More characters might be loaded.
+         *                  Less will be loaded if not enough available.
          *
          * Throws:  ReaderException on Unicode decoding error,
          *          if nonprintable characters are detected, or
@@ -335,96 +286,35 @@ final class Reader
          */
         void loadChars(size_t chars)
         {
-            ///Get next character from the stream.
-            dchar getDChar()
-            {
-                final switch(encoding_)
-                {
-                    case Encoding.UTF_8:
-                        //Temp buffer for moving data in rawBuffer8_.
-                        char[bufferLength8_] temp;
-                        //Shortcut for ASCII.
-                        if(rawUsed_ > 0 && rawBuffer8_[0] < 128)
-                        {
-                            //Get the first byte (one char in ASCII).
-                            const dchar result = rawBuffer8_[0];
-                            --rawUsed_;
-                            //Move the data.
-                            *(cast(ulong*)temp.ptr) = *(cast(ulong*)(rawBuffer8_.ptr + 1));
-                            *(cast(ulong*)rawBuffer8_.ptr) = *(cast(ulong*)temp.ptr);
-                            return result;
-                        }
-
-                        //Bytes to read.
-                        const readBytes = min(available_, bufferLength8_ - rawUsed_);
-                        available_ -= readBytes;
-                        //Length of data in rawBuffer8_ after reading.
-                        const len = rawUsed_ + readBytes;
-                        //Read the data.
-                        stream_.readExact(rawBuffer8_.ptr + rawUsed_, readBytes);
-
-                        //After decoding, this will point to the first byte not decoded.
-                        size_t idx = 0;
-                        const dchar result = decode(rawBuffer8_, idx);
-                        rawUsed_ = cast(uint)(len - idx);
-
-                        //Move the data.
-                        temp[0 .. rawUsed_] = rawBuffer8_[idx .. len];
-                        rawBuffer8_[0 .. rawUsed_] = temp[0 .. rawUsed_];
-                        return result;
-                    case Encoding.UTF_16: 
-                        //Temp buffer for moving data in rawBuffer8_.
-                        wchar[bufferLength16_] temp;
-                        //Words to read.
-                        size_t readWords = min(available_ / 2, bufferLength16_ - rawUsed_);
-                        available_ -= readWords * 2;
-                        //Length of data in rawBuffer16_ after reading.
-                        size_t len = rawUsed_;
-                        //Read the data.
-                        while(readWords > 0)
-                        {
-                            //Due to a bug in std.stream, we have to use getcw here.
-                            rawBuffer16_[len] = stream_.getcw(); 
-                            --readWords;
-                            ++len;
-                        }
-
-                        //After decoding, this will point to the first word not decoded.
-                        size_t idx = 0;
-                        const dchar result = decode(rawBuffer16_, idx);
-                        rawUsed_ = cast(uint)(len - idx);
-
-                        //Move the data.
-                        temp[0 .. rawUsed_] = rawBuffer16_[idx .. len];
-                        rawBuffer16_[0 .. rawUsed_] = temp[0 .. rawUsed_];
-                        return result;
-                    case Encoding.UTF_32:
-                        dchar result;
-                        available_ -= 4;
-                        stream_.read(result);
-                        return result;
-                }
-            }
-
             const oldLength = buffer_.length;
             const oldPosition = stream_.position;
 
-            //Preallocating memory to limit GC reallocations.
-
             bufferReserve(buffer_.length + chars);
             buffer_ = bufferAllocated_[0 .. buffer_.length + chars];
-            scope(exit)
+            scope(success)
             {
                 buffer_ = buffer_[0 .. $ - chars];
                 enforce(printable(buffer_[oldLength .. $]), 
                         new ReaderException("Special unicode characters are not allowed"));
             }
 
-            try for(uint c = 0; chars; --chars, ++c)
+            try for(size_t c = 0; chars && !decoder_.done;)
             {
-                if(done){break;}
+                const slice = decoder_.getDChars(chars);
-                buffer_[oldLength + c] = getDChar();
+                buffer_[oldLength + c .. oldLength + c + slice.length] = slice;
+                c += slice.length;
+                chars -= slice.length;
             }
+            catch(Exception e)
+            {
+                handleLoadCharsException(e, oldPosition);
+            }
+        }
+
+        //Handle an exception thrown in loadChars method of any Reader.
+        void handleLoadCharsException(Exception e, size_t oldPosition)
+        {
+            try{throw e;}
             catch(UtfException e)
             {
                 const position = stream_.position;
@@ -437,94 +327,376 @@ final class Reader
             }
         }
 
-        /**
+        //Code shared by loadEntireFile methods.
-         * Determine if all characters in an array are printable.
+        void loadEntireFile_()
-         *
-         * Params:  chars = Characters to check.
-         *
-         * Returns: True if all the characters are printable, false otherwise.
-         */
-        static bool printable(const ref dchar[] chars) pure
         {
-            foreach(c; chars)
+            const maxChars = decoder_.maxChars;
+            bufferReserve(maxChars + 1);
+            loadChars(maxChars);
+
+            if(buffer_.length == 0 || buffer_[$ - 1] != '\0')
             {
-                if(!((c == 0x09 || c == 0x0A || c == 0x0D || c == 0x85) ||
+                buffer_ = bufferAllocated_[0 .. buffer_.length + 1];
-                     (c >= 0x20 && c <= 0x7E) ||
+                buffer_[$ - 1] = '\0';
-                     (c >= 0xA0 && c <= '\uD7FF') ||
-                     (c >= '\uE000' && c <= '\uFFFD')))
-                {
-                    return false;
-                }
             }
-            return true;
         }
 
-        ///Are we done reading?
+        //Ensure there is space for at least capacity characters in bufferAllocated_.
-        @property bool done() const
-        {   
-            return (available_ == 0 && 
-                    ((encoding_ == Encoding.UTF_8  && rawUsed_ == 0) ||
-                     (encoding_ == Encoding.UTF_16 && rawUsed_ == 0) ||
-                     encoding_ == Encoding.UTF_32));
-        }
-
-        ///Ensure there is space for at least capacity characters in bufferAllocated_.
         void bufferReserve(in size_t capacity)
         {
-            if(bufferAllocated_.length >= capacity){return;}
+            if(bufferAllocated_ !is null && bufferAllocated_.length >= capacity){return;}
 
-            auto newPtr = core.stdc.stdlib.realloc(bufferAllocated_.ptr,  
+            //Handle first allocation as well as reallocation.
-                                                   capacity * dchar.sizeof);
+            auto ptr = bufferAllocated_ !is null 
-            bufferAllocated_ = (cast(dchar*)newPtr)[0 .. capacity];
+                       ? realloc(bufferAllocated_.ptr, capacity * dchar.sizeof)
+                       : malloc(capacity * dchar.sizeof);
+            bufferAllocated_ = (cast(dchar*)ptr)[0 .. capacity];
             buffer_ = bufferAllocated_[0 .. buffer_.length];
         }
+}
 
-    unittest
+private:
-    {
+
-        writeln("D:YAML reader endian unittest");
+alias UTFBlockDecoder!512 UTFFastDecoder;
-        void endian_test(ubyte[] data, Encoding encoding_expected, Endian endian_expected)
+
+///Decodes streams to UTF-32 in blocks.
+struct UTFBlockDecoder(size_t bufferSize_) if (bufferSize_ % 2 == 0)
+{
+    private:
+        //UTF-8 codepoint strides (0xFF are codepoints that can't start a sequence).
+        static immutable ubyte[256] utf8Stride =
+        [
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+            0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+            0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+            0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+            0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+            2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
+            2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
+            3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,
+            4,4,4,4,4,4,4,4,5,5,5,5,6,6,0xFF,0xFF,
+        ];
+
+        //Encoding of the input stream.
+        Encoding encoding_;
+        //Maximum number of characters that might be in the stream.
+        size_t maxChars_;
+        //Bytes available in the stream.
+        size_t available_;
+        //Input stream.
+        EndianStream stream_;
+
+        //Buffer used to store raw UTF-8 or UTF-16 code points.
+        union
         {
-            auto reader = new Reader(new MemoryStream(data));
+            char[bufferSize_] rawBuffer8_;
-            assert(reader.encoding_ == encoding_expected);
+            wchar[bufferSize_ / 2] rawBuffer16_;
-            assert(reader.stream_.endian == endian_expected);
         }
-        ubyte[] little_endian_utf_16 = [0xFF, 0xFE, 0x7A, 0x00];
+        //Used space (in items) in rawBuffer8_/rawBuffer16_.
-        ubyte[] big_endian_utf_16 = [0xFE, 0xFF, 0x00, 0x7A];
+        size_t rawUsed_;
-        endian_test(little_endian_utf_16, Encoding.UTF_16, Endian.littleEndian);
+
-        endian_test(big_endian_utf_16, Encoding.UTF_16, Endian.bigEndian);
+        //Space used by buffer_.
-    }
+        dchar[bufferSize_] bufferSpace_;
-    unittest
+        //Buffer of decoded, UTF-32 characters. This is a slice into bufferSpace_.
+        dchar[] buffer_;
+
+    public:
+        ///Construct a UTFFastDecoder decoding a stream.
+        this(EndianStream stream)
+        {
+            stream_ = stream;
+            available_ = stream_.available;
+
+            //Handle files short enough not to have a BOM.
+            if(available_ < 2)
+            {
+                encoding_ = Encoding.UTF_8;
+                maxChars_ = 0;
+
+                if(available_ == 1)
+                {
+                    bufferSpace_[0] = stream_.getc();
+                    buffer_ = bufferSpace_[0 .. 1];
+                    maxChars_ = 1;
+                }
+                return;
+            }
+
+            char[] rawBuffer8;
+            wchar[] rawBuffer16;
+            //readBOM will determine and set stream endianness.
+            switch(stream_.readBOM(2))
+            {
+                case -1: 
+                    //readBOM() eats two more bytes in this case so get them back.
+                    const wchar bytes = stream_.getcw();
+                    rawBuffer8_[0 .. 2] = [cast(ubyte)(bytes % 256), cast(ubyte)(bytes / 256)];
+                    rawUsed_ = 2;
+                    goto case 0;
+                case 0:  
+                    maxChars_ = available_;
+                    encoding_ = Encoding.UTF_8; 
+                    break;
+                case 1, 2: 
+                    maxChars_ = available_ / 2;
+                    //readBOM() eats two more bytes in this case so get them back.
+                    encoding_ = Encoding.UTF_16; 
+                    rawBuffer16_[0] = stream_.getcw();
+                    rawUsed_ = 1;
+                    enforce(available_ % 2 == 0, 
+                            new ReaderException("Odd byte count in an UTF-16 stream"));
+                    break;
+                case 3, 4: 
+                    maxChars_ = available_ / 4;
+                    encoding_ = Encoding.UTF_32;
+                    enforce(available_ % 4 == 0, 
+                            new ReaderException("Byte count in an UTF-32 stream not divisible by 4"));
+                    break;
+                default: assert(false, "Unknown UTF BOM");
+            }
+            available_ = stream_.available;
+        }
+
+        ///Get maximum number of characters that might be in the stream.
+        @property size_t maxChars() const {return maxChars_;}
+
+        ///Get encoding we're decoding from.
+        @property Encoding encoding() const {return encoding_;}
+
+        ///Are we done decoding?
+        @property bool done() const
+        {   
+            return rawUsed_ == 0 && buffer_.length == 0 && available_ == 0;
+        }
+
+        ///Get next character.
+        dchar getDChar()
+        {
+            if(buffer_.length)
+            {
+                const result = buffer_[0];
+                buffer_ = buffer_[1 .. $];
+                return result;
+            }
+
+            assert(available_ > 0 || rawUsed_ > 0);
+            updateBuffer();
+            return getDChar();
+        }
+
+        ///Get as many characters as possible, but at most maxChars. Slice returned will be invalidated in further calls.
+        const(dchar[]) getDChars(size_t maxChars = size_t.max)
+        {
+            if(buffer_.length)
+            {
+                const slice = min(buffer_.length, maxChars);
+                const result = buffer_[0 .. slice];
+                buffer_ = buffer_[slice .. $];
+                return result;
+            }
+
+            assert(available_ > 0 || rawUsed_ > 0);
+            updateBuffer();
+            return getDChars(maxChars);
+        }
+
+    private:
+        //Read and decode characters from file and store them in the buffer.
+        void updateBuffer()
+        {
+            assert(buffer_.length == 0);
+            final switch(encoding_)
+            {
+                case Encoding.UTF_8:
+                    const bytes = min(bufferSize_ - rawUsed_, available_);
+                    //Current length of valid data in rawBuffer8_.
+                    const rawLength = rawUsed_ + bytes;
+                    stream_.readExact(rawBuffer8_.ptr + rawUsed_, bytes);
+                    available_ -= bytes;
+                    decodeRawBuffer(rawBuffer8_, rawLength);
+                    break;
+
+                case Encoding.UTF_16:
+                    const words = min((bufferSize_ / 2) - rawUsed_, available_ / 2);
+                    //Current length of valid data in rawBuffer16_.
+                    const rawLength = rawUsed_ + words;
+                    foreach(c; rawUsed_ .. rawLength)
+                    {
+                        stream_.read(rawBuffer16_[c]);
+                        available_ -= 2;
+                    }
+                    decodeRawBuffer(rawBuffer16_, rawLength);
+                    break;
+
+                case Encoding.UTF_32:
+                    const chars = min(bufferSize_ / 4, available_ / 4);
+                    foreach(c; 0 .. chars)
+                    {
+                        stream_.read(bufferSpace_[c]);
+                        available_ -= 4;
+                    }
+                    buffer_ = bufferSpace_[0 .. chars];
+                    break;
+            }
+        }
+
+        //Decode contents of a UTF-8 or UTF-16 raw buffer.
+        void decodeRawBuffer(C)(C[] buffer, const size_t length)
+        {
+            //End of part of rawBuffer8_ that contains 
+            //complete characters and can be decoded.
+            const end = endOfLastUTFSequence(buffer, length);
+            //If end is 0, there are no full UTF-8 chars.
+            //This can happen at the end of file if there is an incomplete UTF-8 sequence.
+            enforce(end > 0,
+                    new ReaderException("Invalid UTF-8 character at the end of stream"));
+
+            decodeUTF(buffer[0 .. end]);
+
+            //After decoding, any code points not decoded go to the start of raw buffer.
+            rawUsed_ = length - end;
+            foreach(i; 0 .. rawUsed_){buffer[i] = buffer[i + end];}
+        }
+
+        //Determine the end of last UTF-8 or UTF-16 sequence in a raw buffer.
+        size_t endOfLastUTFSequence(C)(const C[] buffer, const size_t max)
+        {
+            static if(is(C == char))
+            {
+                for(long end = max - 1; end >= 0; --end)
+                {
+                    const s = utf8Stride[buffer[end]];
+                    if(s != 0xFF)
+                    {
+                        //If stride goes beyond end of the buffer (max), return end.
+                        //Otherwise the last sequence ends at max, so we can return that.
+                        //(Unless there is an invalid code point, which is 
+                        //caught at decoding)
+                        return (s > max - end) ?  cast(size_t)end : max;
+                    }
+                }
+                return 0;
+            }
+            else 
+            {
+                size_t end = 0;
+                while(end < max)
+                {
+                    const s = stride(buffer, end);
+                    if(s + end > max){break;}
+                    end += s;
+                }
+                return end;
+            }
+        }
+
+        //Decode a UTF-8 or UTF-16 buffer (with no incomplete sequences at the end).
+        void decodeUTF(C)(const C[] source)
+        {
+            size_t bufpos = 0;
+            const srclength = source.length;
+            for(size_t srcpos = 0; srcpos < srclength;)
+            {
+                const c = source[srcpos];
+                if(c < 0x80)
+                {
+                    bufferSpace_[bufpos++] = c;
+                    ++srcpos;
+                }
+                else
+                {
+                    bufferSpace_[bufpos++] = decode(source, srcpos);
+                }
+            }
+            buffer_ = bufferSpace_[0 .. bufpos]; 
+        }
+}
+
+/**
+ * Determine if all characters in an array are printable.
+ *
+ * Params:  chars = Characters to check.
+ *
+ * Returns: True if all the characters are printable, false otherwise.
+ */
+bool printable(const ref dchar[] chars) pure
+{
+    foreach(c; chars)
     {
-        writeln("D:YAML reader peek/prefix/forward unittest");
+        if(!((c == 0x09 || c == 0x0A || c == 0x0D || c == 0x85) ||
-        ubyte[] data = ByteOrderMarks[BOM.UTF8] ~ cast(ubyte[])"data";
+             (c >= 0x20 && c <= 0x7E) ||
-        auto reader = new Reader(new MemoryStream(data));
+             (c >= 0xA0 && c <= '\uD7FF') ||
+             (c >= '\uE000' && c <= '\uFFFD')))
+        {
+            return false;
+        }
+    }
+    return true;
+}
+
+//Unittests.
+
+void testEndian(R)()
+{
+    writeln(typeid(R).toString() ~ ": endian unittest");
+    void endian_test(ubyte[] data, Encoding encoding_expected, Endian endian_expected)
+    {
+        Reader reader = new R(new MemoryStream(data));
+        assert(reader.encoding == encoding_expected);
+        assert(reader.stream_.endian == endian_expected);
+    }
+    ubyte[] little_endian_utf_16 = [0xFF, 0xFE, 0x7A, 0x00];
+    ubyte[] big_endian_utf_16 = [0xFE, 0xFF, 0x00, 0x7A];
+    endian_test(little_endian_utf_16, Encoding.UTF_16, Endian.littleEndian);
+    endian_test(big_endian_utf_16, Encoding.UTF_16, Endian.bigEndian);
+}
+
+void testPeekPrefixForward(R)()
+{
+    writeln(typeid(R).toString() ~ ": peek/prefix/forward unittest");
+    ubyte[] data = ByteOrderMarks[BOM.UTF8] ~ cast(ubyte[])"data";
+    Reader reader = new R(new MemoryStream(data));
+    assert(reader.peek() == 'd');
+    assert(reader.peek(1) == 'a');
+    assert(reader.peek(2) == 't');
+    assert(reader.peek(3) == 'a');
+    assert(reader.peek(4) == '\0');
+    assert(reader.prefix(4) == "data");
+    assert(reader.prefix(6) == "data\0");
+    reader.forward(2);
+    assert(reader.peek(1) == 'a');
+    assert(collectException(reader.peek(3)));
+}
+
+void testUTF(R)()
+{
+    writeln(typeid(R).toString() ~ ": UTF formats unittest");
+    dchar[] data = cast(dchar[])"data";
+    void utf_test(T)(T[] data, BOM bom)
+    {
+        ubyte[] bytes = ByteOrderMarks[bom] ~ 
+                        (cast(ubyte*)data.ptr)[0 .. data.length * T.sizeof];
+        Reader reader = new R(new MemoryStream(bytes));
         assert(reader.peek() == 'd');
         assert(reader.peek(1) == 'a');
         assert(reader.peek(2) == 't');
         assert(reader.peek(3) == 'a');
-        assert(reader.peek(4) == '\0');
-        assert(reader.prefix(4) == "data");
-        assert(reader.prefix(6) == "data\0");
-        reader.forward(2);
-        assert(reader.peek(1) == 'a');
-        assert(collectException(reader.peek(3)));
-    }
-    unittest
-    {
-        writeln("D:YAML reader UTF formats unittest");
-        dchar[] data = cast(dchar[])"data";
-        void utf_test(T)(T[] data, BOM bom)
-        {
-            ubyte[] bytes = ByteOrderMarks[bom] ~ 
-                            (cast(ubyte*)data.ptr)[0 .. data.length * T.sizeof];
-            auto reader = new Reader(new MemoryStream(bytes));
-            assert(reader.peek() == 'd');
-            assert(reader.peek(1) == 'a');
-            assert(reader.peek(2) == 't');
-            assert(reader.peek(3) == 'a');
-        }
-        utf_test!char(to!(char[])(data), BOM.UTF8);
-        utf_test!wchar(to!(wchar[])(data), endian == Endian.bigEndian ? BOM.UTF16BE : BOM.UTF16LE);
-        utf_test(data, endian == Endian.bigEndian ? BOM.UTF32BE : BOM.UTF32LE);
     }
+    utf_test!char(to!(char[])(data), BOM.UTF8);
+    utf_test!wchar(to!(wchar[])(data), endian == Endian.bigEndian ? BOM.UTF16BE : BOM.UTF16LE);
+    utf_test(data, endian == Endian.bigEndian ? BOM.UTF32BE : BOM.UTF32LE);
+}
+
+unittest
+{
+    testEndian!Reader();
+    testPeekPrefixForward!Reader();
+    testUTF!Reader();
 }

dyaml/representer.d

@@ -65,7 +65,7 @@ final class Representer
          *                                   disabled to use custom representer
          *                                   functions for default types.
          */
-        this(in bool useDefaultRepresenters = true)
+        this(bool useDefaultRepresenters = true)
         {
             if(!useDefaultRepresenters){return;}
             addRepresenter!YAMLNull(&representNull);
@@ -87,13 +87,13 @@ final class Representer
         }
 
         ///Set default _style for scalars. Invalid means the _style is chosen automatically.
-        @property void defaultScalarStyle(in ScalarStyle style)
+        @property void defaultScalarStyle(ScalarStyle style)
         {
             defaultScalarStyle_ = style;
         }
 
         ///Set default _style for collections. Invalid means the _style is chosen automatically. 
-        @property void defaultCollectionStyle(in CollectionStyle style)
+        @property void defaultCollectionStyle(CollectionStyle style)
         {
             defaultCollectionStyle_ = style;
         }
@@ -237,7 +237,7 @@ final class Representer
          * }
          * --------------------
          */
-        Node representScalar(in string tag, string scalar, 
+        Node representScalar(string tag, string scalar, 
                              ScalarStyle style = ScalarStyle.Invalid)
         {
             if(style == ScalarStyle.Invalid){style = defaultScalarStyle_;}
@@ -276,7 +276,7 @@ final class Representer
          * }
          * --------------------
          */
-        Node representSequence(in string tag, Node[] sequence, 
+        Node representSequence(string tag, Node[] sequence, 
                                CollectionStyle style = CollectionStyle.Invalid)
         {
             Node[] value;
@@ -335,7 +335,7 @@ final class Representer
          * }
          * --------------------
          */
-        Node representMapping(in string tag, Node.Pair[] pairs,
+        Node representMapping(string tag, Node.Pair[] pairs,
                               CollectionStyle style = CollectionStyle.Invalid)
         {
             Node.Pair[] value;

dyaml/serializer.d

@@ -23,7 +23,7 @@ import dyaml.exception;
 import dyaml.node;
 import dyaml.resolver;
 import dyaml.tag;
-import dyaml.tagdirectives;
+import dyaml.tagdirective;
 import dyaml.token;
 
 
@@ -46,7 +46,7 @@ struct Serializer
         string YAMLVersion_;
 
         ///Tag directives to emit.
-        TagDirectives tagDirectives_;
+        TagDirective[] tagDirectives_;
 
         //TODO Use something with more deterministic memory usage.
         ///Nodes with assigned anchors.
@@ -70,7 +70,7 @@ struct Serializer
          */
         this(ref Emitter emitter, Resolver resolver, Encoding encoding,
              in bool explicitStart, in bool explicitEnd, string YAMLVersion, 
-             TagDirectives tagDirectives)
+             TagDirective[] tagDirectives)
         {
             emitter_ = &emitter;
             resolver_ = resolver;

dyaml/sharedobject.d

@@ -1,131 +0,0 @@
-
-//          Copyright Ferdinand Majerech 2011.
-// Distributed under the Boost Software License, Version 1.0.
-//    (See accompanying file LICENSE_1_0.txt or copy at
-//          http://www.boost.org/LICENSE_1_0.txt)
-
-///Shared object.
-module dyaml.sharedobject;
-
-
-/**
- * Mixin for shared objects (need a better name). 
- *
- * This works as an index to a static array of type T. Any new object created is 
- * checked for presence in the array to prevent duplication.
- *
- * This is useful for e.g. token/event data that rarely needs to be
- * stored (e.g. tag directives) to prevent inflation of these structs,
- * and when there are many instances of a data type that are mostly
- * duplicates (e.g. tags). 
- *
- * This is not the most elegant way to store the extra data and might change in future.
- */
-template SharedObject(T, MixedIn)
-{
-    private:
-        ///This class stores the data that is shared between the objects.
-        class SharedData
-        {
-            private:
-                /**
-                 * Reference count.
-                 *
-                 * When this reaches zero, objects_ are cleared. This is not
-                 * the number of shared objects, but rather of objects using this kind 
-                 * of shared object. 
-                 */
-                int referenceCount_ = 0;  
-
-                ///All known objects of type T are in this array.
-                T[] objects_;
-
-            public:
-                ///Increment the reference count.
-                void addReference()
-                {
-                    assert(referenceCount_ >= 0);
-                    ++referenceCount_;
-                }
-
-                ///Decrement the reference count and clear the constructed objects if zero.
-                void removeReference()
-                {
-                    --referenceCount_;
-                    assert(referenceCount_ >= 0);
-                    if(referenceCount_ == 0)
-                    {
-                        clear(objects_);
-                        objects_ = [];
-                    }
-                }
-
-                ///Add an object and return its index.
-                uint add(ref T object)
-                {
-                    foreach(index, ref known; objects_) if(object == known)
-                    {
-                        return cast(uint)index;
-                    }
-                    objects_ ~= object;
-                    return cast(uint)objects_.length - 1;
-                }
-
-                ///Get the object at specified object.
-                @property T get(in uint index) 
-                {
-                    return objects_[index];
-                }
-        }
-
-        ///Index of the object in data_.
-        uint index_ = uint.max;
-
-        ///Stores the actual objects.
-        static __gshared SharedData data_;
-
-        static this()
-        {
-            data_ = new SharedData;
-        }
-
-    public:
-        ///Increment the reference count.
-        static void addReference()
-        {
-            synchronized(data_){data_.addReference();}
-        }
-
-        ///Decrement the reference count and clear the constructed objects if zero.
-        static void removeReference()
-        {
-            synchronized(data_){data_.removeReference();}
-        }
-
-        ///Get the object.
-        @property T get() const
-        in{assert(!isNull());}
-        body
-        {
-            T result;
-            synchronized(data_){result = data_.get(index_);}
-            return result;
-        }
-
-        ///Test for equality with another object.
-        bool opEquals(const ref MixedIn object) const
-        {
-            return object.index_ == index_;
-        }
-
-        ///Is this object null (invalid)?
-        @property bool isNull() const {return index_ == uint.max;}
-
-    private:
-        ///Add a new object, checking if identical object already exists.
-        void add(ref T object)
-        {
-            synchronized(data_){index_ = data_.add(object);}
-        }
-}
-

dyaml/tag.d

@@ -7,66 +7,7 @@
 ///YAML tag.
 module dyaml.tag;
 
+import dyaml.zerostring;
 
-import core.stdc.string;
+///YAML tag (data type) struct. Encapsulates a tag to save memory and speed up comparison.
-
+alias ZeroString!"Tag" Tag;
-
-///YAML tag (data type) struct. Encapsulates a tag to save memory and speed-up comparison.
-struct Tag
-{
-    private:
-        ///Zero terminated tag string.
-        immutable(char)* tag_ = null;
-
-    public:
-        @disable int opCmp(ref Tag);
-
-
-        ///Construct a tag from a string representation.
-        this(in string tag)
-        {
-            if(tag is null || tag == "")
-            {
-                tag_ = null;
-                return;
-            }
-
-            tag_ = (tag ~ '\0').ptr;
-        }
-
-        ///Get the tag string.
-        @property string get() const
-        in{assert(!isNull());}
-        body
-        {
-            return cast(string)tag_[0 .. strlen(tag_)];
-        }
-
-        ///Test for equality with another tag.
-        bool opEquals(const ref Tag tag) const
-        {
-            return isNull ? tag.isNull : 
-                   tag.isNull ? false : (0 == strcmp(tag_, tag.tag_));
-        }
-
-        ///Compute a hash.
-        hash_t toHash() const
-        in{assert(!isNull);}
-        body
-        {
-            static type = typeid(string);
-            auto str = get();
-            return type.getHash(&str);
-        }
-
-        ///Compare with another tag.
-        int opCmp(const ref Tag tag) const 
-        in{assert(!isNull && !tag.isNull);}
-        body
-        {
-            return strcmp(tag_, tag.tag_);
-        }
-
-        ///Is this tag null (invalid)?
-        @property bool isNull() const {return tag_ is null;}
-}

dyaml/tagdirective.d

@@ -0,0 +1,15 @@
+
+//          Copyright Ferdinand Majerech 2011.
+// Distributed under the Boost Software License, Version 1.0.
+//    (See accompanying file LICENSE_1_0.txt or copy at
+//          http://www.boost.org/LICENSE_1_0.txt)
+
+///Tag directives.
+module dyaml.tagdirective;
+
+///Single tag directive. handle is the shortcut, prefix is the prefix that replaces it.
+struct TagDirective
+{
+    string handle;
+    string prefix;
+}

dyaml/tagdirectives.d

@@ -1,28 +0,0 @@
-
-//          Copyright Ferdinand Majerech 2011.
-// Distributed under the Boost Software License, Version 1.0.
-//    (See accompanying file LICENSE_1_0.txt or copy at
-//          http://www.boost.org/LICENSE_1_0.txt)
-
-///Tag directives.
-module dyaml.tagdirectives;
-
-import std.typecons;
-
-import dyaml.sharedobject;
-
-///Single tag directive. handle is the shortcut, prefix is the prefix that replaces it.
-alias Tuple!(string, "handle", string, "prefix") tagDirective;
-
-///Tag directives stored in Event.
-struct TagDirectives
-{
-    public:
-        mixin SharedObject!(tagDirective[], TagDirectives);
-
-        ///Construct a tags object from an array of tag directives.
-        this(tagDirective[] tagDirectives)
-        {
-            add(tagDirectives);
-        }
-}

dyaml/zerostring.d

@@ -0,0 +1,73 @@
+
+//          Copyright Ferdinand Majerech 2011.
+// Distributed under the Boost Software License, Version 1.0.
+//    (See accompanying file LICENSE_1_0.txt or copy at
+//          http://www.boost.org/LICENSE_1_0.txt)
+
+///Zero terminated string.
+module dyaml.zerostring;
+
+import core.stdc.string;
+
+/**
+ * Zero terminated string used to decrease data structure size. 
+ *
+ * TypeName is used to differentiate types (better than simple alias).
+ */
+struct ZeroString(string TypeName)
+{
+    private:
+        ///Zero terminated string.
+        immutable(char)* str_ = null;
+
+    public:
+        @disable int opCmp(ref ZeroString);
+
+        ///Construct a string.
+        this(in string str)
+        {
+            if(str is null || str == "")
+            {
+                str_ = null;
+                return;
+            }
+
+            str_ = (str ~ '\0').ptr;
+        }
+
+        ///Get the string.
+        @property string get() const
+        in{assert(!isNull());}
+        body
+        {
+            return cast(string)str_[0 .. strlen(str_)];
+        }
+
+        ///Test for equality with another string.
+        bool opEquals(const ref ZeroString str) const
+        {
+            return isNull ? str.isNull : 
+                   str.isNull ? false : (0 == strcmp(str_, str.str_));
+        }
+
+        ///Compute a hash.
+        hash_t toHash() const
+        in{assert(!isNull);}
+        body
+        {
+            static type = typeid(string);
+            auto str = get();
+            return type.getHash(&str);
+        }
+
+        ///Compare with another string.
+        int opCmp(const ref ZeroString str) const 
+        in{assert(!isNull && !str.isNull);}
+        body
+        {
+            return strcmp(str_, str.str_);
+        }
+
+        ///Is this string null (invalid)?
+        @property bool isNull() const {return str_ is null;}
+}

test/src/compare.d

@@ -42,8 +42,8 @@ void testParser(bool verbose, string dataFilename, string canonicalFilename)
  */
 void testLoader(bool verbose, string dataFilename, string canonicalFilename)
 {
-    auto data = Loader(dataFilename).loadAll;
+    auto data = Loader(dataFilename).loadAll();
-    auto canonical = Loader(canonicalFilename).loadAll;
+    auto canonical = Loader(canonicalFilename).loadAll();
 
     assert(data.length == canonical.length, "Unequal node count");
     foreach(n; 0 .. data.length)
@@ -58,7 +58,7 @@ void testLoader(bool verbose, string dataFilename, string canonicalFilename)
                 writeln("Canonical value:");
                 writeln(canonical[n].debugString);
             }
-            assert(false);
+            assert(false, "testLoader(" ~ dataFilename ~ ", " ~ canonicalFilename ~ ") failed");
         }
     }
 }

test/src/emitter.d

@@ -83,7 +83,7 @@ void testEmitterOnData(bool verbose, string dataFilename, string canonicalFilena
 {
     //Must exist due to Anchor, Tags reference counts.
     auto loader = Loader(dataFilename);
-    auto events = loader.parse();
+    auto events = cast(Event[])loader.parse();
     auto emitStream = new MemoryStream;
     Dumper(emitStream).emit(events);
 
@@ -98,7 +98,7 @@ void testEmitterOnData(bool verbose, string dataFilename, string canonicalFilena
     loader2.name        = "TEST";
     loader2.constructor = new Constructor;
     loader2.resolver    = new Resolver;
-    auto newEvents = loader2.parse();
+    auto newEvents = cast(Event[])loader2.parse();
     assert(compareEvents(events, newEvents));
 }
 
@@ -114,7 +114,7 @@ void testEmitterOnCanonical(bool verbose, string canonicalFilename)
 {
     //Must exist due to Anchor, Tags reference counts.
     auto loader = Loader(canonicalFilename);
-    auto events = loader.parse();
+    auto events = cast(Event[])loader.parse();
     foreach(canonical; [false, true])
     {
         auto emitStream = new MemoryStream;
@@ -130,7 +130,7 @@ void testEmitterOnCanonical(bool verbose, string canonicalFilename)
         loader2.name        = "TEST";
         loader2.constructor = new Constructor;
         loader2.resolver    = new Resolver;
-        auto newEvents = loader2.parse();
+        auto newEvents = cast(Event[])loader2.parse();
         assert(compareEvents(events, newEvents));
     }
 }
@@ -151,7 +151,7 @@ void testEmitterStyles(bool verbose, string dataFilename, string canonicalFilena
     {
         //must exist due to Anchor, Tags reference counts
         auto loader = Loader(canonicalFilename);
-        auto events = loader.parse();
+        auto events = cast(Event[])loader.parse();
         foreach(flowStyle; [CollectionStyle.Block, CollectionStyle.Flow])
         {
             foreach(style; [ScalarStyle.Literal, ScalarStyle.Folded, 
@@ -191,7 +191,7 @@ void testEmitterStyles(bool verbose, string dataFilename, string canonicalFilena
                 loader2.name        = "TEST";
                 loader2.constructor = new Constructor;
                 loader2.resolver    = new Resolver;
-                auto newEvents = loader2.parse();
+                auto newEvents = cast(Event[])loader2.parse();
                 assert(compareEvents(events, newEvents));
             }
         }

test/src/errors.d

@@ -72,7 +72,8 @@ void testLoaderErrorFilename(bool verbose, string errorFilename)
         if(verbose){writeln(typeid(e).toString(), "\n", e);}
         return;
     }
-    assert(false, "Expected an exception");
+    assert(false, "testLoaderErrorSingle(" ~ to!string(verbose) ~ 
+                  ", " ~ errorFilename ~ ") Expected an exception");
 }
 
 /**