Perf: removed redundant lowercase normalization

Author: jhy

src/main/java/org/jsoup/parser/HtmlTreeBuilder.java

@@ -111,7 +111,7 @@ protected void initialiseParse(Reader input, String baseUri, Parser parser) {
 
         if (context != null) {
             final String contextName = context.normalName();
-            contextElement = new Element(tagFor(contextName, settings), baseUri);
+            contextElement = new Element(tagFor(contextName, contextName, defaultNamespace(), settings), baseUri);
             if (context.ownerDocument() != null) // quirks setup:
                 doc.quirksMode(context.ownerDocument().quirksMode());
 
@@ -318,7 +318,7 @@ Element createElementFor(Token.StartTag startTag, String namespace, boolean forc
             }
         }
 
-        Tag tag = tagFor(startTag.tagName, namespace,
+        Tag tag = tagFor(startTag.tagName, startTag.normalName, namespace,
             forcePreserveCase ? ParseSettings.preserveCase : settings);
 
         return (tag.normalName().equals("form")) ?
@@ -948,7 +948,7 @@ void reconstructFormattingElements() {
 
             // 8. create new element from element, 9 insert into current node, onto stack
             skip = false; // can only skip increment from 4.
-            Element newEl = new Element(tagFor(entry.normalName(), settings), null, entry.attributes().clone());
+            Element newEl = new Element(tagFor(entry.nodeName(), entry.normalName(), defaultNamespace(), settings), null, entry.attributes().clone());
             doInsertElement(newEl, null);
 
             // 10. replace entry with new entry

src/main/java/org/jsoup/parser/HtmlTreeBuilderState.java

@@ -950,7 +950,7 @@ private boolean inBodyEndTagAdoption(Token t, HtmlTreeBuilder tb) {
                     }
 
                     //  6. [Create an element for the token] for which the element node was created, in the [HTML namespace], with commonAncestor as the intended parent; replace the entry for node in the [list of active formatting elements] with an entry for the new element, replace the entry for node in the [stack of open elements] with an entry for the new element, and let node be the new element.
-                    Element replacement = new Element(tb.tagFor(node.nodeName(), ParseSettings.preserveCase), tb.getBaseUri());
+                    Element replacement = new Element(tb.tagFor(node.nodeName(), node.normalName(), tb.defaultNamespace(), ParseSettings.preserveCase), tb.getBaseUri());
                     tb.replaceActiveFormattingElement(node, replacement);
                     tb.replaceOnStack(node, replacement);
                     node = replacement;

src/main/java/org/jsoup/parser/ParseSettings.java

@@ -4,6 +4,7 @@
 import org.jspecify.annotations.Nullable;
 
 import static org.jsoup.internal.Normalizer.lowerCase;
+import static org.jsoup.internal.Normalizer.normalize;
 
 /**
  * Controls parser case settings, to optionally preserve tag and/or attribute name case.
@@ -83,6 +84,6 @@ public String normalizeAttribute(String name) {
 
     /** Returns the normal name that a Tag will have (trimmed and lower-cased) */
     static String normalName(String name) {
-        return lowerCase(name.trim());
+        return normalize(name);
     }
 }

src/main/java/org/jsoup/parser/Tag.java

@@ -28,9 +28,9 @@ public class Tag implements Cloneable {
     private boolean formList = false; // a control that appears in forms: input, textarea, output etc
     private boolean formSubmit = false; // a control that can be submitted in a form: input etc
 
-    private Tag(String tagName, String namespace) {
+    private Tag(String tagName, String normalName, String namespace) {
         this.tagName = tagName;
-        normalName = Normalizer.lowerCase(tagName);
+        this.normalName = normalName;
         this.namespace = namespace;
     }
 
@@ -67,15 +67,20 @@ public String namespace() {
      * @return The tag, either defined or new generic.
      */
     public static Tag valueOf(String tagName, String namespace, ParseSettings settings) {
+        return valueOf(tagName, ParseSettings.normalName(tagName), namespace, settings);
+    }
+
+    /** Tag.valueOf with the normalName via the token.normalName, to save redundant lower-casing passes. */
+    static Tag valueOf(String tagName, String normalName, String namespace, ParseSettings settings) {
+        Validate.notNull(tagName);
+        tagName = tagName.trim();
         Validate.notEmpty(tagName);
         Validate.notNull(namespace);
         Tag tag = Tags.get(tagName);
         if (tag != null && tag.namespace.equals(namespace))
             return tag;
 
-        tagName = settings.normalizeTag(tagName); // the name we'll use
-        Validate.notEmpty(tagName);
-        String normalName = Normalizer.lowerCase(tagName); // the lower-case name to get tag settings off
+        tagName = settings.preserveTagCase() ? tagName : normalName;
         tag = Tags.get(normalName);
         if (tag != null && tag.namespace.equals(namespace)) {
             if (settings.preserveTagCase() && !tagName.equals(normalName)) {
@@ -86,12 +91,13 @@ public static Tag valueOf(String tagName, String namespace, ParseSettings settin
         }
 
         // not defined: create default; go anywhere, do anything! (incl be inside a <p>)
-        tag = new Tag(tagName, namespace);
+        tag = new Tag(tagName, normalName, namespace);
         tag.isBlock = false;
 
         return tag;
     }
 
+
     /**
      * Get a Tag by name. If not previously defined (unknown), returns a new generic tag, that can do anything.
      * <p>
@@ -304,7 +310,7 @@ private static void setupTags(String[] tagNames, Consumer<Tag> tagModifier) {
         for (String tagName : tagNames) {
             Tag tag = Tags.get(tagName);
             if (tag == null) {
-                tag = new Tag(tagName, Parser.NamespaceHtml);
+                tag = new Tag(tagName, tagName, Parser.NamespaceHtml);
                 Tags.put(tag.tagName, tag);
             }
             tagModifier.accept(tag);

src/main/java/org/jsoup/parser/TreeBuilder.java

@@ -238,21 +238,17 @@ boolean isContentForTagData(String normalName) {
         return false;
     }
 
-    Tag tagFor(String tagName, String namespace, ParseSettings settings) {
+    Tag tagFor(String tagName, String normalName, String namespace, ParseSettings settings) {
         Tag cached = seenTags.get(tagName); // note that we don't normalize the cache key. But tag via valueOf may be normalized.
         if (cached == null || !cached.namespace().equals(namespace)) {
             // only return from cache if the namespace is the same. not running nested cache to save double hit on the common flow
-            Tag tag = Tag.valueOf(tagName, namespace, settings);
+            Tag tag = Tag.valueOf(tagName, normalName, namespace, settings);
             seenTags.put(tagName, tag);
             return tag;
         }
         return cached;
     }
 
-    Tag tagFor(String tagName, ParseSettings settings) {
-        return tagFor(tagName, defaultNamespace(), settings);
-    }
-
     /**
      Gets the default namespace for this TreeBuilder
      * @return the default namespace

src/main/java/org/jsoup/parser/XmlTreeBuilder.java

@@ -93,7 +93,7 @@ protected boolean process(Token token) {
     }
 
     void insertElementFor(Token.StartTag startTag) {
-        Tag tag = tagFor(startTag.name(), settings);
+        Tag tag = tagFor(startTag.name(), startTag.normalName(), defaultNamespace(), settings);
         if (startTag.attributes != null)
             startTag.attributes.deduplicate(settings);