Regex: Use the native 'd' flag to compute indices when available.

sjrd · sjrd · commit 5f4ce145ae69 · 2021-05-22T19:11:41.000+02:00
That feature is enabled by a flag, and not by default, because of
performance concerns. JS engine implementers have shown that
enabling the feature has a significant performance impact.
Therefore, we only enable it the first time group indices are
required for any given `Pattern` instance. This means that the
first match for which group indices are required will be run twice,
but this is probably the better trade-off.
diff --git a/javalib/src/main/scala/java/util/regex/Pattern.scala b/javalib/src/main/scala/java/util/regex/Pattern.scala
@@ -34,6 +34,9 @@ final class Pattern private[regex] (
   @inline private def jsFlagsForFind: String =
     jsFlags + (if (sticky && supportsSticky) "gy" else "g")
 
+  /** Whether we already added the 'd' flag to the native RegExp's. */
+  private var enabledNativeIndices: Boolean = false
+
   /** The RegExp that is used for `Matcher.find()`.
    *
    *  It receives the 'g' flag so that `lastIndex` is taken into acount.
@@ -45,7 +48,7 @@ final class Pattern private[regex] (
    *  Since that RegExp is only used locally within `execFind()`, we can
    *  always reuse the same instance.
    */
-  private[this] val jsRegExpForFind =
+  private[this] var jsRegExpForFind =
     new js.RegExp(jsPattern, jsFlagsForFind)
 
   /** Another version of the RegExp that is used by `Matcher.matches()`.
@@ -57,7 +60,7 @@ final class Pattern private[regex] (
    *  Since that RegExp is only used locally within `execMatches()`, we can
    *  always reuse the same instance.
    */
-  private[this] val jsRegExpForMatches: js.RegExp =
+  private[this] var jsRegExpForMatches: js.RegExp =
     new js.RegExp("^" + jsPattern + "$", jsFlags)
 
   private lazy val indicesBuilder: IndicesBuilder =
@@ -131,8 +134,20 @@ final class Pattern private[regex] (
 
   private[regex] def getIndices(lastMatch: js.RegExp.ExecResult, forMatches: Boolean): IndicesArray = {
     val lastMatchDyn = lastMatch.asInstanceOf[js.Dynamic]
-    if (js.isUndefined(lastMatchDyn.indices))
-      lastMatchDyn.indices = indicesBuilder(forMatches, lastMatch.input, lastMatch.index)
+    if (js.isUndefined(lastMatchDyn.indices)) {
+      if (supportsIndices) {
+        if (!enabledNativeIndices) {
+          jsRegExpForFind = new js.RegExp(jsPattern, jsFlagsForFind + "d")
+          jsRegExpForMatches = new js.RegExp("^" + jsPattern + "$", jsFlags + "d")
+          enabledNativeIndices = true
+        }
+        val regexp = if (forMatches) jsRegExpForMatches else jsRegExpForFind
+        regexp.lastIndex = lastMatch.index
+        lastMatchDyn.indices = regexp.exec(lastMatch.input).asInstanceOf[js.Dynamic].indices
+      } else {
+        lastMatchDyn.indices = indicesBuilder(forMatches, lastMatch.input, lastMatch.index)
+      }
+    }
     lastMatchDyn.indices.asInstanceOf[IndicesArray]
   }
 
diff --git a/javalib/src/main/scala/java/util/regex/PatternCompiler.scala b/javalib/src/main/scala/java/util/regex/PatternCompiler.scala
@@ -90,6 +90,10 @@ private[regex] object PatternCompiler {
   private val _supportsDotAll =
     (esVersion >= ESVersion.ES2018) || featureTest("", "us")
 
+  /** Cache for `Support.supportsIndices`. */
+  private val _supportsIndices =
+    featureTest("a(b)", "d")
+
   /** Feature-test methods.
    *
    *  They are located in a separate object so that the methods can be fully
@@ -112,6 +116,11 @@ private[regex] object PatternCompiler {
     def supportsDotAll: Boolean =
       (esVersion >= ESVersion.ES2018) || _supportsDotAll
 
+    /** Tests whether the underlying JS RegExp supports the 'd' flag. */
+    @inline
+    def supportsIndices: Boolean =
+      _supportsIndices
+
     /** Tests whether features requiring support for the 'u' flag are enabled.
      *
      *  They are enabled if and only if the project is configured to rely on
