Merge pull request #82851 from glessard/rdar154331399-makeReallyContiguous

[stdlib] Make String.makeContiguousUTF8() strictly true

Author: glessard

stdlib/public/core/SmallString.swift

@@ -79,7 +79,7 @@ internal struct _SmallString {
 extension _SmallString {
   @inlinable @inline(__always)
   internal static var capacity: Int {
-#if _pointerBitWidth(_32) && os(watchOS)
+#if os(watchOS) && _pointerBitWidth(_32)
     return 10
 #elseif _pointerBitWidth(_32) || _pointerBitWidth(_16)
     // Note: changed from 10 for contiguous storage.
@@ -95,7 +95,7 @@ extension _SmallString {
 
   @_alwaysEmitIntoClient @inline(__always)
   internal static func contiguousCapacity() -> Int {
-#if _pointerBitWidth(_32) && os(watchOS)
+#if os(watchOS) && _pointerBitWidth(_32)
     return capacity &- 2
 #else
     return capacity

stdlib/public/core/StringCreate.swift

@@ -312,9 +312,19 @@ extension String {
   @inline(never) // slow-path
   internal static func _copying(_ str: Substring) -> String {
     if _fastPath(str._wholeGuts.isFastUTF8) {
-      return unsafe str._wholeGuts.withFastUTF8(range: str._offsetRange) {
+      var new = unsafe str._wholeGuts.withFastUTF8(range: str._offsetRange) {
         unsafe String._uncheckedFromUTF8($0)
       }
+#if os(watchOS) && _pointerBitWidth(_32)
+      // Required for compatibility with some small strings that
+      // may be encoded in the 32-bit slice of watchOS binaries.
+      if str._wholeGuts.isSmall,
+         str._wholeGuts.count > _SmallString.contiguousCapacity() {
+        new.reserveCapacity(_SmallString.capacity + 1)
+        return new
+      }
+#endif
+      return new
     }
     return unsafe Array(str.utf8).withUnsafeBufferPointer {
       unsafe String._uncheckedFromUTF8($0)

stdlib/public/core/StringProtocol.swift

@@ -177,15 +177,27 @@ extension StringProtocol {
 
 // Contiguous UTF-8 strings
 extension String {
-  /// Returns whether this string is capable of providing access to
-  /// validly-encoded UTF-8 contents in contiguous memory in O(1) time.
+  /// Returns whether this string's storage contains
+  /// validly-encoded UTF-8 contents in contiguous memory.
   ///
-  /// Contiguous strings always operate in O(1) time for withUTF8 and always
-  /// give a result for String.UTF8View.withContiguousStorageIfAvailable.
+  /// Contiguous strings always operate in O(1) time for withUTF8, always give
+  /// a result for String.UTF8View.withContiguousStorageIfAvailable, and always
+  /// return a non-nil value from `String._utf8Span` and `String.UTF8View._span`.
   /// Contiguous strings also benefit from fast-paths and better optimizations.
-  ///
   @_alwaysEmitIntoClient
-  public var isContiguousUTF8: Bool { return _guts.isFastUTF8 }
+  public var isContiguousUTF8: Bool {
+    if _guts.isFastUTF8 {
+#if os(watchOS) && _pointerBitWidth(_32)
+      // Required for compatibility with some small strings that
+      // may be encoded in the 32-bit slice of watchOS binaries.
+      if _guts.isSmall && _guts.count > _SmallString.contiguousCapacity() {
+        return false
+      }
+#endif
+      return true
+    }
+    return false
+  }
 
   /// If this string is not contiguous, make it so. If this mutates the string,
   /// it will invalidate any pre-existing indices.
@@ -223,13 +235,14 @@ extension String {
 
 // Contiguous UTF-8 strings
 extension Substring {
-  /// Returns whether this string is capable of providing access to
-  /// validly-encoded UTF-8 contents in contiguous memory in O(1) time.
+  /// Returns whether this string's storage contains
+  /// validly-encoded UTF-8 contents in contiguous memory.
   ///
-  /// Contiguous strings always operate in O(1) time for withUTF8 and always
-  /// give a result for String.UTF8View.withContiguousStorageIfAvailable.
+  /// Contiguous strings always operate in O(1) time for withUTF8, always give
+  /// a result for Substring.UTF8View.withContiguousStorageIfAvailable, and
+  /// always return a non-nil value from `Substring._utf8Span` and
+  /// `Substring.UTF8View._span`.
   /// Contiguous strings also benefit from fast-paths and better optimizations.
-  ///
   @_alwaysEmitIntoClient
   public var isContiguousUTF8: Bool { return self.base.isContiguousUTF8 }
 

test/stdlib/Span/SmallStringCompatibility.swift

@@ -0,0 +1,67 @@
+//===--- SmallStringCompatibility.swift -----------------------------------===//
+//
+// This source file is part of the Swift.org open source project
+//
+// Copyright (c) 2025 Apple Inc. and the Swift project authors
+// Licensed under Apache License v2.0 with Runtime Library Exception
+//
+// See https://swift.org/LICENSE.txt for license information
+// See https://swift.org/CONTRIBUTORS.txt for the list of Swift project authors
+//
+//===----------------------------------------------------------------------===//
+
+// RUN: %target-run-stdlib-swift
+
+// REQUIRES: executable_test
+// REQUIRES: CPU=wasm32 || CPU=arm64_32
+
+import StdlibUnittest
+
+var suite = TestSuite("SmallStringCompatibility")
+defer { runAllTests() }
+
+var strings = [
+  ("Small", true),
+  ("Less small", true),
+  ("Positively large.", true),
+]
+
+#if os(watchOS)
+strings[1].1 = false
+#endif
+
+strings.forEach { (string, contiguous) in
+  suite.test("Contiguous: \(string)")
+  .require(.stdlib_6_2).code {
+
+    expectEqual(string.isContiguousUTF8, contiguous)
+  }
+}
+
+strings.forEach { (string, contiguous) in
+  suite.test("Contiguous Substring: \(string)")
+  .require(.stdlib_6_2).code {
+    let substring = string[...]
+    expectEqual(substring.isContiguousUTF8, contiguous)
+  }
+}
+
+strings.forEach { (string, contiguous) in
+  suite.test("String.makeContiguousUTF8: \(string)")
+  .require(.stdlib_6_2).code {
+    var s = string
+    s.makeContiguousUTF8()
+    expectTrue(s.isContiguousUTF8)
+    expectEqual(s, string)
+  }
+}
+
+strings.forEach { (string, contiguous) in
+  suite.test("Substring.makeContiguousUTF8: \(string)")
+  .require(.stdlib_6_2).code {
+    var s: Substring = string.dropFirst().dropLast()
+    s.makeContiguousUTF8()
+    expectTrue(s.isContiguousUTF8)
+    expectEqual(s, string.dropFirst().dropLast())
+  }
+}