Merge pull request #998 from pkwarren:pkw/string-ext-quote-reverse

PiperOrigin-RevId: 899240904

Author: copybara-github

conformance/src/test/java/dev/cel/conformance/BUILD.bazel

@@ -119,8 +119,7 @@ _TESTS_TO_SKIP_LEGACY = [
 
     # Skip until fixed.
     "fields/qualified_identifier_resolution/map_value_repeat_key_heterogeneous",
-    # TODO: Add strings.format and strings.quote.
-    "string_ext/quote",
+    # TODO: Add strings.format.quote.
     "string_ext/format",
     "string_ext/format_errors",
 
@@ -148,8 +147,7 @@ _TESTS_TO_SKIP_LEGACY = [
 ]
 
 _TESTS_TO_SKIP_PLANNER = [
-    # TODO: Add strings.format and strings.quote.
-    "string_ext/quote",
+    # TODO: Add strings.format.
     "string_ext/format",
     "string_ext/format_errors",
 

extensions/src/main/java/dev/cel/extensions/CelStringExtensions.java

@@ -137,6 +137,17 @@ public enum Function {
                 SimpleType.STRING,
                 SimpleType.STRING)),
         CelFunctionBinding.from("string_lower_ascii", String.class, Ascii::toLowerCase)),
+    QUOTE(
+        CelFunctionDecl.newFunctionDeclaration(
+            "strings.quote",
+            CelOverloadDecl.newGlobalOverload(
+                "strings_quote",
+                "Takes the given string and makes it safe to print (without any formatting"
+                    + " due to escape sequences). If any invalid UTF-8 characters are"
+                    + " encountered, they are replaced with \\uFFFD.",
+                SimpleType.STRING,
+                ImmutableList.of(SimpleType.STRING))),
+        CelFunctionBinding.from("strings_quote", String.class, CelStringExtensions::quote)),
     REPLACE(
         CelFunctionDecl.newFunctionDeclaration(
             "replace",
@@ -164,6 +175,16 @@ public enum Function {
             "string_replace_string_string_int",
             ImmutableList.of(String.class, String.class, String.class, Long.class),
             CelStringExtensions::replace)),
+    REVERSE(
+        CelFunctionDecl.newFunctionDeclaration(
+            "reverse",
+            CelOverloadDecl.newMemberOverload(
+                "string_reverse",
+                "Returns a new string whose characters are the same as the target string,"
+                    + " only formatted in reverse order.",
+                SimpleType.STRING,
+                SimpleType.STRING)),
+        CelFunctionBinding.from("string_reverse", String.class, CelStringExtensions::reverse)),
     SPLIT(
         CelFunctionDecl.newFunctionDeclaration(
             "split",
@@ -449,6 +470,64 @@ private static Long lastIndexOf(CelCodePointArray str, CelCodePointArray substr,
     return -1L;
   }
 
+  private static String quote(String s) {
+    StringBuilder sb = new StringBuilder(s.length() + 2);
+    sb.append('"');
+    for (int i = 0; i < s.length(); ) {
+      int codePoint = s.codePointAt(i);
+      if (isMalformedUtf16(s, i, codePoint)) {
+        sb.append('\uFFFD');
+        i++;
+        continue;
+      }
+      switch (codePoint) {
+        case '\u0007':
+          sb.append("\\a");
+          break;
+        case '\b':
+          sb.append("\\b");
+          break;
+        case '\f':
+          sb.append("\\f");
+          break;
+        case '\n':
+          sb.append("\\n");
+          break;
+        case '\r':
+          sb.append("\\r");
+          break;
+        case '\t':
+          sb.append("\\t");
+          break;
+        case '\u000B':
+          sb.append("\\v");
+          break;
+        case '\\':
+          sb.append("\\\\");
+          break;
+        case '"':
+          sb.append("\\\"");
+          break;
+        default:
+          sb.appendCodePoint(codePoint);
+          break;
+      }
+      i += Character.charCount(codePoint);
+    }
+    sb.append('"');
+    return sb.toString();
+  }
+
+  private static boolean isMalformedUtf16(String s, int index, int codePoint) {
+    char currentChar = s.charAt(index);
+    if (Character.isLowSurrogate(currentChar)) {
+      return true;
+    }
+    // Check for unpaired high surrogate
+    return Character.isHighSurrogate(currentChar)
+        && (index + 1 >= s.length() || !Character.isLowSurrogate(s.charAt(index + 1)));
+  }
+
   private static String replaceAll(Object[] objects) {
     return replace((String) objects[0], (String) objects[1], (String) objects[2], -1);
   }
@@ -504,6 +583,10 @@ private static String replace(String text, String searchString, String replaceme
     return sb.append(textCpa.slice(start, textCpa.length())).toString();
   }
 
+  private static String reverse(String s) {
+    return new StringBuilder(s).reverse().toString();
+  }
+
   private static List<String> split(String str, String separator) {
     return split(str, separator, Integer.MAX_VALUE);
   }

extensions/src/main/java/dev/cel/extensions/README.md

@@ -474,6 +474,19 @@ Examples:
      'TacoCat'.lowerAscii()      // returns 'tacocat'
      'TacoCÆt Xii'.lowerAscii()  // returns 'tacocÆt xii'
 
+### Quote
+
+Takes the given string and makes it safe to print (without any formatting due
+to escape sequences).
+If any invalid UTF-8 characters are encountered, they are replaced with \uFFFD.
+
+    strings.quote(<string>)
+
+Examples:
+
+    strings.quote('single-quote with "double quote"') // returns '"single-quote with \"double quote\""'
+    strings.quote("two escape sequences \a\n") // returns '"two escape sequences \\a\\n"'
+
 ### Replace
 
 Returns a new string based on the target, which replaces the occurrences of a
@@ -493,6 +506,20 @@ Examples:
     'hello hello'.replace('he', 'we', 1)  // returns 'wello hello'
     'hello hello'.replace('he', 'we', 0)  // returns 'hello hello'
 
+### Reverse
+
+Returns a new string whose characters are the same as the target string, only
+formatted in reverse order.
+This function relies on converting strings to Unicode code point arrays in
+order to reverse.
+
+    <string>.reverse() -> <string>
+
+Examples:
+
+    'gums'.reverse() // returns 'smug'
+    'John Smith'.reverse() // returns 'htimS nhoJ'
+
 ### Split
 
 Returns a mutable list of strings split from the input by the given separator. The

extensions/src/test/java/dev/cel/extensions/CelExtensionsTest.java

@@ -168,6 +168,7 @@ public void getAllFunctionNames() {
             "join",
             "lastIndexOf",
             "lowerAscii",
+            "strings.quote",
             "replace",
             "split",
             "substring",

extensions/src/test/java/dev/cel/extensions/CelStringExtensionsTest.java

@@ -70,7 +70,9 @@ public void library() {
             "lastIndexOf",
             "lowerAscii",
             "replace",
+            "reverse",
             "split",
+            "strings.quote",
             "substring",
             "trim",
             "upperAscii");
@@ -1467,6 +1469,100 @@ public void stringExtension_functionSubset_success() throws Exception {
     assertThat(evaluatedResult).isEqualTo(true);
   }
 
+  @Test
+  @TestParameters("{string: 'abcd', expectedResult: 'dcba'}")
+  @TestParameters("{string: '', expectedResult: ''}")
+  @TestParameters("{string: 'a', expectedResult: 'a'}")
+  @TestParameters("{string: 'hello world', expectedResult: 'dlrow olleh'}")
+  @TestParameters("{string: 'ab가cd', expectedResult: 'dc가ba'}")
+  public void reverse_success(String string, String expectedResult) throws Exception {
+    CelAbstractSyntaxTree ast = COMPILER.compile("s.reverse()").getAst();
+    CelRuntime.Program program = RUNTIME.createProgram(ast);
+
+    Object evaluatedResult = program.eval(ImmutableMap.of("s", string));
+
+    assertThat(evaluatedResult).isEqualTo(expectedResult);
+  }
+
+  @Test
+  @TestParameters("{string: '😁😑😦', expectedResult: '😦😑😁'}")
+  @TestParameters(
+      "{string: '\u180e\u200b\u200c\u200d\u2060\ufeff', expectedResult:"
+          + " '\ufeff\u2060\u200d\u200c\u200b\u180e'}")
+  public void reverse_unicode(String string, String expectedResult) throws Exception {
+    CelAbstractSyntaxTree ast = COMPILER.compile("s.reverse()").getAst();
+    CelRuntime.Program program = RUNTIME.createProgram(ast);
+
+    Object evaluatedResult = program.eval(ImmutableMap.of("s", string));
+
+    assertThat(evaluatedResult).isEqualTo(expectedResult);
+  }
+
+  @Test
+  @TestParameters("{string: 'hello', expectedResult: '\"hello\"'}")
+  @TestParameters("{string: '', expectedResult: '\"\"'}")
+  @TestParameters(
+      "{string: 'contains \\\"quotes\\\"', expectedResult: '\"contains \\\\\\\"quotes\\\\\\\"\"'}")
+  @TestParameters("{string: 'ends with \\\\', expectedResult: '\"ends with \\\\\\\\\"'}")
+  @TestParameters("{string: '\\\\ starts with', expectedResult: '\"\\\\\\\\ starts with\"'}")
+  public void quote_success(String string, String expectedResult) throws Exception {
+    CelAbstractSyntaxTree ast = COMPILER.compile("strings.quote(s)").getAst();
+    CelRuntime.Program program = RUNTIME.createProgram(ast);
+
+    Object evaluatedResult = program.eval(ImmutableMap.of("s", string));
+
+    assertThat(evaluatedResult).isEqualTo(expectedResult);
+  }
+
+  @Test
+  public void quote_singleWithDoubleQuotes() throws Exception {
+    String expr = "strings.quote('single-quote with \"double quote\"')";
+    String expected = "\"\\\"single-quote with \\\\\\\"double quote\\\\\\\"\\\"\"";
+    CelAbstractSyntaxTree ast = COMPILER.compile(expr + " == " + expected).getAst();
+    CelRuntime.Program program = RUNTIME.createProgram(ast);
+
+    Object evaluatedResult = program.eval();
+
+    assertThat(evaluatedResult).isEqualTo(true);
+  }
+
+  @Test
+  public void quote_escapesSpecialCharacters() throws Exception {
+    CelAbstractSyntaxTree ast = COMPILER.compile("strings.quote(s)").getAst();
+    CelRuntime.Program program = RUNTIME.createProgram(ast);
+
+    Object evaluatedResult =
+        program.eval(
+            ImmutableMap.of("s", "\u0007bell\u000Bvtab\bback\ffeed\rret\nline\ttab\\slash 가 😁"));
+
+    assertThat(evaluatedResult)
+        .isEqualTo("\"\\abell\\vvtab\\bback\\ffeed\\rret\\nline\\ttab\\\\slash 가 😁\"");
+  }
+
+  @Test
+  public void quote_escapesMalformed_endWithHighSurrogate() throws Exception {
+    CelRuntime.Program program =
+        RUNTIME.createProgram(COMPILER.compile("strings.quote(s)").getAst());
+    assertThat(program.eval(ImmutableMap.of("s", "end with high surrogate \uD83D")))
+        .isEqualTo("\"end with high surrogate \uFFFD\"");
+  }
+
+  @Test
+  public void quote_escapesMalformed_unpairedHighSurrogate() throws Exception {
+    CelRuntime.Program program =
+        RUNTIME.createProgram(COMPILER.compile("strings.quote(s)").getAst());
+    assertThat(program.eval(ImmutableMap.of("s", "bad pair \uD83DA")))
+        .isEqualTo("\"bad pair \uFFFDA\"");
+  }
+
+  @Test
+  public void quote_escapesMalformed_unpairedLowSurrogate() throws Exception {
+    CelRuntime.Program program =
+        RUNTIME.createProgram(COMPILER.compile("strings.quote(s)").getAst());
+    assertThat(program.eval(ImmutableMap.of("s", "bad pair \uDC00A")))
+        .isEqualTo("\"bad pair \uFFFDA\"");
+  }
+
   @Test
   public void stringExtension_compileUnallowedFunction_throws() {
     CelCompiler celCompiler =