Ziele Untersuchung
mit Columbo Integrität von
Datenbanken Interaktion und
Portierbarkeit Ergonomie der
Schnittstellen

Angebot Produkte Projekt Beratung

Mittel Analytik Modellierung Sprachen Algebra Logik Hardware Denken Kreativität

Zusammenhänge Gesellschaft Wirtschaft Branche Firma


products/sources/formale Sprachen/C/Firefox/js/src/jsapi-tests/ (Browser von der Mozilla Stiftung Version 136.0.1^©) Datei vom 10.2.2025 mit Größe 11 kB

Quelle testCompileUtf8.cpp Sprache: C

/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */

#include "mozilla/ArrayUtils.h"
#include "mozilla/TextUtils.h"
#include "mozilla/Utf8.h"

#include <cstring>

#include "js/CharacterEncoding.h"
#include "js/CompilationAndEvaluation.h"  // JS::Compile
#include "js/Exception.h"
#include "js/friend/ErrorMessages.h"  // JSMSG_*
#include "js/SourceText.h"
#include "jsapi-tests/tests.h"
#include "util/Text.h"
#include "vm/ErrorReporting.h"

using mozilla::ArrayEqual;
using mozilla::IsAsciiHexDigit;
using mozilla::Utf8Unit;

static bool contains(const char* str, const char* substr) {
  return std::strstr(str, substr) != nullptr;
}

BEGIN_TEST(testUtf8BadBytes) {
  static const char badLeadingUnit[] = "var x = \x80";
  CHECK(testBadUtf8(
      badLeadingUnit, JSMSG_BAD_LEADING_UTF8_UNIT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(startsWith(chars, "0x80"));
        CHECK(isBadLeadUnitMessage(chars));
        return true;
      },
      "0x80"));

  static const char badSecondInTwoByte[] = "var x = \xDF\x20";
  CHECK(testBadUtf8(
      badSecondInTwoByte, JSMSG_BAD_TRAILING_UTF8_UNIT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isBadTrailingBytesMessage(chars));
        CHECK(contains(chars, "0x20"));
        return true;
      },
      "0xDF 0x20"));

  static const char badSecondInThreeByte[] = "var x = \xEF\x17\xA7";
  CHECK(testBadUtf8(
      badSecondInThreeByte, JSMSG_BAD_TRAILING_UTF8_UNIT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isBadTrailingBytesMessage(chars));
        CHECK(contains(chars, "0x17"));
        return true;
      },
      // Validating stops with the first invalid code unit and
      // shouldn't go beyond that.
      "0xEF 0x17"));

  static const char lengthTwoTooShort[] = "var x = \xDF";
  CHECK(testBadUtf8(
      lengthTwoTooShort, JSMSG_NOT_ENOUGH_CODE_UNITS,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isNotEnoughUnitsMessage(chars));
        CHECK(contains(chars, "0xDF"));
        CHECK(contains(chars, " 1 byte, but 0 bytes were present"));
        return true;
      },
      "0xDF"));

  static const char forbiddenHighSurrogate[] = "var x = \xED\xA2\x87";
  CHECK(testBadUtf8(
      forbiddenHighSurrogate, JSMSG_FORBIDDEN_UTF8_CODE_POINT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isSurrogateMessage(chars));
        CHECK(contains(chars, "0xD887"));
        return true;
      },
      "0xED 0xA2 0x87"));

  static const char forbiddenLowSurrogate[] = "var x = \xED\xB7\xAF";
  CHECK(testBadUtf8(
      forbiddenLowSurrogate, JSMSG_FORBIDDEN_UTF8_CODE_POINT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isSurrogateMessage(chars));
        CHECK(contains(chars, "0xDDEF"));
        return true;
      },
      "0xED 0xB7 0xAF"));

  static const char oneTooBig[] = "var x = \xF4\x90\x80\x80";
  CHECK(testBadUtf8(
      oneTooBig, JSMSG_FORBIDDEN_UTF8_CODE_POINT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isTooBigMessage(chars));
        CHECK(contains(chars, "0x110000"));
        return true;
      },
      "0xF4 0x90 0x80 0x80"));

  static const char notShortestFormZero[] = "var x = \xC0\x80";
  CHECK(testBadUtf8(
      notShortestFormZero, JSMSG_FORBIDDEN_UTF8_CODE_POINT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isNotShortestFormMessage(chars));
        CHECK(startsWith(chars, "0x0 isn't "));
        return true;
      },
      "0xC0 0x80"));

  static const char notShortestFormNonzero[] = "var x = \xE0\x87\x80";
  CHECK(testBadUtf8(
      notShortestFormNonzero, JSMSG_FORBIDDEN_UTF8_CODE_POINT,
      [this](JS::ConstUTF8CharsZ message) {
        const char* chars = message.c_str();
        CHECK(isNotShortestFormMessage(chars));
        CHECK(startsWith(chars, "0x1C0 isn't "));
        return true;
      },
      "0xE0 0x87 0x80"));

  return true;
}

static constexpr size_t LengthOfByte = js_strlen("0xFF");

static bool startsWithByte(const char* str) {
  return str[0] == '0' && str[1] == 'x' && IsAsciiHexDigit(str[2]) &&
         IsAsciiHexDigit(str[3]);
}

static bool startsWith(const char* str, const char* prefix) {
  return std::strncmp(prefix, str, strlen(prefix)) == 0;
}

static bool equals(const char* str, const char* expected) {
  return std::strcmp(str, expected) == 0;
}

static bool isBadLeadUnitMessage(const char* str) {
  return startsWithByte(str) &&
         equals(str + LengthOfByte,
                " byte doesn't begin a valid UTF-8 code point");
}

static bool isBadTrailingBytesMessage(const char* str) {
  return startsWith(str, "bad trailing UTF-8 byte ");
}

static bool isNotEnoughUnitsMessage(const char* str) {
  return startsWithByte(str) &&
         startsWith(str + LengthOfByte, " byte in UTF-8 must be followed by ");
}

static bool isForbiddenCodePointMessage(const char* str) {
  return contains(str, "isn't a valid code point because");
}

static bool isSurrogateMessage(const char* str) {
  return isForbiddenCodePointMessage(str) &&
         contains(str, " it's a UTF-16 surrogate");
}

static bool isTooBigMessage(const char* str) {
  return isForbiddenCodePointMessage(str) &&
         contains(str, "the maximum code point is U+10FFFF");
}

static bool isNotShortestFormMessage(const char* str) {
  return isForbiddenCodePointMessage(str) &&
         contains(str, "it wasn't encoded in shortest possible form");
}

template <size_t N, typename TestMessage>
bool testBadUtf8(const char (&chars)[N], unsigned errorNumber,
                 TestMessage testMessage, const char* badBytes) {
  JS::Rooted<JSScript*> script(cx);
  {
    JS::CompileOptions options(cx);

    JS::SourceText<mozilla::Utf8Unit> srcBuf;
    CHECK(srcBuf.init(cx, chars, N - 1, JS::SourceOwnership::Borrowed));

    script = JS::Compile(cx, options, srcBuf);
    CHECK(!script);
  }

  JS::ExceptionStack exnStack(cx);
  CHECK(JS::StealPendingExceptionStack(cx, &exnStack));

  JS::ErrorReportBuilder report(cx);
  CHECK(report.init(cx, exnStack, JS::ErrorReportBuilder::WithSideEffects));

  const auto* errorReport = report.report();

  CHECK(errorReport->errorNumber == errorNumber);

  CHECK(testMessage(errorReport->message()));

  {
    const auto& notes = errorReport->notes;
    CHECK(notes != nullptr);

    auto iter = notes->begin();
    CHECK(iter != notes->end());

    const char* noteMessage = (*iter)->message().c_str();

    // The prefix ought always be the same.
    static constexpr char expectedPrefix[] =
        "the code units comprising this invalid code point were: ";
    constexpr size_t expectedPrefixLen = js_strlen(expectedPrefix);

    CHECK(startsWith(noteMessage, expectedPrefix));

    // The end of the prefix is the bad bytes.
    CHECK(equals(noteMessage + expectedPrefixLen, badBytes));

    ++iter;
    CHECK(iter == notes->end());
  }

  static constexpr char16_t expectedContext[] = u"var x = ";
  constexpr size_t expectedContextLen = js_strlen(expectedContext);

  const char16_t* lineOfContext = errorReport->linebuf();
  size_t lineOfContextLength = errorReport->linebufLength();

  CHECK(lineOfContext[lineOfContextLength] == '\0');
  CHECK(lineOfContextLength == expectedContextLen);

  CHECK(std::memcmp(lineOfContext, expectedContext,
                    expectedContextLen * sizeof(char16_t)) == 0);

  return true;
}
END_TEST(testUtf8BadBytes)

BEGIN_TEST(testMultiUnitUtf8InWindow) {
  static const char firstInWindowIsMultiUnit[] =
      "\xCF\x80\xCF\x80 = 6.283185307; @ bad starts HERE:\x80\xFF\xFF";
  CHECK(testContext(firstInWindowIsMultiUnit,
                    u"ππ = 6.283185307; @ bad starts HERE:"));

  static const char atTokenOffsetIsMulti[] = "var z = 💯";
  CHECK(testContext(atTokenOffsetIsMulti, u"var z = 💯"));

  static const char afterTokenOffsetIsMulti[] = "var z = @💯💯💯X";
  CHECK(testContext(afterTokenOffsetIsMulti, u"var z = @💯💯💯X"));

  static const char atEndIsMulti[] = "var z = @@💯💯💯";
  CHECK(testContext(atEndIsMulti, u"var z = @@💯💯💯"));

  return true;
}

template <size_t N, size_t ContextLenWithNull>
bool testContext(const char (&chars)[N],
                 const char16_t (&expectedContext)[ContextLenWithNull]) {
  JS::Rooted<JSScript*> script(cx);
  {
    JS::CompileOptions options(cx);

    JS::SourceText<mozilla::Utf8Unit> srcBuf;
    CHECK(srcBuf.init(cx, chars, N - 1, JS::SourceOwnership::Borrowed));

    script = JS::Compile(cx, options, srcBuf);
    CHECK(!script);
  }

  JS::ExceptionStack exnStack(cx);
  CHECK(JS::StealPendingExceptionStack(cx, &exnStack));

  JS::ErrorReportBuilder report(cx);
  CHECK(report.init(cx, exnStack, JS::ErrorReportBuilder::WithSideEffects));

  const auto* errorReport = report.report();

  CHECK(errorReport->errorNumber == JSMSG_ILLEGAL_CHARACTER);

  const char16_t* lineOfContext = errorReport->linebuf();
  size_t lineOfContextLength = errorReport->linebufLength();

  CHECK(lineOfContext[lineOfContextLength] == '\0');
  CHECK(lineOfContextLength == ContextLenWithNull - 1);

  CHECK(ArrayEqual(lineOfContext, expectedContext, ContextLenWithNull));

  return true;
}
END_TEST(testMultiUnitUtf8InWindow)

BEGIN_TEST(testCompileJsonModule) {
  static const char chars[] = "{ \"a\": 1, \"b\": 2, \"c\": \"foo\" }";
  JS::Rooted<JSObject*> module(cx);
  {
    JS::CompileOptions options(cx);

    JS::SourceText<mozilla::Utf8Unit> srcBuf;
    CHECK(srcBuf.init(cx, chars, js_strlen(chars),
                      JS::SourceOwnership::Borrowed));

    module = JS::CompileJsonModule(cx, options, srcBuf);
    CHECK(module);
  }

  return true;
}
END_TEST(testCompileJsonModule)

BEGIN_TEST(testCompileJsonModuleInvalidJson) {
  static const char chars[] = "{ \"a\": 1, \"b\": 2, \"c\":";
  JS::Rooted<JSObject*> module(cx);
  {
    JS::CompileOptions options(cx);

    JS::SourceText<mozilla::Utf8Unit> srcBuf;
    CHECK(srcBuf.init(cx, chars, js_strlen(chars),
                      JS::SourceOwnership::Borrowed));

    module = JS::CompileJsonModule(cx, options, srcBuf);
    CHECK(!module);
  }

  JS::ExceptionStack exnStack(cx);
  CHECK(JS::StealPendingExceptionStack(cx, &exnStack));

  JS::ErrorReportBuilder report(cx);
  CHECK(report.init(cx, exnStack, JS::ErrorReportBuilder::WithSideEffects));

  const auto* errorReport = report.report();
  CHECK(errorReport->errorNumber == JSMSG_JSON_BAD_PARSE);
  CHECK(contains(errorReport->message().c_str(),
                 "JSON.parse: unexpected end of data"));

  return true;
}
END_TEST(testCompileJsonModuleInvalidJson)

BEGIN_TEST(testCompileJsonModuleBadUtf8) {
  static const char chars[] = "{ \"a\": 1, \"b\": 2, \"c\": \"\xDF\x20\" }";
  JS::Rooted<JSObject*> module(cx);
  {
    JS::CompileOptions options(cx);

    JS::SourceText<mozilla::Utf8Unit> srcBuf;
    CHECK(srcBuf.init(cx, chars, js_strlen(chars),
                      JS::SourceOwnership::Borrowed));

    module = JS::CompileJsonModule(cx, options, srcBuf);
    CHECK(!module);
  }

  JS::ExceptionStack exnStack(cx);
  CHECK(JS::StealPendingExceptionStack(cx, &exnStack));

  JS::ErrorReportBuilder report(cx);
  CHECK(report.init(cx, exnStack, JS::ErrorReportBuilder::WithSideEffects));

  const auto* errorReport = report.report();
  CHECK(errorReport->errorNumber == JSMSG_MALFORMED_UTF8_CHAR);
  CHECK(contains(errorReport->message().c_str(),
                 "malformed UTF-8 character sequence at offset"));

  return true;
}
END_TEST(testCompileJsonModuleBadUtf8)

quality100%

¤ Dauer der Verarbeitung: 0.18 Sekunden (vorverarbeitet) ¤

Wurzel

Suchen

Beweissystem der NASA

Beweissystem Isabelle

NIST Cobol Testsuite

Cephes Mathematical Library

Wiener Entwicklungsmethode

Haftungshinweis

Die Informationen auf dieser Webseite wurden nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit, noch Qualität der bereit gestellten Informationen zugesichert.