src/runtime/runtime-strings.cc

// Copyright 2014 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#include "src/v8.h"

#include "src/arguments.h"
#include "src/jsregexp-inl.h"
#include "src/jsregexp.h"
#include "src/runtime/runtime-utils.h"
#include "src/string-builder.h"
#include "src/string-search.h"

namespace v8 {
namespace internal {


// Perform string match of pattern on subject, starting at start index.
// Caller must ensure that 0 <= start_index <= sub->length(),
// and should check that pat->length() + start_index <= sub->length().
int StringMatch(Isolate* isolate, Handle<String> sub, Handle<String> pat,
                int start_index) {
  DCHECK(0 <= start_index);
  DCHECK(start_index <= sub->length());

  int pattern_length = pat->length();
  if (pattern_length == 0) return start_index;

  int subject_length = sub->length();
  if (start_index + pattern_length > subject_length) return -1;

  sub = String::Flatten(sub);
  pat = String::Flatten(pat);

  DisallowHeapAllocation no_gc;  // ensure vectors stay valid
  // Extract flattened substrings of cons strings before getting encoding.
  String::FlatContent seq_sub = sub->GetFlatContent();
  String::FlatContent seq_pat = pat->GetFlatContent();

  // dispatch on type of strings
  if (seq_pat.IsOneByte()) {
    Vector<const uint8_t> pat_vector = seq_pat.ToOneByteVector();
    if (seq_sub.IsOneByte()) {
      return SearchString(isolate, seq_sub.ToOneByteVector(), pat_vector,
                          start_index);
    }
    return SearchString(isolate, seq_sub.ToUC16Vector(), pat_vector,
                        start_index);
  }
  Vector<const uc16> pat_vector = seq_pat.ToUC16Vector();
  if (seq_sub.IsOneByte()) {
    return SearchString(isolate, seq_sub.ToOneByteVector(), pat_vector,
                        start_index);
  }
  return SearchString(isolate, seq_sub.ToUC16Vector(), pat_vector, start_index);
}


// This may return an empty MaybeHandle if an exception is thrown or
// we abort due to reaching the recursion limit.
MaybeHandle<String> StringReplaceOneCharWithString(
    Isolate* isolate, Handle<String> subject, Handle<String> search,
    Handle<String> replace, bool* found, int recursion_limit) {
  StackLimitCheck stackLimitCheck(isolate);
  if (stackLimitCheck.HasOverflowed() || (recursion_limit == 0)) {
    return MaybeHandle<String>();
  }
  recursion_limit--;
  if (subject->IsConsString()) {
    ConsString* cons = ConsString::cast(*subject);
    Handle<String> first = Handle<String>(cons->first());
    Handle<String> second = Handle<String>(cons->second());
    Handle<String> new_first;
    if (!StringReplaceOneCharWithString(isolate, first, search, replace, found,
                                        recursion_limit).ToHandle(&new_first)) {
      return MaybeHandle<String>();
    }
    if (*found) return isolate->factory()->NewConsString(new_first, second);

    Handle<String> new_second;
    if (!StringReplaceOneCharWithString(isolate, second, search, replace, found,
                                        recursion_limit)
             .ToHandle(&new_second)) {
      return MaybeHandle<String>();
    }
    if (*found) return isolate->factory()->NewConsString(first, new_second);

    return subject;
  } else {
    int index = StringMatch(isolate, subject, search, 0);
    if (index == -1) return subject;
    *found = true;
    Handle<String> first = isolate->factory()->NewSubString(subject, 0, index);
    Handle<String> cons1;
    ASSIGN_RETURN_ON_EXCEPTION(
        isolate, cons1, isolate->factory()->NewConsString(first, replace),
        String);
    Handle<String> second =
        isolate->factory()->NewSubString(subject, index + 1, subject->length());
    return isolate->factory()->NewConsString(cons1, second);
  }
}


RUNTIME_FUNCTION(Runtime_StringReplaceOneCharWithString) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);
  CONVERT_ARG_HANDLE_CHECKED(String, subject, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, search, 1);
  CONVERT_ARG_HANDLE_CHECKED(String, replace, 2);

  // If the cons string tree is too deep, we simply abort the recursion and
  // retry with a flattened subject string.
  const int kRecursionLimit = 0x1000;
  bool found = false;
  Handle<String> result;
  if (StringReplaceOneCharWithString(isolate, subject, search, replace, &found,
                                     kRecursionLimit).ToHandle(&result)) {
    return *result;
  }
  if (isolate->has_pending_exception()) return isolate->heap()->exception();

  subject = String::Flatten(subject);
  ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
      isolate, result,
      StringReplaceOneCharWithString(isolate, subject, search, replace, &found,
                                     kRecursionLimit));
  return *result;
}


RUNTIME_FUNCTION(Runtime_StringIndexOf) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);

  CONVERT_ARG_HANDLE_CHECKED(String, sub, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, pat, 1);
  CONVERT_ARG_HANDLE_CHECKED(Object, index, 2);

  uint32_t start_index;
  if (!index->ToArrayIndex(&start_index)) return Smi::FromInt(-1);

  RUNTIME_ASSERT(start_index <= static_cast<uint32_t>(sub->length()));
  int position = StringMatch(isolate, sub, pat, start_index);
  return Smi::FromInt(position);
}


template <typename schar, typename pchar>
static int StringMatchBackwards(Vector<const schar> subject,
                                Vector<const pchar> pattern, int idx) {
  int pattern_length = pattern.length();
  DCHECK(pattern_length >= 1);
  DCHECK(idx + pattern_length <= subject.length());

  if (sizeof(schar) == 1 && sizeof(pchar) > 1) {
    for (int i = 0; i < pattern_length; i++) {
      uc16 c = pattern[i];
      if (c > String::kMaxOneByteCharCode) {
        return -1;
      }
    }
  }

  pchar pattern_first_char = pattern[0];
  for (int i = idx; i >= 0; i--) {
    if (subject[i] != pattern_first_char) continue;
    int j = 1;
    while (j < pattern_length) {
      if (pattern[j] != subject[i + j]) {
        break;
      }
      j++;
    }
    if (j == pattern_length) {
      return i;
    }
  }
  return -1;
}


RUNTIME_FUNCTION(Runtime_StringLastIndexOf) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);

  CONVERT_ARG_HANDLE_CHECKED(String, sub, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, pat, 1);
  CONVERT_ARG_HANDLE_CHECKED(Object, index, 2);

  uint32_t start_index;
  if (!index->ToArrayIndex(&start_index)) return Smi::FromInt(-1);

  uint32_t pat_length = pat->length();
  uint32_t sub_length = sub->length();

  if (start_index + pat_length > sub_length) {
    start_index = sub_length - pat_length;
  }

  if (pat_length == 0) {
    return Smi::FromInt(start_index);
  }

  sub = String::Flatten(sub);
  pat = String::Flatten(pat);

  int position = -1;
  DisallowHeapAllocation no_gc;  // ensure vectors stay valid

  String::FlatContent sub_content = sub->GetFlatContent();
  String::FlatContent pat_content = pat->GetFlatContent();

  if (pat_content.IsOneByte()) {
    Vector<const uint8_t> pat_vector = pat_content.ToOneByteVector();
    if (sub_content.IsOneByte()) {
      position = StringMatchBackwards(sub_content.ToOneByteVector(), pat_vector,
                                      start_index);
    } else {
      position = StringMatchBackwards(sub_content.ToUC16Vector(), pat_vector,
                                      start_index);
    }
  } else {
    Vector<const uc16> pat_vector = pat_content.ToUC16Vector();
    if (sub_content.IsOneByte()) {
      position = StringMatchBackwards(sub_content.ToOneByteVector(), pat_vector,
                                      start_index);
    } else {
      position = StringMatchBackwards(sub_content.ToUC16Vector(), pat_vector,
                                      start_index);
    }
  }

  return Smi::FromInt(position);
}


RUNTIME_FUNCTION(Runtime_StringLocaleCompare) {
  HandleScope handle_scope(isolate);
  DCHECK(args.length() == 2);

  CONVERT_ARG_HANDLE_CHECKED(String, str1, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, str2, 1);

  if (str1.is_identical_to(str2)) return Smi::FromInt(0);  // Equal.
  int str1_length = str1->length();
  int str2_length = str2->length();

  // Decide trivial cases without flattening.
  if (str1_length == 0) {
    if (str2_length == 0) return Smi::FromInt(0);  // Equal.
    return Smi::FromInt(-str2_length);
  } else {
    if (str2_length == 0) return Smi::FromInt(str1_length);
  }

  int end = str1_length < str2_length ? str1_length : str2_length;

  // No need to flatten if we are going to find the answer on the first
  // character.  At this point we know there is at least one character
  // in each string, due to the trivial case handling above.
  int d = str1->Get(0) - str2->Get(0);
  if (d != 0) return Smi::FromInt(d);

  str1 = String::Flatten(str1);
  str2 = String::Flatten(str2);

  DisallowHeapAllocation no_gc;
  String::FlatContent flat1 = str1->GetFlatContent();
  String::FlatContent flat2 = str2->GetFlatContent();

  for (int i = 0; i < end; i++) {
    if (flat1.Get(i) != flat2.Get(i)) {
      return Smi::FromInt(flat1.Get(i) - flat2.Get(i));
    }
  }

  return Smi::FromInt(str1_length - str2_length);
}


RUNTIME_FUNCTION(Runtime_SubString) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);

  CONVERT_ARG_HANDLE_CHECKED(String, string, 0);
  int start, end;
  // We have a fast integer-only case here to avoid a conversion to double in
  // the common case where from and to are Smis.
  if (args[1]->IsSmi() && args[2]->IsSmi()) {
    CONVERT_SMI_ARG_CHECKED(from_number, 1);
    CONVERT_SMI_ARG_CHECKED(to_number, 2);
    start = from_number;
    end = to_number;
  } else {
    CONVERT_DOUBLE_ARG_CHECKED(from_number, 1);
    CONVERT_DOUBLE_ARG_CHECKED(to_number, 2);
    start = FastD2IChecked(from_number);
    end = FastD2IChecked(to_number);
  }
  RUNTIME_ASSERT(end >= start);
  RUNTIME_ASSERT(start >= 0);
  RUNTIME_ASSERT(end <= string->length());
  isolate->counters()->sub_string_runtime()->Increment();

  return *isolate->factory()->NewSubString(string, start, end);
}


RUNTIME_FUNCTION(Runtime_StringAdd) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 2);
  CONVERT_ARG_HANDLE_CHECKED(String, str1, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, str2, 1);
  isolate->counters()->string_add_runtime()->Increment();
  Handle<String> result;
  ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
      isolate, result, isolate->factory()->NewConsString(str1, str2));
  return *result;
}


RUNTIME_FUNCTION(Runtime_InternalizeString) {
  HandleScope handles(isolate);
  RUNTIME_ASSERT(args.length() == 1);
  CONVERT_ARG_HANDLE_CHECKED(String, string, 0);
  return *isolate->factory()->InternalizeString(string);
}


RUNTIME_FUNCTION(Runtime_StringMatch) {
  HandleScope handles(isolate);
  DCHECK(args.length() == 3);

  CONVERT_ARG_HANDLE_CHECKED(String, subject, 0);
  CONVERT_ARG_HANDLE_CHECKED(JSRegExp, regexp, 1);
  CONVERT_ARG_HANDLE_CHECKED(JSArray, regexp_info, 2);

  RUNTIME_ASSERT(regexp_info->HasFastObjectElements());

  RegExpImpl::GlobalCache global_cache(regexp, subject, true, isolate);
  if (global_cache.HasException()) return isolate->heap()->exception();

  int capture_count = regexp->CaptureCount();

  ZoneScope zone_scope(isolate->runtime_zone());
  ZoneList<int> offsets(8, zone_scope.zone());

  while (true) {
    int32_t* match = global_cache.FetchNext();
    if (match == NULL) break;
    offsets.Add(match[0], zone_scope.zone());  // start
    offsets.Add(match[1], zone_scope.zone());  // end
  }

  if (global_cache.HasException()) return isolate->heap()->exception();

  if (offsets.length() == 0) {
    // Not a single match.
    return isolate->heap()->null_value();
  }

  RegExpImpl::SetLastMatchInfo(regexp_info, subject, capture_count,
                               global_cache.LastSuccessfulMatch());

  int matches = offsets.length() / 2;
  Handle<FixedArray> elements = isolate->factory()->NewFixedArray(matches);
  Handle<String> substring =
      isolate->factory()->NewSubString(subject, offsets.at(0), offsets.at(1));
  elements->set(0, *substring);
  for (int i = 1; i < matches; i++) {
    HandleScope temp_scope(isolate);
    int from = offsets.at(i * 2);
    int to = offsets.at(i * 2 + 1);
    Handle<String> substring =
        isolate->factory()->NewProperSubString(subject, from, to);
    elements->set(i, *substring);
  }
  Handle<JSArray> result = isolate->factory()->NewJSArrayWithElements(elements);
  result->set_length(Smi::FromInt(matches));
  return *result;
}


RUNTIME_FUNCTION(Runtime_StringCharCodeAtRT) {
  HandleScope handle_scope(isolate);
  DCHECK(args.length() == 2);

  CONVERT_ARG_HANDLE_CHECKED(String, subject, 0);
  CONVERT_NUMBER_CHECKED(uint32_t, i, Uint32, args[1]);

  // Flatten the string.  If someone wants to get a char at an index
  // in a cons string, it is likely that more indices will be
  // accessed.
  subject = String::Flatten(subject);

  if (i >= static_cast<uint32_t>(subject->length())) {
    return isolate->heap()->nan_value();
  }

  return Smi::FromInt(subject->Get(i));
}


RUNTIME_FUNCTION(Runtime_CharFromCode) {
  HandleScope handlescope(isolate);
  DCHECK(args.length() == 1);
  if (args[0]->IsNumber()) {
    CONVERT_NUMBER_CHECKED(uint32_t, code, Uint32, args[0]);
    code &= 0xffff;
    return *isolate->factory()->LookupSingleCharacterStringFromCode(code);
  }
  return isolate->heap()->empty_string();
}


RUNTIME_FUNCTION(Runtime_StringCompare) {
  HandleScope handle_scope(isolate);
  DCHECK(args.length() == 2);

  CONVERT_ARG_HANDLE_CHECKED(String, x, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, y, 1);

  isolate->counters()->string_compare_runtime()->Increment();

  // A few fast case tests before we flatten.
  if (x.is_identical_to(y)) return Smi::FromInt(EQUAL);
  if (y->length() == 0) {
    if (x->length() == 0) return Smi::FromInt(EQUAL);
    return Smi::FromInt(GREATER);
  } else if (x->length() == 0) {
    return Smi::FromInt(LESS);
  }

  int d = x->Get(0) - y->Get(0);
  if (d < 0)
    return Smi::FromInt(LESS);
  else if (d > 0)
    return Smi::FromInt(GREATER);

  // Slow case.
  x = String::Flatten(x);
  y = String::Flatten(y);

  DisallowHeapAllocation no_gc;
  Object* equal_prefix_result = Smi::FromInt(EQUAL);
  int prefix_length = x->length();
  if (y->length() < prefix_length) {
    prefix_length = y->length();
    equal_prefix_result = Smi::FromInt(GREATER);
  } else if (y->length() > prefix_length) {
    equal_prefix_result = Smi::FromInt(LESS);
  }
  int r;
  String::FlatContent x_content = x->GetFlatContent();
  String::FlatContent y_content = y->GetFlatContent();
  if (x_content.IsOneByte()) {
    Vector<const uint8_t> x_chars = x_content.ToOneByteVector();
    if (y_content.IsOneByte()) {
      Vector<const uint8_t> y_chars = y_content.ToOneByteVector();
      r = CompareChars(x_chars.start(), y_chars.start(), prefix_length);
    } else {
      Vector<const uc16> y_chars = y_content.ToUC16Vector();
      r = CompareChars(x_chars.start(), y_chars.start(), prefix_length);
    }
  } else {
    Vector<const uc16> x_chars = x_content.ToUC16Vector();
    if (y_content.IsOneByte()) {
      Vector<const uint8_t> y_chars = y_content.ToOneByteVector();
      r = CompareChars(x_chars.start(), y_chars.start(), prefix_length);
    } else {
      Vector<const uc16> y_chars = y_content.ToUC16Vector();
      r = CompareChars(x_chars.start(), y_chars.start(), prefix_length);
    }
  }
  Object* result;
  if (r == 0) {
    result = equal_prefix_result;
  } else {
    result = (r < 0) ? Smi::FromInt(LESS) : Smi::FromInt(GREATER);
  }
  return result;
}


RUNTIME_FUNCTION(Runtime_StringBuilderConcat) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);
  CONVERT_ARG_HANDLE_CHECKED(JSArray, array, 0);
  int32_t array_length;
  if (!args[1]->ToInt32(&array_length)) {
    THROW_NEW_ERROR_RETURN_FAILURE(isolate, NewInvalidStringLengthError());
  }
  CONVERT_ARG_HANDLE_CHECKED(String, special, 2);

  size_t actual_array_length = 0;
  RUNTIME_ASSERT(
      TryNumberToSize(isolate, array->length(), &actual_array_length));
  RUNTIME_ASSERT(array_length >= 0);
  RUNTIME_ASSERT(static_cast<size_t>(array_length) <= actual_array_length);

  // This assumption is used by the slice encoding in one or two smis.
  DCHECK(Smi::kMaxValue >= String::kMaxLength);

  RUNTIME_ASSERT(array->HasFastElements());
  JSObject::EnsureCanContainHeapObjectElements(array);

  int special_length = special->length();
  if (!array->HasFastObjectElements()) {
    return isolate->Throw(isolate->heap()->illegal_argument_string());
  }

  int length;
  bool one_byte = special->HasOnlyOneByteChars();

  {
    DisallowHeapAllocation no_gc;
    FixedArray* fixed_array = FixedArray::cast(array->elements());
    if (fixed_array->length() < array_length) {
      array_length = fixed_array->length();
    }

    if (array_length == 0) {
      return isolate->heap()->empty_string();
    } else if (array_length == 1) {
      Object* first = fixed_array->get(0);
      if (first->IsString()) return first;
    }
    length = StringBuilderConcatLength(special_length, fixed_array,
                                       array_length, &one_byte);
  }

  if (length == -1) {
    return isolate->Throw(isolate->heap()->illegal_argument_string());
  }

  if (one_byte) {
    Handle<SeqOneByteString> answer;
    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
        isolate, answer, isolate->factory()->NewRawOneByteString(length));
    StringBuilderConcatHelper(*special, answer->GetChars(),
                              FixedArray::cast(array->elements()),
                              array_length);
    return *answer;
  } else {
    Handle<SeqTwoByteString> answer;
    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
        isolate, answer, isolate->factory()->NewRawTwoByteString(length));
    StringBuilderConcatHelper(*special, answer->GetChars(),
                              FixedArray::cast(array->elements()),
                              array_length);
    return *answer;
  }
}


RUNTIME_FUNCTION(Runtime_StringBuilderJoin) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);
  CONVERT_ARG_HANDLE_CHECKED(JSArray, array, 0);
  int32_t array_length;
  if (!args[1]->ToInt32(&array_length)) {
    THROW_NEW_ERROR_RETURN_FAILURE(isolate, NewInvalidStringLengthError());
  }
  CONVERT_ARG_HANDLE_CHECKED(String, separator, 2);
  RUNTIME_ASSERT(array->HasFastObjectElements());
  RUNTIME_ASSERT(array_length >= 0);

  Handle<FixedArray> fixed_array(FixedArray::cast(array->elements()));
  if (fixed_array->length() < array_length) {
    array_length = fixed_array->length();
  }

  if (array_length == 0) {
    return isolate->heap()->empty_string();
  } else if (array_length == 1) {
    Object* first = fixed_array->get(0);
    RUNTIME_ASSERT(first->IsString());
    return first;
  }

  int separator_length = separator->length();
  RUNTIME_ASSERT(separator_length > 0);
  int max_nof_separators =
      (String::kMaxLength + separator_length - 1) / separator_length;
  if (max_nof_separators < (array_length - 1)) {
    THROW_NEW_ERROR_RETURN_FAILURE(isolate, NewInvalidStringLengthError());
  }
  int length = (array_length - 1) * separator_length;
  for (int i = 0; i < array_length; i++) {
    Object* element_obj = fixed_array->get(i);
    RUNTIME_ASSERT(element_obj->IsString());
    String* element = String::cast(element_obj);
    int increment = element->length();
    if (increment > String::kMaxLength - length) {
      STATIC_ASSERT(String::kMaxLength < kMaxInt);
      length = kMaxInt;  // Provoke exception;
      break;
    }
    length += increment;
  }

  Handle<SeqTwoByteString> answer;
  ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
      isolate, answer, isolate->factory()->NewRawTwoByteString(length));

  DisallowHeapAllocation no_gc;

  uc16* sink = answer->GetChars();
#ifdef DEBUG
  uc16* end = sink + length;
#endif

  RUNTIME_ASSERT(fixed_array->get(0)->IsString());
  String* first = String::cast(fixed_array->get(0));
  String* separator_raw = *separator;
  int first_length = first->length();
  String::WriteToFlat(first, sink, 0, first_length);
  sink += first_length;

  for (int i = 1; i < array_length; i++) {
    DCHECK(sink + separator_length <= end);
    String::WriteToFlat(separator_raw, sink, 0, separator_length);
    sink += separator_length;

    RUNTIME_ASSERT(fixed_array->get(i)->IsString());
    String* element = String::cast(fixed_array->get(i));
    int element_length = element->length();
    DCHECK(sink + element_length <= end);
    String::WriteToFlat(element, sink, 0, element_length);
    sink += element_length;
  }
  DCHECK(sink == end);

  // Use %_FastOneByteArrayJoin instead.
  DCHECK(!answer->IsOneByteRepresentation());
  return *answer;
}

template <typename Char>
static void JoinSparseArrayWithSeparator(FixedArray* elements,
                                         int elements_length,
                                         uint32_t array_length,
                                         String* separator,
                                         Vector<Char> buffer) {
  DisallowHeapAllocation no_gc;
  int previous_separator_position = 0;
  int separator_length = separator->length();
  int cursor = 0;
  for (int i = 0; i < elements_length; i += 2) {
    int position = NumberToInt32(elements->get(i));
    String* string = String::cast(elements->get(i + 1));
    int string_length = string->length();
    if (string->length() > 0) {
      while (previous_separator_position < position) {
        String::WriteToFlat<Char>(separator, &buffer[cursor], 0,
                                  separator_length);
        cursor += separator_length;
        previous_separator_position++;
      }
      String::WriteToFlat<Char>(string, &buffer[cursor], 0, string_length);
      cursor += string->length();
    }
  }
  if (separator_length > 0) {
    // Array length must be representable as a signed 32-bit number,
    // otherwise the total string length would have been too large.
    DCHECK(array_length <= 0x7fffffff);  // Is int32_t.
    int last_array_index = static_cast<int>(array_length - 1);
    while (previous_separator_position < last_array_index) {
      String::WriteToFlat<Char>(separator, &buffer[cursor], 0,
                                separator_length);
      cursor += separator_length;
      previous_separator_position++;
    }
  }
  DCHECK(cursor <= buffer.length());
}


RUNTIME_FUNCTION(Runtime_SparseJoinWithSeparator) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);
  CONVERT_ARG_HANDLE_CHECKED(JSArray, elements_array, 0);
  CONVERT_NUMBER_CHECKED(uint32_t, array_length, Uint32, args[1]);
  CONVERT_ARG_HANDLE_CHECKED(String, separator, 2);
  // elements_array is fast-mode JSarray of alternating positions
  // (increasing order) and strings.
  RUNTIME_ASSERT(elements_array->HasFastSmiOrObjectElements());
  // array_length is length of original array (used to add separators);
  // separator is string to put between elements. Assumed to be non-empty.
  RUNTIME_ASSERT(array_length > 0);

  // Find total length of join result.
  int string_length = 0;
  bool is_one_byte = separator->IsOneByteRepresentation();
  bool overflow = false;
  CONVERT_NUMBER_CHECKED(int, elements_length, Int32, elements_array->length());
  RUNTIME_ASSERT(elements_length <= elements_array->elements()->length());
  RUNTIME_ASSERT((elements_length & 1) == 0);  // Even length.
  FixedArray* elements = FixedArray::cast(elements_array->elements());
  for (int i = 0; i < elements_length; i += 2) {
    RUNTIME_ASSERT(elements->get(i)->IsNumber());
    CONVERT_NUMBER_CHECKED(uint32_t, position, Uint32, elements->get(i));
    RUNTIME_ASSERT(position < array_length);
    RUNTIME_ASSERT(elements->get(i + 1)->IsString());
  }

  {
    DisallowHeapAllocation no_gc;
    for (int i = 0; i < elements_length; i += 2) {
      String* string = String::cast(elements->get(i + 1));
      int length = string->length();
      if (is_one_byte && !string->IsOneByteRepresentation()) {
        is_one_byte = false;
      }
      if (length > String::kMaxLength ||
          String::kMaxLength - length < string_length) {
        overflow = true;
        break;
      }
      string_length += length;
    }
  }

  int separator_length = separator->length();
  if (!overflow && separator_length > 0) {
    if (array_length <= 0x7fffffffu) {
      int separator_count = static_cast<int>(array_length) - 1;
      int remaining_length = String::kMaxLength - string_length;
      if ((remaining_length / separator_length) >= separator_count) {
        string_length += separator_length * (array_length - 1);
      } else {
        // Not room for the separators within the maximal string length.
        overflow = true;
      }
    } else {
      // Nonempty separator and at least 2^31-1 separators necessary
      // means that the string is too large to create.
      STATIC_ASSERT(String::kMaxLength < 0x7fffffff);
      overflow = true;
    }
  }
  if (overflow) {
    // Throw an exception if the resulting string is too large. See
    // https://code.google.com/p/chromium/issues/detail?id=336820
    // for details.
    THROW_NEW_ERROR_RETURN_FAILURE(isolate, NewInvalidStringLengthError());
  }

  if (is_one_byte) {
    Handle<SeqOneByteString> result = isolate->factory()
                                          ->NewRawOneByteString(string_length)
                                          .ToHandleChecked();
    JoinSparseArrayWithSeparator<uint8_t>(
        FixedArray::cast(elements_array->elements()), elements_length,
        array_length, *separator,
        Vector<uint8_t>(result->GetChars(), string_length));
    return *result;
  } else {
    Handle<SeqTwoByteString> result = isolate->factory()
                                          ->NewRawTwoByteString(string_length)
                                          .ToHandleChecked();
    JoinSparseArrayWithSeparator<uc16>(
        FixedArray::cast(elements_array->elements()), elements_length,
        array_length, *separator,
        Vector<uc16>(result->GetChars(), string_length));
    return *result;
  }
}


// Copies Latin1 characters to the given fixed array looking up
// one-char strings in the cache. Gives up on the first char that is
// not in the cache and fills the remainder with smi zeros. Returns
// the length of the successfully copied prefix.
static int CopyCachedOneByteCharsToArray(Heap* heap, const uint8_t* chars,
                                         FixedArray* elements, int length) {
  DisallowHeapAllocation no_gc;
  FixedArray* one_byte_cache = heap->single_character_string_cache();
  Object* undefined = heap->undefined_value();
  int i;
  WriteBarrierMode mode = elements->GetWriteBarrierMode(no_gc);
  for (i = 0; i < length; ++i) {
    Object* value = one_byte_cache->get(chars[i]);
    if (value == undefined) break;
    elements->set(i, value, mode);
  }
  if (i < length) {
    DCHECK(Smi::FromInt(0) == 0);
    memset(elements->data_start() + i, 0, kPointerSize * (length - i));
  }
#ifdef DEBUG
  for (int j = 0; j < length; ++j) {
    Object* element = elements->get(j);
    DCHECK(element == Smi::FromInt(0) ||
           (element->IsString() && String::cast(element)->LooksValid()));
  }
#endif
  return i;
}


// Converts a String to JSArray.
// For example, "foo" => ["f", "o", "o"].
RUNTIME_FUNCTION(Runtime_StringToArray) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 2);
  CONVERT_ARG_HANDLE_CHECKED(String, s, 0);
  CONVERT_NUMBER_CHECKED(uint32_t, limit, Uint32, args[1]);

  s = String::Flatten(s);
  const int length = static_cast<int>(Min<uint32_t>(s->length(), limit));

  Handle<FixedArray> elements;
  int position = 0;
  if (s->IsFlat() && s->IsOneByteRepresentation()) {
    // Try using cached chars where possible.
    elements = isolate->factory()->NewUninitializedFixedArray(length);

    DisallowHeapAllocation no_gc;
    String::FlatContent content = s->GetFlatContent();
    if (content.IsOneByte()) {
      Vector<const uint8_t> chars = content.ToOneByteVector();
      // Note, this will initialize all elements (not only the prefix)
      // to prevent GC from seeing partially initialized array.
      position = CopyCachedOneByteCharsToArray(isolate->heap(), chars.start(),
                                               *elements, length);
    } else {
      MemsetPointer(elements->data_start(), isolate->heap()->undefined_value(),
                    length);
    }
  } else {
    elements = isolate->factory()->NewFixedArray(length);
  }
  for (int i = position; i < length; ++i) {
    Handle<Object> str =
        isolate->factory()->LookupSingleCharacterStringFromCode(s->Get(i));
    elements->set(i, *str);
  }

#ifdef DEBUG
  for (int i = 0; i < length; ++i) {
    DCHECK(String::cast(elements->get(i))->length() == 1);
  }
#endif

  return *isolate->factory()->NewJSArrayWithElements(elements);
}


static inline bool ToUpperOverflows(uc32 character) {
  // y with umlauts and the micro sign are the only characters that stop
  // fitting into one-byte when converting to uppercase.
  static const uc32 yuml_code = 0xff;
  static const uc32 micro_code = 0xb5;
  return (character == yuml_code || character == micro_code);
}


template <class Converter>
MUST_USE_RESULT static Object* ConvertCaseHelper(
    Isolate* isolate, String* string, SeqString* result, int result_length,
    unibrow::Mapping<Converter, 128>* mapping) {
  DisallowHeapAllocation no_gc;
  // We try this twice, once with the assumption that the result is no longer
  // than the input and, if that assumption breaks, again with the exact
  // length.  This may not be pretty, but it is nicer than what was here before
  // and I hereby claim my vaffel-is.
  //
  // NOTE: This assumes that the upper/lower case of an ASCII
  // character is also ASCII.  This is currently the case, but it
  // might break in the future if we implement more context and locale
  // dependent upper/lower conversions.
  bool has_changed_character = false;

  // Convert all characters to upper case, assuming that they will fit
  // in the buffer
  StringCharacterStream stream(string);
  unibrow::uchar chars[Converter::kMaxWidth];
  // We can assume that the string is not empty
  uc32 current = stream.GetNext();
  bool ignore_overflow = Converter::kIsToLower || result->IsSeqTwoByteString();
  for (int i = 0; i < result_length;) {
    bool has_next = stream.HasMore();
    uc32 next = has_next ? stream.GetNext() : 0;
    int char_length = mapping->get(current, next, chars);
    if (char_length == 0) {
      // The case conversion of this character is the character itself.
      result->Set(i, current);
      i++;
    } else if (char_length == 1 &&
               (ignore_overflow || !ToUpperOverflows(current))) {
      // Common case: converting the letter resulted in one character.
      DCHECK(static_cast<uc32>(chars[0]) != current);
      result->Set(i, chars[0]);
      has_changed_character = true;
      i++;
    } else if (result_length == string->length()) {
      bool overflows = ToUpperOverflows(current);
      // We've assumed that the result would be as long as the
      // input but here is a character that converts to several
      // characters.  No matter, we calculate the exact length
      // of the result and try the whole thing again.
      //
      // Note that this leaves room for optimization.  We could just
      // memcpy what we already have to the result string.  Also,
      // the result string is the last object allocated we could
      // "realloc" it and probably, in the vast majority of cases,
      // extend the existing string to be able to hold the full
      // result.
      int next_length = 0;
      if (has_next) {
        next_length = mapping->get(next, 0, chars);
        if (next_length == 0) next_length = 1;
      }
      int current_length = i + char_length + next_length;
      while (stream.HasMore()) {
        current = stream.GetNext();
        overflows |= ToUpperOverflows(current);
        // NOTE: we use 0 as the next character here because, while
        // the next character may affect what a character converts to,
        // it does not in any case affect the length of what it convert
        // to.
        int char_length = mapping->get(current, 0, chars);
        if (char_length == 0) char_length = 1;
        current_length += char_length;
        if (current_length > String::kMaxLength) {
          AllowHeapAllocation allocate_error_and_return;
          THROW_NEW_ERROR_RETURN_FAILURE(isolate,
                                         NewInvalidStringLengthError());
        }
      }
      // Try again with the real length.  Return signed if we need
      // to allocate a two-byte string for to uppercase.
      return (overflows && !ignore_overflow) ? Smi::FromInt(-current_length)
                                             : Smi::FromInt(current_length);
    } else {
      for (int j = 0; j < char_length; j++) {
        result->Set(i, chars[j]);
        i++;
      }
      has_changed_character = true;
    }
    current = next;
  }
  if (has_changed_character) {
    return result;
  } else {
    // If we didn't actually change anything in doing the conversion
    // we simple return the result and let the converted string
    // become garbage; there is no reason to keep two identical strings
    // alive.
    return string;
  }
}


static const uintptr_t kOneInEveryByte = kUintptrAllBitsSet / 0xFF;
static const uintptr_t kAsciiMask = kOneInEveryByte << 7;

// Given a word and two range boundaries returns a word with high bit
// set in every byte iff the corresponding input byte was strictly in
// the range (m, n). All the other bits in the result are cleared.
// This function is only useful when it can be inlined and the
// boundaries are statically known.
// Requires: all bytes in the input word and the boundaries must be
// ASCII (less than 0x7F).
static inline uintptr_t AsciiRangeMask(uintptr_t w, char m, char n) {
  // Use strict inequalities since in edge cases the function could be
  // further simplified.
  DCHECK(0 < m && m < n);
  // Has high bit set in every w byte less than n.
  uintptr_t tmp1 = kOneInEveryByte * (0x7F + n) - w;
  // Has high bit set in every w byte greater than m.
  uintptr_t tmp2 = w + kOneInEveryByte * (0x7F - m);
  return (tmp1 & tmp2 & (kOneInEveryByte * 0x80));
}


#ifdef DEBUG
static bool CheckFastAsciiConvert(char* dst, const char* src, int length,
                                  bool changed, bool is_to_lower) {
  bool expected_changed = false;
  for (int i = 0; i < length; i++) {
    if (dst[i] == src[i]) continue;
    expected_changed = true;
    if (is_to_lower) {
      DCHECK('A' <= src[i] && src[i] <= 'Z');
      DCHECK(dst[i] == src[i] + ('a' - 'A'));
    } else {
      DCHECK('a' <= src[i] && src[i] <= 'z');
      DCHECK(dst[i] == src[i] - ('a' - 'A'));
    }
  }
  return (expected_changed == changed);
}
#endif


template <class Converter>
static bool FastAsciiConvert(char* dst, const char* src, int length,
                             bool* changed_out) {
#ifdef DEBUG
  char* saved_dst = dst;
  const char* saved_src = src;
#endif
  DisallowHeapAllocation no_gc;
  // We rely on the distance between upper and lower case letters
  // being a known power of 2.
  DCHECK('a' - 'A' == (1 << 5));
  // Boundaries for the range of input characters than require conversion.
  static const char lo = Converter::kIsToLower ? 'A' - 1 : 'a' - 1;
  static const char hi = Converter::kIsToLower ? 'Z' + 1 : 'z' + 1;
  bool changed = false;
  uintptr_t or_acc = 0;
  const char* const limit = src + length;

  // dst is newly allocated and always aligned.
  DCHECK(IsAligned(reinterpret_cast<intptr_t>(dst), sizeof(uintptr_t)));
  // Only attempt processing one word at a time if src is also aligned.
  if (IsAligned(reinterpret_cast<intptr_t>(src), sizeof(uintptr_t))) {
    // Process the prefix of the input that requires no conversion one aligned
    // (machine) word at a time.
    while (src <= limit - sizeof(uintptr_t)) {
      const uintptr_t w = *reinterpret_cast<const uintptr_t*>(src);
      or_acc |= w;
      if (AsciiRangeMask(w, lo, hi) != 0) {
        changed = true;
        break;
      }
      *reinterpret_cast<uintptr_t*>(dst) = w;
      src += sizeof(uintptr_t);
      dst += sizeof(uintptr_t);
    }
    // Process the remainder of the input performing conversion when
    // required one word at a time.
    while (src <= limit - sizeof(uintptr_t)) {
      const uintptr_t w = *reinterpret_cast<const uintptr_t*>(src);
      or_acc |= w;
      uintptr_t m = AsciiRangeMask(w, lo, hi);
      // The mask has high (7th) bit set in every byte that needs
      // conversion and we know that the distance between cases is
      // 1 << 5.
      *reinterpret_cast<uintptr_t*>(dst) = w ^ (m >> 2);
      src += sizeof(uintptr_t);
      dst += sizeof(uintptr_t);
    }
  }
  // Process the last few bytes of the input (or the whole input if
  // unaligned access is not supported).
  while (src < limit) {
    char c = *src;
    or_acc |= c;
    if (lo < c && c < hi) {
      c ^= (1 << 5);
      changed = true;
    }
    *dst = c;
    ++src;
    ++dst;
  }

  if ((or_acc & kAsciiMask) != 0) return false;

  DCHECK(CheckFastAsciiConvert(saved_dst, saved_src, length, changed,
                               Converter::kIsToLower));

  *changed_out = changed;
  return true;
}


template <class Converter>
MUST_USE_RESULT static Object* ConvertCase(
    Handle<String> s, Isolate* isolate,
    unibrow::Mapping<Converter, 128>* mapping) {
  s = String::Flatten(s);
  int length = s->length();
  // Assume that the string is not empty; we need this assumption later
  if (length == 0) return *s;

  // Simpler handling of ASCII strings.
  //
  // NOTE: This assumes that the upper/lower case of an ASCII
  // character is also ASCII.  This is currently the case, but it
  // might break in the future if we implement more context and locale
  // dependent upper/lower conversions.
  if (s->IsOneByteRepresentationUnderneath()) {
    // Same length as input.
    Handle<SeqOneByteString> result =
        isolate->factory()->NewRawOneByteString(length).ToHandleChecked();
    DisallowHeapAllocation no_gc;
    String::FlatContent flat_content = s->GetFlatContent();
    DCHECK(flat_content.IsFlat());
    bool has_changed_character = false;
    bool is_ascii = FastAsciiConvert<Converter>(
        reinterpret_cast<char*>(result->GetChars()),
        reinterpret_cast<const char*>(flat_content.ToOneByteVector().start()),
        length, &has_changed_character);
    // If not ASCII, we discard the result and take the 2 byte path.
    if (is_ascii) return has_changed_character ? *result : *s;
  }

  Handle<SeqString> result;  // Same length as input.
  if (s->IsOneByteRepresentation()) {
    result = isolate->factory()->NewRawOneByteString(length).ToHandleChecked();
  } else {
    result = isolate->factory()->NewRawTwoByteString(length).ToHandleChecked();
  }

  Object* answer = ConvertCaseHelper(isolate, *s, *result, length, mapping);
  if (answer->IsException() || answer->IsString()) return answer;

  DCHECK(answer->IsSmi());
  length = Smi::cast(answer)->value();
  if (s->IsOneByteRepresentation() && length > 0) {
    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
        isolate, result, isolate->factory()->NewRawOneByteString(length));
  } else {
    if (length < 0) length = -length;
    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
        isolate, result, isolate->factory()->NewRawTwoByteString(length));
  }
  return ConvertCaseHelper(isolate, *s, *result, length, mapping);
}


RUNTIME_FUNCTION(Runtime_StringToLowerCase) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 1);
  CONVERT_ARG_HANDLE_CHECKED(String, s, 0);
  return ConvertCase(s, isolate, isolate->runtime_state()->to_lower_mapping());
}


RUNTIME_FUNCTION(Runtime_StringToUpperCase) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 1);
  CONVERT_ARG_HANDLE_CHECKED(String, s, 0);
  return ConvertCase(s, isolate, isolate->runtime_state()->to_upper_mapping());
}


RUNTIME_FUNCTION(Runtime_StringTrim) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 3);

  CONVERT_ARG_HANDLE_CHECKED(String, string, 0);
  CONVERT_BOOLEAN_ARG_CHECKED(trimLeft, 1);
  CONVERT_BOOLEAN_ARG_CHECKED(trimRight, 2);

  string = String::Flatten(string);
  int length = string->length();

  int left = 0;
  UnicodeCache* unicode_cache = isolate->unicode_cache();
  if (trimLeft) {
    while (left < length &&
           unicode_cache->IsWhiteSpaceOrLineTerminator(string->Get(left))) {
      left++;
    }
  }

  int right = length;
  if (trimRight) {
    while (
        right > left &&
        unicode_cache->IsWhiteSpaceOrLineTerminator(string->Get(right - 1))) {
      right--;
    }
  }

  return *isolate->factory()->NewSubString(string, left, right);
}


RUNTIME_FUNCTION(Runtime_TruncateString) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 2);
  CONVERT_ARG_HANDLE_CHECKED(SeqString, string, 0);
  CONVERT_INT32_ARG_CHECKED(new_length, 1);
  RUNTIME_ASSERT(new_length >= 0);
  return *SeqString::Truncate(string, new_length);
}


RUNTIME_FUNCTION(Runtime_NewString) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 2);
  CONVERT_INT32_ARG_CHECKED(length, 0);
  CONVERT_BOOLEAN_ARG_CHECKED(is_one_byte, 1);
  if (length == 0) return isolate->heap()->empty_string();
  Handle<String> result;
  if (is_one_byte) {
    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
        isolate, result, isolate->factory()->NewRawOneByteString(length));
  } else {
    ASSIGN_RETURN_FAILURE_ON_EXCEPTION(
        isolate, result, isolate->factory()->NewRawTwoByteString(length));
  }
  return *result;
}


RUNTIME_FUNCTION(Runtime_StringEquals) {
  HandleScope handle_scope(isolate);
  DCHECK(args.length() == 2);

  CONVERT_ARG_HANDLE_CHECKED(String, x, 0);
  CONVERT_ARG_HANDLE_CHECKED(String, y, 1);

  bool not_equal = !String::Equals(x, y);
  // This is slightly convoluted because the value that signifies
  // equality is 0 and inequality is 1 so we have to negate the result
  // from String::Equals.
  DCHECK(not_equal == 0 || not_equal == 1);
  STATIC_ASSERT(EQUAL == 0);
  STATIC_ASSERT(NOT_EQUAL == 1);
  return Smi::FromInt(not_equal);
}


RUNTIME_FUNCTION(Runtime_FlattenString) {
  HandleScope scope(isolate);
  DCHECK(args.length() == 1);
  CONVERT_ARG_HANDLE_CHECKED(String, str, 0);
  return *String::Flatten(str);
}


RUNTIME_FUNCTION(RuntimeReference_StringCharFromCode) {
  SealHandleScope shs(isolate);
  return __RT_impl_Runtime_CharFromCode(args, isolate);
}


RUNTIME_FUNCTION(RuntimeReference_StringCharAt) {
  SealHandleScope shs(isolate);
  DCHECK(args.length() == 2);
  if (!args[0]->IsString()) return Smi::FromInt(0);
  if (!args[1]->IsNumber()) return Smi::FromInt(0);
  if (std::isinf(args.number_at(1))) return isolate->heap()->empty_string();
  Object* code = __RT_impl_Runtime_StringCharCodeAtRT(args, isolate);
  if (code->IsNaN()) return isolate->heap()->empty_string();
  return __RT_impl_Runtime_CharFromCode(Arguments(1, &code), isolate);
}


RUNTIME_FUNCTION(RuntimeReference_OneByteSeqStringSetChar) {
  SealHandleScope shs(isolate);
  DCHECK(args.length() == 3);
  CONVERT_INT32_ARG_CHECKED(index, 0);
  CONVERT_INT32_ARG_CHECKED(value, 1);
  CONVERT_ARG_CHECKED(SeqOneByteString, string, 2);
  string->SeqOneByteStringSet(index, value);
  return string;
}


RUNTIME_FUNCTION(RuntimeReference_TwoByteSeqStringSetChar) {
  SealHandleScope shs(isolate);
  DCHECK(args.length() == 3);
  CONVERT_INT32_ARG_CHECKED(index, 0);
  CONVERT_INT32_ARG_CHECKED(value, 1);
  CONVERT_ARG_CHECKED(SeqTwoByteString, string, 2);
  string->SeqTwoByteStringSet(index, value);
  return string;
}


RUNTIME_FUNCTION(RuntimeReference_StringCompare) {
  SealHandleScope shs(isolate);
  return __RT_impl_Runtime_StringCompare(args, isolate);
}


RUNTIME_FUNCTION(RuntimeReference_StringCharCodeAt) {
  SealHandleScope shs(isolate);
  DCHECK(args.length() == 2);
  if (!args[0]->IsString()) return isolate->heap()->undefined_value();
  if (!args[1]->IsNumber()) return isolate->heap()->undefined_value();
  if (std::isinf(args.number_at(1))) return isolate->heap()->nan_value();
  return __RT_impl_Runtime_StringCharCodeAtRT(args, isolate);
}


RUNTIME_FUNCTION(RuntimeReference_SubString) {
  SealHandleScope shs(isolate);
  return __RT_impl_Runtime_SubString(args, isolate);
}


RUNTIME_FUNCTION(RuntimeReference_StringAdd) {
  SealHandleScope shs(isolate);
  return __RT_impl_Runtime_StringAdd(args, isolate);
}


RUNTIME_FUNCTION(RuntimeReference_IsStringWrapperSafeForDefaultValueOf) {
  UNIMPLEMENTED();
  return NULL;
}
}
}  // namespace v8::internal