Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(259)

Side by Side Diff: src/scanner.cc

Issue 198583003: Convert scanner buffers to use standard character types. (Closed) Base URL: https://v8.googlecode.com/svn/branches/bleeding_edge
Patch Set: Created 6 years, 9 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View unified diff | Download patch | Annotate | Revision Log
« no previous file with comments | « src/scanner.h ('k') | no next file » | no next file with comments »
Toggle Intra-line Diffs ('i') | Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
OLDNEW
1 // Copyright 2011 the V8 project authors. All rights reserved. 1 // Copyright 2011 the V8 project authors. All rights reserved.
2 // Redistribution and use in source and binary forms, with or without 2 // Redistribution and use in source and binary forms, with or without
3 // modification, are permitted provided that the following conditions are 3 // modification, are permitted provided that the following conditions are
4 // met: 4 // met:
5 // 5 //
6 // * Redistributions of source code must retain the above copyright 6 // * Redistributions of source code must retain the above copyright
7 // notice, this list of conditions and the following disclaimer. 7 // notice, this list of conditions and the following disclaimer.
8 // * Redistributions in binary form must reproduce the above 8 // * Redistributions in binary form must reproduce the above
9 // copyright notice, this list of conditions and the following 9 // copyright notice, this list of conditions and the following
10 // disclaimer in the documentation and/or other materials provided 10 // disclaimer in the documentation and/or other materials provided
(...skipping 891 matching lines...) Expand 10 before | Expand all | Expand 10 after
902 KEYWORD_GROUP('v') \ 902 KEYWORD_GROUP('v') \
903 KEYWORD("var", Token::VAR) \ 903 KEYWORD("var", Token::VAR) \
904 KEYWORD("void", Token::VOID) \ 904 KEYWORD("void", Token::VOID) \
905 KEYWORD_GROUP('w') \ 905 KEYWORD_GROUP('w') \
906 KEYWORD("while", Token::WHILE) \ 906 KEYWORD("while", Token::WHILE) \
907 KEYWORD("with", Token::WITH) \ 907 KEYWORD("with", Token::WITH) \
908 KEYWORD_GROUP('y') \ 908 KEYWORD_GROUP('y') \
909 KEYWORD("yield", Token::YIELD) 909 KEYWORD("yield", Token::YIELD)
910 910
911 911
912 static Token::Value KeywordOrIdentifierToken(const char* input, 912 static Token::Value KeywordOrIdentifierToken(const uint8_t* input,
913 int input_length, 913 int input_length,
914 bool harmony_scoping, 914 bool harmony_scoping,
915 bool harmony_modules) { 915 bool harmony_modules) {
916 ASSERT(input_length >= 1); 916 ASSERT(input_length >= 1);
917 const int kMinLength = 2; 917 const int kMinLength = 2;
918 const int kMaxLength = 10; 918 const int kMaxLength = 10;
919 if (input_length < kMinLength || input_length > kMaxLength) { 919 if (input_length < kMinLength || input_length > kMaxLength) {
920 return Token::IDENTIFIER; 920 return Token::IDENTIFIER;
921 } 921 }
922 switch (input[0]) { 922 switch (input[0]) {
(...skipping 55 matching lines...) Expand 10 before | Expand all | Expand 10 after
978 AddLiteralChar(next_char); 978 AddLiteralChar(next_char);
979 continue; 979 continue;
980 } 980 }
981 // Fallthrough if no longer able to complete keyword. 981 // Fallthrough if no longer able to complete keyword.
982 return ScanIdentifierSuffix(&literal); 982 return ScanIdentifierSuffix(&literal);
983 } 983 }
984 984
985 literal.Complete(); 985 literal.Complete();
986 986
987 if (next_.literal_chars->is_one_byte()) { 987 if (next_.literal_chars->is_one_byte()) {
988 Vector<const char> chars = next_.literal_chars->one_byte_literal(); 988 Vector<const uint8_t> chars = next_.literal_chars->one_byte_literal();
989 return KeywordOrIdentifierToken(chars.start(), 989 return KeywordOrIdentifierToken(chars.start(),
990 chars.length(), 990 chars.length(),
991 harmony_scoping_, 991 harmony_scoping_,
992 harmony_modules_); 992 harmony_modules_);
993 } 993 }
994 994
995 return Token::IDENTIFIER; 995 return Token::IDENTIFIER;
996 } 996 }
997 997
998 998
(...skipping 117 matching lines...) Expand 10 before | Expand all | Expand 10 after
1116 } 1116 }
1117 1117
1118 1118
1119 Handle<String> Scanner::AllocateNextLiteralString(Isolate* isolate, 1119 Handle<String> Scanner::AllocateNextLiteralString(Isolate* isolate,
1120 PretenureFlag tenured) { 1120 PretenureFlag tenured) {
1121 if (is_next_literal_one_byte()) { 1121 if (is_next_literal_one_byte()) {
1122 return isolate->factory()->NewStringFromOneByte( 1122 return isolate->factory()->NewStringFromOneByte(
1123 Vector<const uint8_t>::cast(next_literal_one_byte_string()), tenured); 1123 Vector<const uint8_t>::cast(next_literal_one_byte_string()), tenured);
1124 } else { 1124 } else {
1125 return isolate->factory()->NewStringFromTwoByte( 1125 return isolate->factory()->NewStringFromTwoByte(
1126 next_literal_utf16_string(), tenured); 1126 next_literal_two_byte_string(), tenured);
1127 } 1127 }
1128 } 1128 }
1129 1129
1130 1130
1131 Handle<String> Scanner::AllocateInternalizedString(Isolate* isolate) { 1131 Handle<String> Scanner::AllocateInternalizedString(Isolate* isolate) {
1132 if (is_literal_one_byte()) { 1132 if (is_literal_one_byte()) {
1133 return isolate->factory()->InternalizeOneByteString( 1133 return isolate->factory()->InternalizeOneByteString(
1134 Vector<const uint8_t>::cast(literal_one_byte_string())); 1134 literal_one_byte_string());
1135 } else { 1135 } else {
1136 return isolate->factory()->InternalizeTwoByteString( 1136 return isolate->factory()->InternalizeTwoByteString(
1137 literal_utf16_string()); 1137 literal_two_byte_string());
1138 } 1138 }
1139 } 1139 }
1140 1140
1141 1141
1142 double Scanner::DoubleValue() { 1142 double Scanner::DoubleValue() {
1143 ASSERT(is_literal_one_byte()); 1143 ASSERT(is_literal_one_byte());
1144 return StringToDouble( 1144 return StringToDouble(
1145 unicode_cache_, literal_one_byte_string(), 1145 unicode_cache_, Vector<const char>::cast(literal_one_byte_string()),
1146 ALLOW_HEX | ALLOW_OCTAL | ALLOW_IMPLICIT_OCTAL | ALLOW_BINARY); 1146 ALLOW_HEX | ALLOW_OCTAL | ALLOW_IMPLICIT_OCTAL | ALLOW_BINARY);
1147 } 1147 }
1148 1148
1149 1149
1150 int Scanner::FindNumber(DuplicateFinder* finder, int value) { 1150 int Scanner::FindNumber(DuplicateFinder* finder, int value) {
1151 return finder->AddNumber(literal_one_byte_string(), value); 1151 return finder->AddNumber(literal_one_byte_string(), value);
1152 } 1152 }
1153 1153
1154 1154
1155 int Scanner::FindSymbol(DuplicateFinder* finder, int value) { 1155 int Scanner::FindSymbol(DuplicateFinder* finder, int value) {
1156 if (is_literal_one_byte()) { 1156 if (is_literal_one_byte()) {
1157 return finder->AddAsciiSymbol(literal_one_byte_string(), value); 1157 return finder->AddOneByteSymbol(literal_one_byte_string(), value);
1158 } 1158 }
1159 return finder->AddUtf16Symbol(literal_utf16_string(), value); 1159 return finder->AddTwoByteSymbol(literal_two_byte_string(), value);
1160 } 1160 }
1161 1161
1162 1162
1163 void Scanner::LogSymbol(ParserRecorder* log, int position) { 1163 void Scanner::LogSymbol(ParserRecorder* log, int position) {
1164 if (is_literal_one_byte()) { 1164 if (is_literal_one_byte()) {
1165 log->LogAsciiSymbol(position, literal_one_byte_string()); 1165 log->LogOneByteSymbol(position, literal_one_byte_string());
1166 } else { 1166 } else {
1167 log->LogUtf16Symbol(position, literal_utf16_string()); 1167 log->LogTwoByteSymbol(position, literal_two_byte_string());
1168 } 1168 }
1169 } 1169 }
1170 1170
1171 1171
1172 int DuplicateFinder::AddAsciiSymbol(Vector<const char> key, int value) { 1172 int DuplicateFinder::AddOneByteSymbol(Vector<const uint8_t> key, int value) {
1173 return AddSymbol(Vector<const byte>::cast(key), true, value); 1173 return AddSymbol(key, true, value);
1174 } 1174 }
1175 1175
1176 1176
1177 int DuplicateFinder::AddUtf16Symbol(Vector<const uint16_t> key, int value) { 1177 int DuplicateFinder::AddTwoByteSymbol(Vector<const uint16_t> key, int value) {
1178 return AddSymbol(Vector<const byte>::cast(key), false, value); 1178 return AddSymbol(Vector<const uint8_t>::cast(key), false, value);
1179 } 1179 }
1180 1180
1181 1181
1182 int DuplicateFinder::AddSymbol(Vector<const byte> key, 1182 int DuplicateFinder::AddSymbol(Vector<const uint8_t> key,
1183 bool is_one_byte, 1183 bool is_one_byte,
1184 int value) { 1184 int value) {
1185 uint32_t hash = Hash(key, is_one_byte); 1185 uint32_t hash = Hash(key, is_one_byte);
1186 byte* encoding = BackupKey(key, is_one_byte); 1186 byte* encoding = BackupKey(key, is_one_byte);
1187 HashMap::Entry* entry = map_.Lookup(encoding, hash, true); 1187 HashMap::Entry* entry = map_.Lookup(encoding, hash, true);
1188 int old_value = static_cast<int>(reinterpret_cast<intptr_t>(entry->value)); 1188 int old_value = static_cast<int>(reinterpret_cast<intptr_t>(entry->value));
1189 entry->value = 1189 entry->value =
1190 reinterpret_cast<void*>(static_cast<intptr_t>(value | old_value)); 1190 reinterpret_cast<void*>(static_cast<intptr_t>(value | old_value));
1191 return old_value; 1191 return old_value;
1192 } 1192 }
1193 1193
1194 1194
1195 int DuplicateFinder::AddNumber(Vector<const char> key, int value) { 1195 int DuplicateFinder::AddNumber(Vector<const uint8_t> key, int value) {
1196 ASSERT(key.length() > 0); 1196 ASSERT(key.length() > 0);
1197 // Quick check for already being in canonical form. 1197 // Quick check for already being in canonical form.
1198 if (IsNumberCanonical(key)) { 1198 if (IsNumberCanonical(key)) {
1199 return AddAsciiSymbol(key, value); 1199 return AddOneByteSymbol(key, value);
1200 } 1200 }
1201 1201
1202 int flags = ALLOW_HEX | ALLOW_OCTAL | ALLOW_IMPLICIT_OCTAL | ALLOW_BINARY; 1202 int flags = ALLOW_HEX | ALLOW_OCTAL | ALLOW_IMPLICIT_OCTAL | ALLOW_BINARY;
1203 double double_value = StringToDouble(unicode_constants_, key, flags, 0.0); 1203 double double_value = StringToDouble(
1204 unicode_constants_, Vector<const char>::cast(key), flags, 0.0);
1204 int length; 1205 int length;
1205 const char* string; 1206 const char* string;
1206 if (!std::isfinite(double_value)) { 1207 if (!std::isfinite(double_value)) {
1207 string = "Infinity"; 1208 string = "Infinity";
1208 length = 8; // strlen("Infinity"); 1209 length = 8; // strlen("Infinity");
1209 } else { 1210 } else {
1210 string = DoubleToCString(double_value, 1211 string = DoubleToCString(double_value,
1211 Vector<char>(number_buffer_, kBufferSize)); 1212 Vector<char>(number_buffer_, kBufferSize));
1212 length = StrLength(string); 1213 length = StrLength(string);
1213 } 1214 }
1214 return AddSymbol(Vector<const byte>(reinterpret_cast<const byte*>(string), 1215 return AddSymbol(Vector<const byte>(reinterpret_cast<const byte*>(string),
1215 length), true, value); 1216 length), true, value);
1216 } 1217 }
1217 1218
1218 1219
1219 bool DuplicateFinder::IsNumberCanonical(Vector<const char> number) { 1220 bool DuplicateFinder::IsNumberCanonical(Vector<const uint8_t> number) {
1220 // Test for a safe approximation of number literals that are already 1221 // Test for a safe approximation of number literals that are already
1221 // in canonical form: max 15 digits, no leading zeroes, except an 1222 // in canonical form: max 15 digits, no leading zeroes, except an
1222 // integer part that is a single zero, and no trailing zeros below 1223 // integer part that is a single zero, and no trailing zeros below
1223 // the decimal point. 1224 // the decimal point.
1224 int pos = 0; 1225 int pos = 0;
1225 int length = number.length(); 1226 int length = number.length();
1226 if (number.length() > 15) return false; 1227 if (number.length() > 15) return false;
1227 if (number[pos] == '0') { 1228 if (number[pos] == '0') {
1228 pos++; 1229 pos++;
1229 } else { 1230 } else {
1230 while (pos < length && 1231 while (pos < length &&
1231 static_cast<unsigned>(number[pos] - '0') <= ('9' - '0')) pos++; 1232 static_cast<unsigned>(number[pos] - '0') <= ('9' - '0')) pos++;
1232 } 1233 }
1233 if (length == pos) return true; 1234 if (length == pos) return true;
1234 if (number[pos] != '.') return false; 1235 if (number[pos] != '.') return false;
1235 pos++; 1236 pos++;
1236 bool invalid_last_digit = true; 1237 bool invalid_last_digit = true;
1237 while (pos < length) { 1238 while (pos < length) {
1238 byte digit = number[pos] - '0'; 1239 uint8_t digit = number[pos] - '0';
1239 if (digit > '9' - '0') return false; 1240 if (digit > '9' - '0') return false;
1240 invalid_last_digit = (digit == 0); 1241 invalid_last_digit = (digit == 0);
1241 pos++; 1242 pos++;
1242 } 1243 }
1243 return !invalid_last_digit; 1244 return !invalid_last_digit;
1244 } 1245 }
1245 1246
1246 1247
1247 uint32_t DuplicateFinder::Hash(Vector<const byte> key, bool is_one_byte) { 1248 uint32_t DuplicateFinder::Hash(Vector<const uint8_t> key, bool is_one_byte) {
1248 // Primitive hash function, almost identical to the one used 1249 // Primitive hash function, almost identical to the one used
1249 // for strings (except that it's seeded by the length and ASCII-ness). 1250 // for strings (except that it's seeded by the length and ASCII-ness).
1250 int length = key.length(); 1251 int length = key.length();
1251 uint32_t hash = (length << 1) | (is_one_byte ? 1 : 0) ; 1252 uint32_t hash = (length << 1) | (is_one_byte ? 1 : 0) ;
1252 for (int i = 0; i < length; i++) { 1253 for (int i = 0; i < length; i++) {
1253 uint32_t c = key[i]; 1254 uint32_t c = key[i];
1254 hash = (hash + c) * 1025; 1255 hash = (hash + c) * 1025;
1255 hash ^= (hash >> 6); 1256 hash ^= (hash >> 6);
1256 } 1257 }
1257 return hash; 1258 return hash;
(...skipping 15 matching lines...) Expand all
1273 if (c1 != *s2) return false; 1274 if (c1 != *s2) return false;
1274 length_one_byte_field = (length_one_byte_field << 7) | (c1 & 0x7f); 1275 length_one_byte_field = (length_one_byte_field << 7) | (c1 & 0x7f);
1275 s1++; 1276 s1++;
1276 s2++; 1277 s2++;
1277 } while ((c1 & 0x80) != 0); 1278 } while ((c1 & 0x80) != 0);
1278 int length = static_cast<int>(length_one_byte_field >> 1); 1279 int length = static_cast<int>(length_one_byte_field >> 1);
1279 return memcmp(s1, s2, length) == 0; 1280 return memcmp(s1, s2, length) == 0;
1280 } 1281 }
1281 1282
1282 1283
1283 byte* DuplicateFinder::BackupKey(Vector<const byte> bytes, 1284 byte* DuplicateFinder::BackupKey(Vector<const uint8_t> bytes,
1284 bool is_one_byte) { 1285 bool is_one_byte) {
1285 uint32_t one_byte_length = (bytes.length() << 1) | (is_one_byte ? 1 : 0); 1286 uint32_t one_byte_length = (bytes.length() << 1) | (is_one_byte ? 1 : 0);
1286 backing_store_.StartSequence(); 1287 backing_store_.StartSequence();
1287 // Emit one_byte_length as base-128 encoded number, with the 7th bit set 1288 // Emit one_byte_length as base-128 encoded number, with the 7th bit set
1288 // on the byte of every heptet except the last, least significant, one. 1289 // on the byte of every heptet except the last, least significant, one.
1289 if (one_byte_length >= (1 << 7)) { 1290 if (one_byte_length >= (1 << 7)) {
1290 if (one_byte_length >= (1 << 14)) { 1291 if (one_byte_length >= (1 << 14)) {
1291 if (one_byte_length >= (1 << 21)) { 1292 if (one_byte_length >= (1 << 21)) {
1292 if (one_byte_length >= (1 << 28)) { 1293 if (one_byte_length >= (1 << 28)) {
1293 backing_store_.Add(static_cast<byte>((one_byte_length >> 28) | 0x80)); 1294 backing_store_.Add(
1295 static_cast<uint8_t>((one_byte_length >> 28) | 0x80));
1294 } 1296 }
1295 backing_store_.Add(static_cast<byte>((one_byte_length >> 21) | 0x80u)); 1297 backing_store_.Add(
1298 static_cast<uint8_t>((one_byte_length >> 21) | 0x80u));
1296 } 1299 }
1297 backing_store_.Add(static_cast<byte>((one_byte_length >> 14) | 0x80u)); 1300 backing_store_.Add(
1301 static_cast<uint8_t>((one_byte_length >> 14) | 0x80u));
1298 } 1302 }
1299 backing_store_.Add(static_cast<byte>((one_byte_length >> 7) | 0x80u)); 1303 backing_store_.Add(static_cast<uint8_t>((one_byte_length >> 7) | 0x80u));
1300 } 1304 }
1301 backing_store_.Add(static_cast<byte>(one_byte_length & 0x7f)); 1305 backing_store_.Add(static_cast<uint8_t>(one_byte_length & 0x7f));
1302 1306
1303 backing_store_.AddBlock(bytes); 1307 backing_store_.AddBlock(bytes);
1304 return backing_store_.EndSequence().start(); 1308 return backing_store_.EndSequence().start();
1305 } 1309 }
1306 1310
1307 } } // namespace v8::internal 1311 } } // namespace v8::internal
OLDNEW
« no previous file with comments | « src/scanner.h ('k') | no next file » | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698