diff options
Diffstat (limited to 'base/strings/string_util.cc')
| -rw-r--r-- | base/strings/string_util.cc | 629 |
1 files changed, 290 insertions, 339 deletions
diff --git a/base/strings/string_util.cc b/base/strings/string_util.cc index 738d32e045..e8000abd40 100644 --- a/base/strings/string_util.cc +++ b/base/strings/string_util.cc @@ -8,6 +8,7 @@ #include <errno.h> #include <math.h> #include <stdarg.h> +#include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string.h> @@ -16,10 +17,11 @@ #include <wctype.h> #include <algorithm> +#include <limits> #include <vector> -#include "base/basictypes.h" #include "base/logging.h" +#include "base/macros.h" #include "base/memory/singleton.h" #include "base/strings/string_split.h" #include "base/strings/utf_string_conversion_utils.h" @@ -27,9 +29,7 @@ #include "base/third_party/icu/icu_utf.h" #include "build/build_config.h" -// Remove when this entire file is in the base namespace. -using base::char16; -using base::string16; +namespace base { namespace { @@ -80,13 +80,13 @@ template<typename T> inline T* AlignToMachineWord(T* pointer) { } template<size_t size, typename CharacterType> struct NonASCIIMask; -template<> struct NonASCIIMask<4, base::char16> { +template<> struct NonASCIIMask<4, char16> { static inline uint32_t value() { return 0xFF80FF80U; } }; template<> struct NonASCIIMask<4, char> { static inline uint32_t value() { return 0x80808080U; } }; -template<> struct NonASCIIMask<8, base::char16> { +template<> struct NonASCIIMask<8, char16> { static inline uint64_t value() { return 0xFF80FF80FF80FF80ULL; } }; template<> struct NonASCIIMask<8, char> { @@ -103,8 +103,6 @@ template<> struct NonASCIIMask<8, wchar_t> { } // namespace -namespace base { - bool IsWprintfFormatPortable(const wchar_t* format) { for (const wchar_t* position = format; *position != '\0'; ++position) { if (*position == '%') { @@ -140,6 +138,91 @@ bool IsWprintfFormatPortable(const wchar_t* format) { return true; } +namespace { + +template<typename StringType> +StringType ToLowerASCIIImpl(BasicStringPiece<StringType> str) { + StringType ret; + ret.reserve(str.size()); + for (size_t i = 0; i < str.size(); i++) + ret.push_back(ToLowerASCII(str[i])); + return ret; +} + +template<typename StringType> +StringType ToUpperASCIIImpl(BasicStringPiece<StringType> str) { + StringType ret; + ret.reserve(str.size()); + for (size_t i = 0; i < str.size(); i++) + ret.push_back(ToUpperASCII(str[i])); + return ret; +} + +} // namespace + +std::string ToLowerASCII(StringPiece str) { + return ToLowerASCIIImpl<std::string>(str); +} + +string16 ToLowerASCII(StringPiece16 str) { + return ToLowerASCIIImpl<string16>(str); +} + +std::string ToUpperASCII(StringPiece str) { + return ToUpperASCIIImpl<std::string>(str); +} + +string16 ToUpperASCII(StringPiece16 str) { + return ToUpperASCIIImpl<string16>(str); +} + +template<class StringType> +int CompareCaseInsensitiveASCIIT(BasicStringPiece<StringType> a, + BasicStringPiece<StringType> b) { + // Find the first characters that aren't equal and compare them. If the end + // of one of the strings is found before a nonequal character, the lengths + // of the strings are compared. + size_t i = 0; + while (i < a.length() && i < b.length()) { + typename StringType::value_type lower_a = ToLowerASCII(a[i]); + typename StringType::value_type lower_b = ToLowerASCII(b[i]); + if (lower_a < lower_b) + return -1; + if (lower_a > lower_b) + return 1; + i++; + } + + // End of one string hit before finding a different character. Expect the + // common case to be "strings equal" at this point so check that first. + if (a.length() == b.length()) + return 0; + + if (a.length() < b.length()) + return -1; + return 1; +} + +int CompareCaseInsensitiveASCII(StringPiece a, StringPiece b) { + return CompareCaseInsensitiveASCIIT<std::string>(a, b); +} + +int CompareCaseInsensitiveASCII(StringPiece16 a, StringPiece16 b) { + return CompareCaseInsensitiveASCIIT<string16>(a, b); +} + +bool EqualsCaseInsensitiveASCII(StringPiece a, StringPiece b) { + if (a.length() != b.length()) + return false; + return CompareCaseInsensitiveASCIIT<std::string>(a, b) == 0; +} + +bool EqualsCaseInsensitiveASCII(StringPiece16 a, StringPiece16 b) { + if (a.length() != b.length()) + return false; + return CompareCaseInsensitiveASCIIT<string16>(a, b) == 0; +} + const std::string& EmptyString() { return EmptyStrings::GetInstance()->s; } @@ -169,27 +252,27 @@ bool ReplaceCharsT(const STR& input, } bool ReplaceChars(const string16& input, - const base::StringPiece16& replace_chars, + const StringPiece16& replace_chars, const string16& replace_with, string16* output) { return ReplaceCharsT(input, replace_chars.as_string(), replace_with, output); } bool ReplaceChars(const std::string& input, - const base::StringPiece& replace_chars, + const StringPiece& replace_chars, const std::string& replace_with, std::string* output) { return ReplaceCharsT(input, replace_chars.as_string(), replace_with, output); } bool RemoveChars(const string16& input, - const base::StringPiece16& remove_chars, + const StringPiece16& remove_chars, string16* output) { return ReplaceChars(input, remove_chars.as_string(), string16(), output); } bool RemoveChars(const std::string& input, - const base::StringPiece& remove_chars, + const StringPiece& remove_chars, std::string* output) { return ReplaceChars(input, remove_chars.as_string(), std::string(), output); } @@ -231,13 +314,13 @@ TrimPositions TrimStringT(const Str& input, } bool TrimString(const string16& input, - base::StringPiece16 trim_chars, + StringPiece16 trim_chars, string16* output) { return TrimStringT(input, trim_chars, TRIM_ALL, output) != TRIM_NONE; } bool TrimString(const std::string& input, - base::StringPiece trim_chars, + StringPiece trim_chars, std::string* output) { return TrimStringT(input, trim_chars, TRIM_ALL, output) != TRIM_NONE; } @@ -254,13 +337,13 @@ BasicStringPiece<Str> TrimStringPieceT(BasicStringPiece<Str> input, } StringPiece16 TrimString(StringPiece16 input, - const base::StringPiece16& trim_chars, + const StringPiece16& trim_chars, TrimPositions positions) { return TrimStringPieceT(input, trim_chars, positions); } StringPiece TrimString(StringPiece input, - const base::StringPiece& trim_chars, + const StringPiece& trim_chars, TrimPositions positions) { return TrimStringPieceT(input, trim_chars, positions); } @@ -273,10 +356,11 @@ void TruncateUTF8ToByteSize(const std::string& input, *output = input; return; } - DCHECK_LE(byte_size, static_cast<uint32>(kint32max)); - // Note: This cast is necessary because CBU8_NEXT uses int32s. - int32 truncation_length = static_cast<int32>(byte_size); - int32 char_index = truncation_length - 1; + DCHECK_LE(byte_size, + static_cast<uint32_t>(std::numeric_limits<int32_t>::max())); + // Note: This cast is necessary because CBU8_NEXT uses int32_ts. + int32_t truncation_length = static_cast<int32_t>(byte_size); + int32_t char_index = truncation_length - 1; const char* data = input.data(); // Using CBU8, we will move backwards from the truncation point @@ -284,7 +368,7 @@ void TruncateUTF8ToByteSize(const std::string& input, // character. Once a full UTF8 character is found, we will // truncate the string to the end of that character. while (char_index >= 0) { - int32 prev = char_index; + int32_t prev = char_index; base_icu::UChar32 code_point = 0; CBU8_NEXT(data, char_index, truncation_length, code_point); if (!IsValidCharacter(code_point) || @@ -307,18 +391,19 @@ TrimPositions TrimWhitespace(const string16& input, return TrimStringT(input, StringPiece16(kWhitespaceUTF16), positions, output); } +StringPiece16 TrimWhitespace(StringPiece16 input, + TrimPositions positions) { + return TrimStringPieceT(input, StringPiece16(kWhitespaceUTF16), positions); +} + TrimPositions TrimWhitespaceASCII(const std::string& input, TrimPositions positions, std::string* output) { return TrimStringT(input, StringPiece(kWhitespaceASCII), positions, output); } -// This function is only for backward-compatibility. -// To be removed when all callers are updated. -TrimPositions TrimWhitespace(const std::string& input, - TrimPositions positions, - std::string* output) { - return TrimWhitespaceASCII(input, positions, output); +StringPiece TrimWhitespaceASCII(StringPiece input, TrimPositions positions) { + return TrimStringPieceT(input, StringPiece(kWhitespaceASCII), positions); } template<typename STR> @@ -334,7 +419,7 @@ STR CollapseWhitespaceT(const STR& text, int chars_written = 0; for (typename STR::const_iterator i(text.begin()); i != text.end(); ++i) { - if (IsWhitespace(*i)) { + if (IsUnicodeWhitespace(*i)) { if (!in_whitespace) { // Reduce all whitespace sequences to a single space. in_whitespace = true; @@ -433,11 +518,11 @@ bool IsStringASCII(const std::wstring& str) { bool IsStringUTF8(const StringPiece& str) { const char *src = str.data(); - int32 src_len = static_cast<int32>(str.length()); - int32 char_index = 0; + int32_t src_len = static_cast<int32_t>(str.length()); + int32_t char_index = 0; while (char_index < src_len) { - int32 code_point; + int32_t code_point; CBU8_NEXT(src, char_index, src_len, code_point); if (!IsValidCharacter(code_point)) return false; @@ -445,113 +530,141 @@ bool IsStringUTF8(const StringPiece& str) { return true; } -template<typename Iter> -static inline bool DoLowerCaseEqualsASCII(Iter a_begin, - Iter a_end, - const char* b) { - for (Iter it = a_begin; it != a_end; ++it, ++b) { - if (!*b || ToLowerASCII(*it) != *b) +// Implementation note: Normally this function will be called with a hardcoded +// constant for the lowercase_ascii parameter. Constructing a StringPiece from +// a C constant requires running strlen, so the result will be two passes +// through the buffers, one to file the length of lowercase_ascii, and one to +// compare each letter. +// +// This function could have taken a const char* to avoid this and only do one +// pass through the string. But the strlen is faster than the case-insensitive +// compares and lets us early-exit in the case that the strings are different +// lengths (will often be the case for non-matches). So whether one approach or +// the other will be faster depends on the case. +// +// The hardcoded strings are typically very short so it doesn't matter, and the +// string piece gives additional flexibility for the caller (doesn't have to be +// null terminated) so we choose the StringPiece route. +template<typename Str> +static inline bool DoLowerCaseEqualsASCII(BasicStringPiece<Str> str, + StringPiece lowercase_ascii) { + if (str.size() != lowercase_ascii.size()) + return false; + for (size_t i = 0; i < str.size(); i++) { + if (ToLowerASCII(str[i]) != lowercase_ascii[i]) return false; } - return *b == 0; + return true; } -// Front-ends for LowerCaseEqualsASCII. -bool LowerCaseEqualsASCII(const std::string& a, const char* b) { - return DoLowerCaseEqualsASCII(a.begin(), a.end(), b); +bool LowerCaseEqualsASCII(StringPiece str, StringPiece lowercase_ascii) { + return DoLowerCaseEqualsASCII<std::string>(str, lowercase_ascii); } -bool LowerCaseEqualsASCII(const string16& a, const char* b) { - return DoLowerCaseEqualsASCII(a.begin(), a.end(), b); +bool LowerCaseEqualsASCII(StringPiece16 str, StringPiece lowercase_ascii) { + return DoLowerCaseEqualsASCII<string16>(str, lowercase_ascii); } -bool LowerCaseEqualsASCII(std::string::const_iterator a_begin, - std::string::const_iterator a_end, - const char* b) { - return DoLowerCaseEqualsASCII(a_begin, a_end, b); +bool EqualsASCII(StringPiece16 str, StringPiece ascii) { + if (str.length() != ascii.length()) + return false; + return std::equal(ascii.begin(), ascii.end(), str.begin()); } -bool LowerCaseEqualsASCII(string16::const_iterator a_begin, - string16::const_iterator a_end, - const char* b) { - return DoLowerCaseEqualsASCII(a_begin, a_end, b); -} +template<typename Str> +bool StartsWithT(BasicStringPiece<Str> str, + BasicStringPiece<Str> search_for, + CompareCase case_sensitivity) { + if (search_for.size() > str.size()) + return false; -bool LowerCaseEqualsASCII(const char* a_begin, - const char* a_end, - const char* b) { - return DoLowerCaseEqualsASCII(a_begin, a_end, b); -} + BasicStringPiece<Str> source = str.substr(0, search_for.size()); + + switch (case_sensitivity) { + case CompareCase::SENSITIVE: + return source == search_for; -bool LowerCaseEqualsASCII(const char* a_begin, - const char* a_end, - const char* b_begin, - const char* b_end) { - while (a_begin != a_end && b_begin != b_end && - ToLowerASCII(*a_begin) == *b_begin) { - a_begin++; - b_begin++; + case CompareCase::INSENSITIVE_ASCII: + return std::equal( + search_for.begin(), search_for.end(), + source.begin(), + CaseInsensitiveCompareASCII<typename Str::value_type>()); + + default: + NOTREACHED(); + return false; } - return a_begin == a_end && b_begin == b_end; } -bool LowerCaseEqualsASCII(const char16* a_begin, - const char16* a_end, - const char* b) { - return DoLowerCaseEqualsASCII(a_begin, a_end, b); +bool StartsWith(StringPiece str, + StringPiece search_for, + CompareCase case_sensitivity) { + return StartsWithT<std::string>(str, search_for, case_sensitivity); } -bool EqualsASCII(const string16& a, const StringPiece& b) { - if (a.length() != b.length()) - return false; - return std::equal(b.begin(), b.end(), a.begin()); +bool StartsWith(StringPiece16 str, + StringPiece16 search_for, + CompareCase case_sensitivity) { + return StartsWithT<string16>(str, search_for, case_sensitivity); } -bool StartsWithASCII(const std::string& str, - const std::string& search, - bool case_sensitive) { - if (case_sensitive) - return str.compare(0, search.length(), search) == 0; - else - return base::strncasecmp(str.c_str(), search.c_str(), search.length()) == 0; -} +template <typename Str> +bool EndsWithT(BasicStringPiece<Str> str, + BasicStringPiece<Str> search_for, + CompareCase case_sensitivity) { + if (search_for.size() > str.size()) + return false; + + BasicStringPiece<Str> source = str.substr(str.size() - search_for.size(), + search_for.size()); + + switch (case_sensitivity) { + case CompareCase::SENSITIVE: + return source == search_for; + + case CompareCase::INSENSITIVE_ASCII: + return std::equal( + source.begin(), source.end(), + search_for.begin(), + CaseInsensitiveCompareASCII<typename Str::value_type>()); -bool StartsWith(const string16& str, - const string16& search, - bool case_sensitive) { - if (case_sensitive) { - return str.compare(0, search.length(), search) == 0; + default: + NOTREACHED(); + return false; } - if (search.size() > str.size()) - return false; - return std::equal(search.begin(), search.end(), str.begin(), - CaseInsensitiveCompare<char16>()); } -template <typename STR> -bool EndsWithT(const STR& str, const STR& search, bool case_sensitive) { - size_t str_length = str.length(); - size_t search_length = search.length(); - if (search_length > str_length) - return false; - if (case_sensitive) - return str.compare(str_length - search_length, search_length, search) == 0; - return std::equal(search.begin(), search.end(), - str.begin() + (str_length - search_length), - base::CaseInsensitiveCompare<typename STR::value_type>()); +bool EndsWith(StringPiece str, + StringPiece search_for, + CompareCase case_sensitivity) { + return EndsWithT<std::string>(str, search_for, case_sensitivity); } -bool EndsWith(const std::string& str, const std::string& search, - bool case_sensitive) { - return EndsWithT(str, search, case_sensitive); +bool EndsWith(StringPiece16 str, + StringPiece16 search_for, + CompareCase case_sensitivity) { + return EndsWithT<string16>(str, search_for, case_sensitivity); } -bool EndsWith(const string16& str, const string16& search, - bool case_sensitive) { - return EndsWithT(str, search, case_sensitive); +char HexDigitToInt(wchar_t c) { + DCHECK(IsHexDigit(c)); + if (c >= '0' && c <= '9') + return static_cast<char>(c - '0'); + if (c >= 'A' && c <= 'F') + return static_cast<char>(c - 'A' + 10); + if (c >= 'a' && c <= 'f') + return static_cast<char>(c - 'a' + 10); + return 0; } -} // namespace base +bool IsUnicodeWhitespace(wchar_t c) { + // kWhitespaceWide is a NULL-terminated string + for (const wchar_t* cur = kWhitespaceWide; *cur; ++cur) { + if (*cur == c) + return true; + } + return false; +} static const char* const kByteStringsUnlocalized[] = { " B", @@ -562,7 +675,7 @@ static const char* const kByteStringsUnlocalized[] = { " PB" }; -string16 FormatBytesUnlocalized(int64 bytes) { +string16 FormatBytesUnlocalized(int64_t bytes) { double unit_amount = static_cast<double>(bytes); size_t dimension = 0; const int kKilo = 1024; @@ -581,20 +694,20 @@ string16 FormatBytesUnlocalized(int64 bytes) { kByteStringsUnlocalized[dimension]); } - return base::ASCIIToUTF16(buf); + return ASCIIToUTF16(buf); } // Runs in O(n) time in the length of |str|. template<class StringType> void DoReplaceSubstringsAfterOffset(StringType* str, size_t offset, - const StringType& find_this, - const StringType& replace_with, + BasicStringPiece<StringType> find_this, + BasicStringPiece<StringType> replace_with, bool replace_all) { DCHECK(!find_this.empty()); // If the find string doesn't appear, there's nothing to do. - offset = str->find(find_this, offset); + offset = str->find(find_this.data(), offset, find_this.size()); if (offset == StringType::npos) return; @@ -602,7 +715,7 @@ void DoReplaceSubstringsAfterOffset(StringType* str, // complicated. size_t find_length = find_this.length(); if (!replace_all) { - str->replace(offset, find_length, replace_with); + str->replace(offset, find_length, replace_with.data(), replace_with.size()); return; } @@ -611,8 +724,10 @@ void DoReplaceSubstringsAfterOffset(StringType* str, size_t replace_length = replace_with.length(); if (find_length == replace_length) { do { - str->replace(offset, find_length, replace_with); - offset = str->find(find_this, offset + replace_length); + str->replace(offset, find_length, + replace_with.data(), replace_with.size()); + offset = str->find(find_this.data(), offset + replace_length, + find_this.size()); } while (offset != StringType::npos); return; } @@ -629,11 +744,14 @@ void DoReplaceSubstringsAfterOffset(StringType* str, size_t write_offset = offset; do { if (replace_length) { - str->replace(write_offset, replace_length, replace_with); + str->replace(write_offset, replace_length, + replace_with.data(), replace_with.size()); write_offset += replace_length; } size_t read_offset = offset + find_length; - offset = std::min(str->find(find_this, read_offset), str_length); + offset = std::min( + str->find(find_this.data(), read_offset, find_this.size()), + str_length); size_t length = offset - read_offset; if (length) { memmove(&(*str)[write_offset], &(*str)[read_offset], @@ -662,13 +780,15 @@ void DoReplaceSubstringsAfterOffset(StringType* str, // exit from the loop, |current_match| will point at the last instance of // the find string, and we won't need to find() it again immediately. current_match = offset; - offset = str->find(find_this, offset + find_length); + offset = str->find(find_this.data(), offset + find_length, + find_this.size()); } while (offset != StringType::npos); str->resize(final_length); // Now do the replacement loop, working backwards through the string. for (size_t prev_match = str_length, write_offset = final_length; ; - current_match = str->rfind(find_this, current_match - 1)) { + current_match = str->rfind(find_this.data(), current_match - 1, + find_this.size())) { size_t read_offset = current_match + find_length; size_t length = prev_match - read_offset; if (length) { @@ -677,7 +797,8 @@ void DoReplaceSubstringsAfterOffset(StringType* str, length * sizeof(typename StringType::value_type)); } write_offset -= replace_length; - str->replace(write_offset, replace_length, replace_with); + str->replace(write_offset, replace_length, + replace_with.data(), replace_with.size()); if (current_match == first_match) return; prev_match = current_match; @@ -686,112 +807,97 @@ void DoReplaceSubstringsAfterOffset(StringType* str, void ReplaceFirstSubstringAfterOffset(string16* str, size_t start_offset, - const string16& find_this, - const string16& replace_with) { - DoReplaceSubstringsAfterOffset(str, start_offset, find_this, replace_with, - false); // replace first instance + StringPiece16 find_this, + StringPiece16 replace_with) { + DoReplaceSubstringsAfterOffset<string16>( + str, start_offset, find_this, replace_with, false); // Replace first. } void ReplaceFirstSubstringAfterOffset(std::string* str, size_t start_offset, - const std::string& find_this, - const std::string& replace_with) { - DoReplaceSubstringsAfterOffset(str, start_offset, find_this, replace_with, - false); // replace first instance + StringPiece find_this, + StringPiece replace_with) { + DoReplaceSubstringsAfterOffset<std::string>( + str, start_offset, find_this, replace_with, false); // Replace first. } void ReplaceSubstringsAfterOffset(string16* str, size_t start_offset, - const string16& find_this, - const string16& replace_with) { - DoReplaceSubstringsAfterOffset(str, start_offset, find_this, replace_with, - true); // replace all instances + StringPiece16 find_this, + StringPiece16 replace_with) { + DoReplaceSubstringsAfterOffset<string16>( + str, start_offset, find_this, replace_with, true); // Replace all. } void ReplaceSubstringsAfterOffset(std::string* str, size_t start_offset, - const std::string& find_this, - const std::string& replace_with) { - DoReplaceSubstringsAfterOffset(str, start_offset, find_this, replace_with, - true); // replace all instances + StringPiece find_this, + StringPiece replace_with) { + DoReplaceSubstringsAfterOffset<std::string>( + str, start_offset, find_this, replace_with, true); // Replace all. } -size_t Tokenize(const base::string16& str, - const base::string16& delimiters, - std::vector<base::string16>* tokens) { - *tokens = base::SplitString( - str, delimiters, base::KEEP_WHITESPACE, base::SPLIT_WANT_NONEMPTY); - return tokens->size(); +template <class string_type> +inline typename string_type::value_type* WriteIntoT(string_type* str, + size_t length_with_null) { + DCHECK_GT(length_with_null, 1u); + str->reserve(length_with_null); + str->resize(length_with_null - 1); + return &((*str)[0]); } -size_t Tokenize(const std::string& str, - const std::string& delimiters, - std::vector<std::string>* tokens) { - *tokens = base::SplitString( - str, delimiters, base::KEEP_WHITESPACE, base::SPLIT_WANT_NONEMPTY); - return tokens->size(); +char* WriteInto(std::string* str, size_t length_with_null) { + return WriteIntoT(str, length_with_null); } -size_t Tokenize(const base::StringPiece& str, - const base::StringPiece& delimiters, - std::vector<base::StringPiece>* tokens) { - *tokens = base::SplitStringPiece( - str, delimiters, base::KEEP_WHITESPACE, base::SPLIT_WANT_NONEMPTY); - return tokens->size(); +char16* WriteInto(string16* str, size_t length_with_null) { + return WriteIntoT(str, length_with_null); } template<typename STR> -static STR JoinStringT(const std::vector<STR>& parts, const STR& sep) { +static STR JoinStringT(const std::vector<STR>& parts, + BasicStringPiece<STR> sep) { if (parts.empty()) return STR(); STR result(parts[0]); - typename std::vector<STR>::const_iterator iter = parts.begin(); + auto iter = parts.begin(); ++iter; for (; iter != parts.end(); ++iter) { - result += sep; + sep.AppendToString(&result); result += *iter; } return result; } -std::string JoinString(const std::vector<std::string>& parts, char sep) { - return JoinStringT(parts, std::string(1, sep)); -} - -string16 JoinString(const std::vector<string16>& parts, char16 sep) { - return JoinStringT(parts, string16(1, sep)); -} - std::string JoinString(const std::vector<std::string>& parts, - const std::string& separator) { + StringPiece separator) { return JoinStringT(parts, separator); } string16 JoinString(const std::vector<string16>& parts, - const string16& separator) { + StringPiece16 separator) { return JoinStringT(parts, separator); } template<class FormatStringType, class OutStringType> -OutStringType DoReplaceStringPlaceholders(const FormatStringType& format_string, - const std::vector<OutStringType>& subst, std::vector<size_t>* offsets) { +OutStringType DoReplaceStringPlaceholders( + const FormatStringType& format_string, + const std::vector<OutStringType>& subst, + std::vector<size_t>* offsets) { size_t substitutions = subst.size(); size_t sub_length = 0; - for (typename std::vector<OutStringType>::const_iterator iter = subst.begin(); - iter != subst.end(); ++iter) { - sub_length += iter->length(); - } + for (const auto& cur : subst) + sub_length += cur.length(); OutStringType formatted; formatted.reserve(format_string.length() + sub_length); std::vector<ReplacementOffset> r_offsets; - for (typename FormatStringType::const_iterator i = format_string.begin(); - i != format_string.end(); ++i) { + for (auto i = format_string.begin(); i != format_string.end(); ++i) { if ('$' == *i) { if (i + 1 != format_string.end()) { ++i; @@ -829,10 +935,8 @@ OutStringType DoReplaceStringPlaceholders(const FormatStringType& format_string, } } if (offsets) { - for (std::vector<ReplacementOffset>::const_iterator i = r_offsets.begin(); - i != r_offsets.end(); ++i) { - offsets->push_back(i->offset); - } + for (const auto& cur : r_offsets) + offsets->push_back(cur.offset); } return formatted; } @@ -843,7 +947,7 @@ string16 ReplaceStringPlaceholders(const string16& format_string, return DoReplaceStringPlaceholders(format_string, subst, offsets); } -std::string ReplaceStringPlaceholders(const base::StringPiece& format_string, +std::string ReplaceStringPlaceholders(const StringPiece& format_string, const std::vector<std::string>& subst, std::vector<size_t>* offsets) { return DoReplaceStringPlaceholders(format_string, subst, offsets); @@ -863,161 +967,6 @@ string16 ReplaceStringPlaceholders(const string16& format_string, return result; } -static bool IsWildcard(base_icu::UChar32 character) { - return character == '*' || character == '?'; -} - -// Move the strings pointers to the point where they start to differ. -template <typename CHAR, typename NEXT> -static void EatSameChars(const CHAR** pattern, const CHAR* pattern_end, - const CHAR** string, const CHAR* string_end, - NEXT next) { - const CHAR* escape = NULL; - while (*pattern != pattern_end && *string != string_end) { - if (!escape && IsWildcard(**pattern)) { - // We don't want to match wildcard here, except if it's escaped. - return; - } - - // Check if the escapement char is found. If so, skip it and move to the - // next character. - if (!escape && **pattern == '\\') { - escape = *pattern; - next(pattern, pattern_end); - continue; - } - - // Check if the chars match, if so, increment the ptrs. - const CHAR* pattern_next = *pattern; - const CHAR* string_next = *string; - base_icu::UChar32 pattern_char = next(&pattern_next, pattern_end); - if (pattern_char == next(&string_next, string_end) && - pattern_char != CBU_SENTINEL) { - *pattern = pattern_next; - *string = string_next; - } else { - // Uh oh, it did not match, we are done. If the last char was an - // escapement, that means that it was an error to advance the ptr here, - // let's put it back where it was. This also mean that the MatchPattern - // function will return false because if we can't match an escape char - // here, then no one will. - if (escape) { - *pattern = escape; - } - return; - } - - escape = NULL; - } -} - -template <typename CHAR, typename NEXT> -static void EatWildcard(const CHAR** pattern, const CHAR* end, NEXT next) { - while (*pattern != end) { - if (!IsWildcard(**pattern)) - return; - next(pattern, end); - } -} - -template <typename CHAR, typename NEXT> -static bool MatchPatternT(const CHAR* eval, const CHAR* eval_end, - const CHAR* pattern, const CHAR* pattern_end, - int depth, - NEXT next) { - const int kMaxDepth = 16; - if (depth > kMaxDepth) - return false; - - // Eat all the matching chars. - EatSameChars(&pattern, pattern_end, &eval, eval_end, next); - - // If the string is empty, then the pattern must be empty too, or contains - // only wildcards. - if (eval == eval_end) { - EatWildcard(&pattern, pattern_end, next); - return pattern == pattern_end; - } - - // Pattern is empty but not string, this is not a match. - if (pattern == pattern_end) - return false; - - // If this is a question mark, then we need to compare the rest with - // the current string or the string with one character eaten. - const CHAR* next_pattern = pattern; - next(&next_pattern, pattern_end); - if (pattern[0] == '?') { - if (MatchPatternT(eval, eval_end, next_pattern, pattern_end, - depth + 1, next)) - return true; - const CHAR* next_eval = eval; - next(&next_eval, eval_end); - if (MatchPatternT(next_eval, eval_end, next_pattern, pattern_end, - depth + 1, next)) - return true; - } - - // This is a *, try to match all the possible substrings with the remainder - // of the pattern. - if (pattern[0] == '*') { - // Collapse duplicate wild cards (********** into *) so that the - // method does not recurse unnecessarily. http://crbug.com/52839 - EatWildcard(&next_pattern, pattern_end, next); - - while (eval != eval_end) { - if (MatchPatternT(eval, eval_end, next_pattern, pattern_end, - depth + 1, next)) - return true; - eval++; - } - - // We reached the end of the string, let see if the pattern contains only - // wildcards. - if (eval == eval_end) { - EatWildcard(&pattern, pattern_end, next); - if (pattern != pattern_end) - return false; - return true; - } - } - - return false; -} - -struct NextCharUTF8 { - base_icu::UChar32 operator()(const char** p, const char* end) { - base_icu::UChar32 c; - int offset = 0; - CBU8_NEXT(*p, offset, end - *p, c); - *p += offset; - return c; - } -}; - -struct NextCharUTF16 { - base_icu::UChar32 operator()(const char16** p, const char16* end) { - base_icu::UChar32 c; - int offset = 0; - CBU16_NEXT(*p, offset, end - *p, c); - *p += offset; - return c; - } -}; - -bool MatchPattern(const base::StringPiece& eval, - const base::StringPiece& pattern) { - return MatchPatternT(eval.data(), eval.data() + eval.size(), - pattern.data(), pattern.data() + pattern.size(), - 0, NextCharUTF8()); -} - -bool MatchPattern(const string16& eval, const string16& pattern) { - return MatchPatternT(eval.c_str(), eval.c_str() + eval.size(), - pattern.c_str(), pattern.c_str() + pattern.size(), - 0, NextCharUTF16()); -} - // The following code is compatible with the OpenBSD lcpy interface. See: // http://www.gratisoft.us/todd/papers/strlcpy.html // ftp://ftp.openbsd.org/pub/OpenBSD/src/lib/libc/string/{wcs,str}lcpy.c @@ -1042,9 +991,11 @@ size_t lcpyT(CHAR* dst, const CHAR* src, size_t dst_size) { } // namespace -size_t base::strlcpy(char* dst, const char* src, size_t dst_size) { +size_t strlcpy(char* dst, const char* src, size_t dst_size) { return lcpyT<char>(dst, src, dst_size); } -size_t base::wcslcpy(wchar_t* dst, const wchar_t* src, size_t dst_size) { +size_t wcslcpy(wchar_t* dst, const wchar_t* src, size_t dst_size) { return lcpyT<wchar_t>(dst, src, dst_size); } + +} // namespace base |