From cd46b267f5715c0900b409d7bcb97f6bc9a43d9e Mon Sep 17 00:00:00 2001
From: Lioncash <mathew1800@gmail.com>
Date: Sun, 22 Jul 2018 15:33:07 -0400
Subject: [PATCH 1/3] string_util: Remove unnecessary std::string instance in
TabsToSpaces()
We can just use the variant of std::string's replace() function that can
replace an occurrence with N copies of the same character, eliminating
the need to allocate a std::string containing a buffer of spaces.
---
src/common/string_util.cpp | 13 ++++++-------
src/common/string_util.h | 2 +-
2 files changed, 7 insertions(+), 8 deletions(-)
diff --git a/src/common/string_util.cpp b/src/common/string_util.cpp
index f3ad3d68a..2099eebb8 100644
--- a/src/common/string_util.cpp
+++ b/src/common/string_util.cpp
@@ -168,15 +168,14 @@ void SplitString(const std::string& str, const char delim, std::vector<std::stri
output.pop_back();
}
-std::string TabsToSpaces(int tab_size, const std::string& in) {
- const std::string spaces(tab_size, ' ');
- std::string out(in);
-
+std::string TabsToSpaces(int tab_size, std::string in) {
size_t i = 0;
- while (out.npos != (i = out.find('\t')))
- out.replace(i, 1, spaces);
- return out;
+ while ((i = in.find('\t')) != std::string::npos) {
+ in.replace(i, 1, tab_size, ' ');
+ }
+
+ return in;
}
std::string ReplaceAll(std::string result, const std::string& src, const std::string& dest) {
diff --git a/src/common/string_util.h b/src/common/string_util.h
index daa071f83..4a2143b59 100644
--- a/src/common/string_util.h
+++ b/src/common/string_util.h
@@ -57,7 +57,7 @@ static bool TryParse(const std::string& str, N* const output) {
return false;
}
-std::string TabsToSpaces(int tab_size, const std::string& in);
+std::string TabsToSpaces(int tab_size, std::string in);
void SplitString(const std::string& str, char delim, std::vector<std::string>& output);
From 26a157cd31676f419420078e7d99df90d9fbc910 Mon Sep 17 00:00:00 2001
From: Lioncash <mathew1800@gmail.com>
Date: Sun, 22 Jul 2018 15:36:30 -0400
Subject: [PATCH 2/3] string_util: Use emplace_back() in SplitString() instead
of push_back()
This is equivalent to doing:
push_back(std::string(""));
which is likely not to cause issues, assuming a decent std::string
implementation with small-string optimizations implemented in its
design, however it's still a little unnecessary to copy that buffer
regardless. Instead, we can use emplace_back() to directly construct the
empty string within the std::vector instance, eliminating any possible
overhead from the copy.
---
src/common/string_util.cpp | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/common/string_util.cpp b/src/common/string_util.cpp
index 2099eebb8..6737655a5 100644
--- a/src/common/string_util.cpp
+++ b/src/common/string_util.cpp
@@ -162,8 +162,9 @@ void SplitString(const std::string& str, const char delim, std::vector<std::stri
std::istringstream iss(str);
output.resize(1);
- while (std::getline(iss, *output.rbegin(), delim))
- output.push_back("");
+ while (std::getline(iss, *output.rbegin(), delim)) {
+ output.emplace_back();
+ }
output.pop_back();
}
From 9d331221970a26a2d2f68fe51c0249b62fafd0a2 Mon Sep 17 00:00:00 2001
From: Lioncash <mathew1800@gmail.com>
Date: Sun, 22 Jul 2018 15:47:37 -0400
Subject: [PATCH 3/3] string_util: Get rid of separate resize() in CPToUTF16(),
UTF16ToUTF8(), CodeToUTF8() and UTF8ToUTF16()
There's no need to perform the resize separately here, since the
constructor allows presizing the buffer.
Also move the empty string check before the construction of the string
to make the early out more straightforward.
---
src/common/string_util.cpp | 42 ++++++++++++++++++++------------------
1 file changed, 22 insertions(+), 20 deletions(-)
diff --git a/src/common/string_util.cpp b/src/common/string_util.cpp
index 6737655a5..1f0456aee 100644
--- a/src/common/string_util.cpp
+++ b/src/common/string_util.cpp
@@ -220,31 +220,37 @@ std::u16string UTF8ToUTF16(const std::string& input) {
}
static std::wstring CPToUTF16(u32 code_page, const std::string& input) {
- auto const size =
+ const auto size =
MultiByteToWideChar(code_page, 0, input.data(), static_cast<int>(input.size()), nullptr, 0);
- std::wstring output;
- output.resize(size);
+ if (size == 0) {
+ return L"";
+ }
- if (size == 0 ||
- size != MultiByteToWideChar(code_page, 0, input.data(), static_cast<int>(input.size()),
- &output[0], static_cast<int>(output.size())))
+ std::wstring output(size, L'\0');
+
+ if (size != MultiByteToWideChar(code_page, 0, input.data(), static_cast<int>(input.size()),
+ &output[0], static_cast<int>(output.size()))) {
output.clear();
+ }
return output;
}
std::string UTF16ToUTF8(const std::wstring& input) {
- auto const size = WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
+ const auto size = WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
nullptr, 0, nullptr, nullptr);
+ if (size == 0) {
+ return "";
+ }
- std::string output;
- output.resize(size);
+ std::string output(size, '\0');
- if (size == 0 ||
- size != WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
- &output[0], static_cast<int>(output.size()), nullptr, nullptr))
+ if (size != WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
+ &output[0], static_cast<int>(output.size()), nullptr,
+ nullptr)) {
output.clear();
+ }
return output;
}
@@ -265,8 +271,6 @@ std::string CP1252ToUTF8(const std::string& input) {
template <typename T>
static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input) {
- std::string result;
-
iconv_t const conv_desc = iconv_open("UTF-8", fromcode);
if ((iconv_t)(-1) == conv_desc) {
LOG_ERROR(Common, "Iconv initialization failure [{}]: {}", fromcode, strerror(errno));
@@ -278,8 +282,7 @@ static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>&
// Multiply by 4, which is the max number of bytes to encode a codepoint
const size_t out_buffer_size = 4 * in_bytes;
- std::string out_buffer;
- out_buffer.resize(out_buffer_size);
+ std::string out_buffer(out_buffer_size, '\0');
auto src_buffer = &input[0];
size_t src_bytes = in_bytes;
@@ -304,6 +307,7 @@ static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>&
}
}
+ std::string result;
out_buffer.resize(out_buffer_size - dst_bytes);
out_buffer.swap(result);
@@ -313,8 +317,6 @@ static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>&
}
std::u16string UTF8ToUTF16(const std::string& input) {
- std::u16string result;
-
iconv_t const conv_desc = iconv_open("UTF-16LE", "UTF-8");
if ((iconv_t)(-1) == conv_desc) {
LOG_ERROR(Common, "Iconv initialization failure [UTF-8]: {}", strerror(errno));
@@ -326,8 +328,7 @@ std::u16string UTF8ToUTF16(const std::string& input) {
// Multiply by 4, which is the max number of bytes to encode a codepoint
const size_t out_buffer_size = 4 * sizeof(char16_t) * in_bytes;
- std::u16string out_buffer;
- out_buffer.resize(out_buffer_size);
+ std::u16string out_buffer(out_buffer_size, char16_t{});
char* src_buffer = const_cast<char*>(&input[0]);
size_t src_bytes = in_bytes;
@@ -352,6 +353,7 @@ std::u16string UTF8ToUTF16(const std::string& input) {
}
}
+ std::u16string result;
out_buffer.resize(out_buffer_size - dst_bytes);
out_buffer.swap(result);