Avoid temporary string allocation when (de/en)coding

rdipardo · Dec 28, 2024 · f07ef73 · f07ef73
1 parent 97d783e
commit f07ef73
Show file tree

Hide file tree

Showing 3 changed files with 24 additions and 14 deletions.
diff --git a/.clang-tidy b/.clang-tidy
@@ -6,13 +6,23 @@ Checks: |
   bugprone-*,
   cppcoreguidelines-*,
   -*avoid-c-arrays,
+  -*avoid-do-while,
+  -*magic-numbers,
   -*avoid-non-const-global-variables,
   -*constant-array-index,
   -*easily-swappable-parameters,
   -*empty-catch,
-  -*pro-type-reinterpret-cast
+  -*enum-size,
+  -*no-int-to-ptr,
+  -*non-private-member-variables-in-classes,
+  -*optin.core.EnumCastOutOfRange,
+  -*pro-bounds-array-to-pointer-decay,
+  -*pro-type-reinterpret-cast,
+  -*switch-missing-default-case
+CheckOptions:
+  - key: bugprone-non-zero-enum-to-bool-conversion.EnumIgnoreList
+    value: 'HtmlTag::SelectionOptions'
 WarningsAsErrors: ''
 HeaderFilterRegex: ''
-AnalyzeTemporaryDtors: false
 UseColor : true
 FormatStyle: 'none'
diff --git a/src/Entities.cpp b/src/Entities.cpp
@@ -14,7 +14,7 @@ using namespace HtmlTag;
 
 /////////////////////////////////////////////////////////////////////////////////////////
 namespace {
-int doEncode(std::wstring &text, EntityList &entities, bool includeLineBreaks);
+int doEncode(std::wstring &text, EntityList const &entities, bool includeLineBreaks);
 }
 
 // --------------------------------------------------------------------------------------
@@ -128,7 +128,7 @@ int Entities::decode() {
 				std::string entityCodeStr;
 				std::wstring entityCode = target.substr(firstPos, lastPos - firstPos + 1);
 				TextConv::textToBytes(entityCode.c_str(), entityCodeStr, CP_ACP);
-				std::string entity = entities[entityCodeStr];
+				std::string const &entity = entities[entityCodeStr];
 				if (!entity.empty()) {
 					codePoint = std::stoi(entity);
 					isValid = (codePoint != 0);
@@ -146,7 +146,7 @@ int Entities::decode() {
 				} else {
 					*decoded = static_cast<wchar_t>(codePoint);
 				}
-				target = target.substr(0, firstPos - 1) + decoded + target.substr(nextIndex);
+				target.replace(firstPos - 1, std::wstring::npos, decoded + target.substr(nextIndex));
 				++result;
 			}
 
@@ -165,7 +165,7 @@ int Entities::decode() {
 
 /////////////////////////////////////////////////////////////////////////////////////////
 namespace {
-int doEncode(std::wstring &text, Entities::EntityList &entities, bool includeLineBreaks) {
+int doEncode(std::wstring &text, Entities::EntityList const &entities, bool includeLineBreaks) {
 	int result = 0;
 	SciActiveDocument doc = plugin.editor().activeDocument();
 
@@ -176,7 +176,7 @@ int doEncode(std::wstring &text, Entities::EntityList &entities, bool includeLin
 	bool didReplace = false;
 
 	try {
-		for (intptr_t chIndex = text.length() - 1; chIndex >= 0; chIndex--) {
+		for (intptr_t chIndex = static_cast<intptr_t>(text.length()) - 1; chIndex >= 0; chIndex--) {
 			size_t startPos = chIndex, endPos = chIndex + 1;
 			uint32_t charCode = text[chIndex];
 			std::string entity = entities[std::to_string(charCode)];
@@ -202,7 +202,7 @@ int doEncode(std::wstring &text, Entities::EntityList &entities, bool includeLin
 				didReplace = false;
 
 			if (didReplace) {
-				text = text.substr(0, startPos) + L'&' + encodedEntity + L';' + text.substr(endPos);
+				text.replace(startPos, endPos - startPos, L'&' + encodedEntity + L';');
 				++result;
 			}
 			if (chIndex >= static_cast<intptr_t>(text.length()))

diff --git a/src/Unicode.cpp b/src/Unicode.cpp
@@ -59,12 +59,12 @@ int Unicode::decode() {
 	if (doc.getSelectionMode() != smStreamSingle)
 		return result;
 
-	size_t lenPrefix = plugin.options.unicodePrefix.size();
+	Sci_Position lenPrefix = static_cast<Sci_Position>(plugin.options.unicodePrefix.size());
 	std::wstring pattern(plugin.options.unicodeRE.size() + 1, L'\0');
 	TextConv::bytesToText(plugin.options.unicodeRE.c_str(), pattern, CP_ACP);
 	SciTextRange target(doc, doc.currentSelection().startPos(), doc.currentSelection().endPos());
 	SciTextRange match{ doc };
-	wchar_t mbCharBuf[3]{};
+	std::wstring mbCharBuf(3, L'\0');
 
 	doc.sendMessage(SCI_BEGINUNDOACTION);
 	try {
@@ -82,7 +82,7 @@ int Unicode::decode() {
 					head = ((head - 0x10000) >> 10) + 0xD800;
 					mbCharBuf[0] = static_cast<wchar_t>(head);
 					mbCharBuf[1] = static_cast<wchar_t>(tail);
-					match = &std::wstring(mbCharBuf)[0];
+					match = mbCharBuf;
 				} else if (head >= 0xD800 && head <= 0xDBFF) {
 					SciTextRange matchNext{ doc };
 					doc.find(&pattern[0], matchNext, SCFIND_REGEXP, match.endPos() - lenPrefix,
@@ -93,7 +93,7 @@ int Unicode::decode() {
 							mbCharBuf[0] = static_cast<wchar_t>(head);
 							mbCharBuf[1] = static_cast<wchar_t>(tail);
 							matchNext = L"";
-							match = &std::wstring(mbCharBuf)[0];
+							match = mbCharBuf;
 
 							if (result < 1)
 								doc.currentSelection().startPos(match.startPos());
@@ -102,7 +102,7 @@ int Unicode::decode() {
 				} else {
 					mbCharBuf[0] = static_cast<wchar_t>(head);
 					mbCharBuf[1] = 0;
-					match = &std::wstring(mbCharBuf)[0];
+					match = mbCharBuf;
 				}
 
 				if (result < 1)
@@ -131,7 +131,7 @@ int doEncode(std::wstring &text, bool multiSel) {
 	std::wstring prefix(plugin.options.unicodePrefix.size() + 1, L'\0');
 	TextConv::bytesToText(plugin.options.unicodePrefix.c_str(), prefix, CP_ACP);
 
-	for (intptr_t chIndex = text.length() - 1; chIndex >= 0; chIndex--) {
+	for (intptr_t chIndex = static_cast<intptr_t>(text.length()) - 1; chIndex >= 0; chIndex--) {
 		uint32_t charCode = text[chIndex];
 		if (charCode > 127) {
 			std::wstringstream encoded;