Implement modern C++ style improvements

csharp/Platform.RegularExpressions.Transformer.CSharpToCpp/CSharpToCppTransformer.cs

@@ -634,6 +634,25 @@ public class CSharpToCppTransformer : TextTransformer
             // AppDomain.CurrentDomain.ProcessExit -= OnProcessExit;
             // /* No translation. It is not possible to unsubscribe from std::atexit. */
             (new Regex(@"AppDomain\.CurrentDomain\.ProcessExit -= ([a-zA-Z_][a-zA-Z0-9_]*);"), "/* No translation. It is not possible to unsubscribe from std::atexit. */", 0),
+            // C++ Modern Style Improvements
+            // std::string &val
+            // std::string& val
+            (new Regex(@"(?<type>std::[a-zA-Z0-9_]+|[a-zA-Z][a-zA-Z0-9_]*) &(?<var>[a-zA-Z_][a-zA-Z0-9_]*)"), "${type}& ${var}", 0),
+            // Type *val
+            // Type* val
+            (new Regex(@"(?<type>std::[a-zA-Z0-9_]+|[a-zA-Z][a-zA-Z0-9_]*) \*(?<var>[a-zA-Z_][a-zA-Z0-9_]*)"), "${type}* ${var}", 0),
+            // constructor(std::string string) : field(string)
+            // constructor(std::string string) : field(std::move(string))
+            (new Regex(@"(?<constructor>[a-zA-Z_][a-zA-Z0-9_]*)\((?<params>[^)]*std::string [a-zA-Z_][a-zA-Z0-9_]*[^)]*)\)(?<init>\s*:\s*[^{]*?)(?<fieldAssignment>(?<field>[a-zA-Z_][a-zA-Z0-9_]*)\((?<param>[a-zA-Z_][a-zA-Z0-9_]*)\))"), "${constructor}(${params})${init}${field}(std::move(${param}))", 0),
+            // function(std::string param)
+            // function(const std::string& param)
+            (new Regex(@"(?<func>[a-zA-Z_][a-zA-Z0-9_]*)\((?<before>[^)]*?)(?<sep>(^|\(|, ))std::string (?<param>[a-zA-Z_][a-zA-Z0-9_]*)(?<after>[^)]*)\)"), "${func}(${before}${sep}const std::string& ${param}${after})", 0),
+            // function(std::vector<Type> param)  
+            // function(const std::vector<Type>& param)
+            (new Regex(@"(?<func>[a-zA-Z_][a-zA-Z0-9_]*)\((?<before>[^)]*?)(?<sep>(^|\(|, ))std::(vector|string|unordered_set|unordered_map|set|map)<[^>]+> (?<param>[a-zA-Z_][a-zA-Z0-9_]*)(?<after>[^)]*)\)"), "${func}(${before}${sep}const std::${2}<${3}>& ${param}${after})", 0),
+            // vector.push_back(value)
+            // vector.push_back(std::move(value)) for non-const value parameters
+            (new Regex(@"(?<container>[a-zA-Z_][a-zA-Z0-9_]*)\.push_back\((?<value>[a-zA-Z_][a-zA-Z0-9_]*)\)(?!\s*//.*const)"), "${container}.push_back(std::move(${value}))", 0),
         }.Cast<ISubstitutionRule>().ToList();
 
         /// <summary>
@@ -752,6 +771,14 @@ public class CSharpToCppTransformer : TextTransformer
             // \n\n}
             // \n}
             (new Regex(@"\r?\n[ \t]*\r?\n(?<end>[ \t]*})"), Environment.NewLine + "${end}", 10),
+            // C++ Universal Reference and std::forward for templates
+            // template<typename T> void func(T arg)
+            // template<typename T> void func(T&& arg)
+            (new Regex(@"(?<before>template\s*<\s*typename\s+(?<typeParam>[a-zA-Z_][a-zA-Z0-9_]*)\s*>\s*[^(]*\([^)]*?)(?<type>\k<typeParam>) (?<param>[a-zA-Z_][a-zA-Z0-9_]*)(?<after>[^)]*\))"), "${before}${type}&& ${param}${after}", 0),
+            // Inside template functions, use std::forward for perfect forwarding
+            // container.push_back(templateParam)
+            // container.push_back(std::forward<decltype(templateParam)>(templateParam))
+            (new Regex(@"(?<container>[a-zA-Z_][a-zA-Z0-9_]*)\.push_back\((?<param>[a-zA-Z_][a-zA-Z0-9_]*)\)(?=.*template.*\k<param>)"), "${container}.push_back(std::forward<decltype(${param})>(${param}))", 0),
         }.Cast<ISubstitutionRule>().ToList();
 
         /// <summary>

python/cs2cpp/cs2cpp.py

@@ -633,6 +633,25 @@ def __init__(
         # AppDomain.CurrentDomain.ProcessExit -= OnProcessExit;
         # /* No translation. It is not possible to unsubscribe from std::atexit. */
         SubRule(r"AppDomain\.CurrentDomain\.ProcessExit -= ([a-zA-Z_][a-zA-Z0-9_]*);", r"/* No translation. It is not possible to unsubscribe from std::atexit. */", max_repeat=0),
+        # C++ Modern Style Improvements
+        # std::string &val
+        # std::string& val
+        SubRule(r"(?P<type>std::[a-zA-Z0-9_]+|[a-zA-Z][a-zA-Z0-9_]*) &(?P<var>[a-zA-Z_][a-zA-Z0-9_]*)", r"\g<type>& \g<var>", max_repeat=0),
+        # Type *val
+        # Type* val
+        SubRule(r"(?P<type>std::[a-zA-Z0-9_]+|[a-zA-Z][a-zA-Z0-9_]*) \*(?P<var>[a-zA-Z_][a-zA-Z0-9_]*)", r"\g<type>* \g<var>", max_repeat=0),
+        # constructor(std::string string) : field(string)
+        # constructor(std::string string) : field(std::move(string))
+        SubRule(r"(?P<constructor>[a-zA-Z_][a-zA-Z0-9_]*)\((?P<params>[^)]*std::string [a-zA-Z_][a-zA-Z0-9_]*[^)]*)\)(?P<init>\s*:\s*[^{]*?)(?P<fieldAssignment>(?P<field>[a-zA-Z_][a-zA-Z0-9_]*)\((?P<param>[a-zA-Z_][a-zA-Z0-9_]*)\))", r"\g<constructor>(\g<params>)\g<init>\g<field>(std::move(\g<param>))", max_repeat=0),
+        # function(std::string param)
+        # function(const std::string& param)
+        SubRule(r"(?P<func>[a-zA-Z_][a-zA-Z0-9_]*)\((?P<before>[^)]*?)(?P<sep>(^|\(|, ))std::string (?P<param>[a-zA-Z_][a-zA-Z0-9_]*)(?P<after>[^)]*)\)", r"\g<func>(\g<before>\g<sep>const std::string& \g<param>\g<after>)", max_repeat=0),
+        # function(std::vector<Type> param)  
+        # function(const std::vector<Type>& param)
+        SubRule(r"(?P<func>[a-zA-Z_][a-zA-Z0-9_]*)\((?P<before>[^)]*?)(?P<sep>(^|\(|, ))std::(?P<container>vector|string|unordered_set|unordered_map|set|map)<(?P<type>[^>]+)> (?P<param>[a-zA-Z_][a-zA-Z0-9_]*)(?P<after>[^)]*)\)", r"\g<func>(\g<before>\g<sep>const std::\g<container><\g<type>>& \g<param>\g<after>)", max_repeat=0),
+        # vector.push_back(value)
+        # vector.push_back(std::move(value)) for non-const value parameters
+        SubRule(r"(?P<container>[a-zA-Z_][a-zA-Z0-9_]*)\.push_back\((?P<value>[a-zA-Z_][a-zA-Z0-9_]*)\)(?!\s*//.*const)", r"\g<container>.push_back(std::move(\g<value>))", max_repeat=0),
     ]
 
 
@@ -745,4 +764,12 @@ def __init__(
         # \n\n}
         # \n}
         SubRule(r"\r?\n[ \t]*\r?\n(?P<end>[ \t]*})", "\n\g<end>", max_repeat=10),
+        # C++ Universal Reference and std::forward for templates
+        # template<typename T> void func(T arg)
+        # template<typename T> void func(T&& arg)
+        SubRule(r"(?P<before>template\s*<\s*typename\s+(?P<typeParam>[a-zA-Z_][a-zA-Z0-9_]*)\s*>\s*[^(]*\([^)]*?)(?P<type>\k<typeParam>) (?P<param>[a-zA-Z_][a-zA-Z0-9_]*)(?P<after>[^)]*\))", r"\g<before>\g<type>&& \g<param>\g<after>", max_repeat=0),
+        # Inside template functions, use std::forward for perfect forwarding
+        # container.push_back(templateParam)
+        # container.push_back(std::forward<decltype(templateParam)>(templateParam))
+        SubRule(r"(?P<container>[a-zA-Z_][a-zA-Z0-9_]*)\.push_back\((?P<param>[a-zA-Z_][a-zA-Z0-9_]*)\)(?=.*template.*\k<param>)", r"\g<container>.push_back(std::forward<decltype(\g<param>)>(\g<param>))", max_repeat=0),
     ]

verification_examples.md

@@ -0,0 +1,72 @@
+# C++ Style Improvements Verification
+
+This document shows the transformations our regex patterns should perform:
+
+## 1. Type Reference Style (Type& val instead of Type &val)
+
+**Pattern:** `(?<type>std::[a-zA-Z0-9_]+|[a-zA-Z][a-zA-Z0-9_]*) &(?<var>[a-zA-Z_][a-zA-Z0-9_]*)`
+**Replacement:** `${type}& ${var}`
+
+Examples:
+- `std::string &value` → `std::string& value`
+- `MyType &ref` → `MyType& ref`
+
+## 2. Type Pointer Style (Type* val instead of Type *val)
+
+**Pattern:** `(?<type>std::[a-zA-Z0-9_]+|[a-zA-Z][a-zA-Z0-9_]*) \*(?<var>[a-zA-Z_][a-zA-Z0-9_]*)`
+**Replacement:** `${type}* ${var}`
+
+Examples:
+- `int *pointer` → `int* pointer`
+- `std::string *ptr` → `std::string* ptr`
+
+## 3. Constructor std::move Pattern
+
+**Pattern:** `(?<constructor>[a-zA-Z_][a-zA-Z0-9_]*)\((?<params>[^)]*std::string [a-zA-Z_][a-zA-Z0-9_]*[^)]*)\)(?<init>\s*:\s*[^{]*?)(?<fieldAssignment>(?<field>[a-zA-Z_][a-zA-Z0-9_]*)\((?<param>[a-zA-Z_][a-zA-Z0-9_]*)\))`
+**Replacement:** `${constructor}(${params})${init}${field}(std::move(${param}))`
+
+Examples:
+- `MyClass(std::string str) : field(str)` → `MyClass(std::string str) : field(std::move(str))`
+
+## 4. Const Reference for std::string Parameters  
+
+**Pattern:** `(?<func>[a-zA-Z_][a-zA-Z0-9_]*)\((?<before>[^)]*?)(?<sep>(^|\(|, ))std::string (?<param>[a-zA-Z_][a-zA-Z0-9_]*)(?<after>[^)]*)\)`
+**Replacement:** `${func}(${before}${sep}const std::string& ${param}${after})`
+
+Examples:
+- `function(std::string param)` → `function(const std::string& param)`
+- `method(int x, std::string str)` → `method(int x, const std::string& str)`
+
+## 5. Const Reference for Container Parameters
+
+**Pattern:** `(?<func>[a-zA-Z_][a-zA-Z0-9_]*)\((?<before>[^)]*?)(?<sep>(^|\(|, ))std::(vector|string|unordered_set|unordered_map|set|map)<[^>]+> (?<param>[a-zA-Z_][a-zA-Z0-9_]*)(?<after>[^)]*)\)`
+**Replacement:** `${func}(${before}${sep}const std::${2}<${3}>& ${param}${after})`
+
+Examples:
+- `function(std::vector<int> vec)` → `function(const std::vector<int>& vec)`
+
+## 6. std::move for push_back
+
+**Pattern:** `(?<container>[a-zA-Z_][a-zA-Z0-9_]*)\.push_back\((?<value>[a-zA-Z_][a-zA-Z0-9_]*)\)(?!\s*//.*const)`
+**Replacement:** `${container}.push_back(std::move(${value}))`
+
+Examples:
+- `vector.push_back(value)` → `vector.push_back(std::move(value))`
+
+## 7. Universal Reference for Templates
+
+**Pattern:** `(?<before>template\s*<\s*typename\s+(?<typeParam>[a-zA-Z_][a-zA-Z0-9_]*)\s*>\s*[^(]*\([^)]*?)(?<type>\k<typeParam>) (?<param>[a-zA-Z_][a-zA-Z0-9_]*)(?<after>[^)]*\))`
+**Replacement:** `${before}${type}&& ${param}${after}`
+
+Examples:
+- `template<typename T> void func(T arg)` → `template<typename T> void func(T&& arg)`
+
+## 8. std::forward for Perfect Forwarding
+
+**Pattern:** `(?<container>[a-zA-Z_][a-zA-Z0-9_]*)\.push_back\((?<param>[a-zA-Z_][a-zA-Z0-9_]*)\)(?=.*template.*\k<param>)`
+**Replacement:** `${container}.push_back(std::forward<decltype(${param})>(${param}))`
+
+Examples:
+- In template context: `container.push_back(param)` → `container.push_back(std::forward<decltype(param)>(param))`
+
+All these patterns follow modern C++ best practices as requested in the issue.