move string utility to `util.cc` + small script identifier fix

6 files changed, 123 insertions, 110 deletions

diff --git a/src/common.hh b/src/common.hh
index 02cad88..219ff22 100644
--- a/src/common.hh
+++ b/src/common.hh
@@ -61,7 +61,7 @@ STATIC_ASSERT(sizeof(int) == 4);
 #endif
 
 #if COMPILER_MSVC
-#	define TRAP() __debugbreak();
+#	define TRAP() __debugbreak()
 #elif COMPILER_GCC || COMPILER_CLANG
 #	define TRAP() __builtin_trap()
 #else
@@ -137,6 +137,18 @@ void SetPrintFunction(TPrintFunction *Function);
 void error(char *Text, ...) ATTR_PRINTF(1, 2);
 void print(int Level, char *Text, ...) ATTR_PRINTF(1, 2);
 
+bool isSpace(int c);
+bool isAlpha(int c);
+bool isEngAlpha(int c);
+bool isDigit(int c);
+int toLower(int c);
+int toUpper(int c);
+char *strLower(char *s);
+char *strUpper(char *s);
+int stricmp(const char *s1, const char *s2, int Pos);
+char *findFirst(char *s, char c);
+char *findLast(char *s, char c);
+
 struct TReadStream {
 	// VIRTUAL FUNCTIONS
 	// =========================================================================
diff --git a/src/config.cc b/src/config.cc
index 0be4eec..e3e3c85 100644
--- a/src/config.cc
+++ b/src/config.cc
@@ -128,10 +128,8 @@ void ReadConfig(void){
 			return;
 		}
 
-		// TODO(fusion): If the following `readSymbol` calls `nextToken`, then the
-		// buffer the identifier is in will get wiped. (UPDATE: Yes, the original
-		// version copies it into a local stack buffer, lol.)
-		const char *Identifier = Script.readIdentifier();
+		char Identifier[MAX_IDENT_LENGTH];
+		strcpy(Identifier, Script.readIdentifier());
 		Script.readSymbol('=');
 
 		// TODO(fusion): Ughh... Get rid of all `strcpy`s. A malicious configuration
diff --git a/src/map.cc b/src/map.cc
index 604fd6b..0d3f69b 100644
--- a/src/map.cc
+++ b/src/map.cc
@@ -735,7 +735,8 @@ static void ReadMapConfig(void){
 			break;
 		}
 
-		const char *Identifier = Script.readIdentifier();
+		char Identifier[MAX_IDENT_LENGTH];
+		strcpy(Identifier, Script.readIdentifier());
 		Script.readSymbol('=');
 
 		if(strcmp(Identifier, "sectorxmin") == 0){
diff --git a/src/script.cc b/src/script.cc
index 5dd3751..39141f7 100644
--- a/src/script.cc
+++ b/src/script.cc
@@ -15,108 +15,6 @@ static char ErrorString[100];
 
 // Helper Functions
 // =============================================================================
-// TODO(fusion): Move to `util.cc`?
-
-static bool isSpace(int c){
-	return c == ' '
-		|| c == '\t'
-		|| c == '\n'
-		|| c == '\r'
-		|| c == '\v'
-		|| c == '\f';
-}
-
-static bool isAlpha(int c){
-	// TODO(fusion): This is most likely wrong! We're assuming a direct conversion
-	// from `char` to `int` which will cause sign extension for negative values. This
-	// wouldn't be a problem if we expected to parse only streams of `char[]` but can
-	// be problematic for the output of `getc` which returns bytes as `unsigned char`
-	// converted to `int`.
-	//	TLDR: The parameter `c` should be `uint8`.
-	return ('A' <= c && c <= 'Z')
-		|| ('a' <= c && c <= 'z')
-		|| c == -0x1C	// E4 => ä
-		|| c == -0x0A	// F6 => ö
-		|| c == -0x04	// FC => ü
-		|| c == -0x3C	// C4 => Ä
-		|| c == -0x2A	// D6 => Ö
-		|| c == -0x24	// DC => Ü
-		|| c == -0x21;	// DF => ß
-}
-
-static bool isEngAlpha(int c){
-	return ('A' <= c && c <= 'Z')
-		|| ('a' <= c && c <= 'z');
-}
-
-static bool isDigit(int c){
-	return ('0' <= c && c <= '9');
-}
-
-static int toLower(int c){
-	// TODO(fusion): Same problem as `isAlpha`.
-	if(('A' <= c && c <= 'Z') || (0xC0 <= c && c <= 0xDE && c != 0xD7)){
-		c += 32;
-	}
-	return c;
-}
-
-static int toUpper(int c){
-	// TODO(fusion): Same problem as `isAlpha`.
-	if(('A' <= c && c <= 'Z') || (0xE0 <= c && c <= 0xFE && c != 0xF7)){
-		c -= 32;
-	}
-	return c;
-}
-
-static char *strLower(char *s){
-	for(int i = 0; s[i] != 0; i += 1){
-		s[i] = (char)toLower(s[i]);
-	}
-	return s;
-}
-
-static char *strUpper(char *s){
-	for(int i = 0; s[i] != 0; i += 1){
-		s[i] = (char)toUpper(s[i]);
-	}
-	return s;
-}
-
-static int stricmp(const char *s1, const char *s2, int Pos){
-	for(int i = 0; i < Pos; i += 1){
-		int c1 = toLower(s1[i]);
-		int c2 = toLower(s2[i]);
-		if(c1 > c2){
-			return 1;
-		}else if(c1 < c2){
-			return -1;
-		}else{
-			ASSERT(c1 == c2);
-			if(c1 == 0){
-				return 0;
-			}
-		}
-	}
-}
-
-static char *findFirst(char *s, char c){
-	return strchr(s, (int)c);
-}
-
-static char *findLast(char *s, char c){
-	char *Current = s;
-	char *Last = NULL;
-	while(true){
-		Current = strchr(Current, (int)c);
-		if(Current == NULL)
-			break;
-		Last = Current;
-		Current += 1; // skip character
-	}
-	return Last;
-}
-
 static void SaveFile(const char *Filename){
 	if(Filename == NULL){
 		error("SaveFile: Filename ist NULL.\n");
@@ -143,7 +41,7 @@ static void SaveFile(const char *Filename){
 
 	// NOTE(fusion): This function was in pretty bad shape after this point. The
 	// reason is probably because GHIDRA didn't make sense of it using `alloca`
-	// or some other mechanism to allocate a buffer on the stack to copy the whole
+	// or some other mechanism to allocate a buffer on the stack and copy the whole
 	// file in one go. It's probably not a good idea to mimic that here so we'll
 	// do it gradually.
 
@@ -535,7 +433,7 @@ void TReadScriptFile::nextToken(void){
 					while(true){
 						int next = getc(File);
 						if(isAlpha(next) || isDigit(next) || next == '_'){
-							if(IdentLength >= 30){
+							if(IdentLength >= (MAX_IDENT_LENGTH - 1)){
 								this->error("identifier too long");
 							}
 							this->String[IdentLength] = (char)next;
diff --git a/src/script.hh b/src/script.hh
index 59df86d..0aa6a78 100644
--- a/src/script.hh
+++ b/src/script.hh
@@ -4,6 +4,8 @@
 #include "common.hh"
 #include "enums.hh"
 
+#define MAX_IDENT_LENGTH 30
+
 struct TReadScriptFile {
 	// REGULAR FUNCTIONS
 	// =========================================================================
diff --git a/src/util.cc b/src/util.cc
index 30e1ba5..017e98b 100644
--- a/src/util.cc
+++ b/src/util.cc
@@ -41,6 +41,108 @@ void print(int Level, char *Text, ...){
 	}
 }
 
+// String Utility
+// =============================================================================
+bool isSpace(int c){
+	return c == ' '
+		|| c == '\t'
+		|| c == '\n'
+		|| c == '\r'
+		|| c == '\v'
+		|| c == '\f';
+}
+
+bool isAlpha(int c){
+	// TODO(fusion): This is most likely wrong! We're assuming a direct conversion
+	// from `char` to `int` which will cause sign extension for negative values. This
+	// wouldn't be a problem if we expected to parse only streams of `char[]` but can
+	// be problematic for the output of `getc` which returns bytes as `unsigned char`
+	// converted to `int`.
+	//	TLDR: The parameter `c` should be `uint8`.
+	return ('A' <= c && c <= 'Z')
+		|| ('a' <= c && c <= 'z')
+		|| c == -0x1C	// E4 => ä
+		|| c == -0x0A	// F6 => ö
+		|| c == -0x04	// FC => ü
+		|| c == -0x3C	// C4 => Ä
+		|| c == -0x2A	// D6 => Ö
+		|| c == -0x24	// DC => Ü
+		|| c == -0x21;	// DF => ß
+}
+
+bool isEngAlpha(int c){
+	return ('A' <= c && c <= 'Z')
+		|| ('a' <= c && c <= 'z');
+}
+
+bool isDigit(int c){
+	return ('0' <= c && c <= '9');
+}
+
+int toLower(int c){
+	// TODO(fusion): Same problem as `isAlpha`.
+	if(('A' <= c && c <= 'Z') || (0xC0 <= c && c <= 0xDE && c != 0xD7)){
+		c += 32;
+	}
+	return c;
+}
+
+int toUpper(int c){
+	// TODO(fusion): Same problem as `isAlpha`.
+	if(('A' <= c && c <= 'Z') || (0xE0 <= c && c <= 0xFE && c != 0xF7)){
+		c -= 32;
+	}
+	return c;
+}
+
+char *strLower(char *s){
+	for(int i = 0; s[i] != 0; i += 1){
+		s[i] = (char)toLower(s[i]);
+	}
+	return s;
+}
+
+char *strUpper(char *s){
+	for(int i = 0; s[i] != 0; i += 1){
+		s[i] = (char)toUpper(s[i]);
+	}
+	return s;
+}
+
+int stricmp(const char *s1, const char *s2, int Pos){
+	for(int i = 0; i < Pos; i += 1){
+		int c1 = toLower(s1[i]);
+		int c2 = toLower(s2[i]);
+		if(c1 > c2){
+			return 1;
+		}else if(c1 < c2){
+			return -1;
+		}else{
+			ASSERT(c1 == c2);
+			if(c1 == 0){
+				return 0;
+			}
+		}
+	}
+}
+
+char *findFirst(char *s, char c){
+	return strchr(s, (int)c);
+}
+
+char *findLast(char *s, char c){
+	char *Current = s;
+	char *Last = NULL;
+	while(true){
+		Current = strchr(Current, (int)c);
+		if(Current == NULL)
+			break;
+		Last = Current;
+		Current += 1; // skip character
+	}
+	return Last;
+}
+
 // TReadStream
 // =============================================================================
 bool TReadStream::readFlag(void){