int checkSegment(FILE *f, char ending)
{
struct ht_Table *status = ht_create(NUMBER_OF_SYMBOLS);
struct ht_Table *opposites = ht_create(NUMBER_OF_OPPOSITES);
initializeStatus(status);
initializeOpposites(opposites);
char CLOSERS[] = {')', '>', ']', '}'};
char OPENERS[] = {'(', '<', '[', '{'};
const int LISTS_LENGTH = 4;
char last = 'a';
char curr;
while((curr = fgetc(f)) != EOF)
{
printf("%c", curr);
if(curr == '#')
skipLineComment(f);
else if(curr == '*' && last == '/')
skipBlockComment(f);
else if(curr == '/' && last == '/')
skipLineComment(f);
else if(curr == '\'' || curr == '"')
{
if(skipString(f, curr) != 0)
return 1;
}
else if(search(OPENERS, LISTS_LENGTH, curr) != -1)
{
(*ht_get(status, curr))++;
}
// Struct pointer references screw up the normal algorithm. Don't tell Tim Peters or Linus Torvalds
else if(curr == '>' && last == '-')
{}
else if(search(CLOSERS, LISTS_LENGTH, curr) != -1)
{
char correspondingOpener = (char) (*ht_get(opposites, curr));
int *num = ht_get(status, correspondingOpener);
if(*num == 0)
{
printf("There is an unmatched %c!\n", curr);
return 1;
}
else
{
(*num)--;
}
}
last = curr;
}
return 0;
}
void getChar(void)
{
if (*bufferp == nullptr)
{
if (!getSourceLine())
{
if (NumOpenFiles > 1)
{
closeSourceFile();
return;
}
else
{
curChar = CHAR_EOF;
return;
}
}
bufferp = sourceBuffer;
bufferOffset = 0;
}
curChar = *bufferp;
bufferp++;
if (DumbGetCharOn)
{
bufferOffset++;
return;
}
//-------------------
// Special Characters
switch (curChar)
{
case '\t':
// TAB
bufferOffset += TAB_SIZE - bufferOffset % TAB_SIZE;
curChar = ' ';
break;
case '\n':
case '\r':
// NEWLINE
bufferOffset++;
curChar = ' ';
break;
case '/':
if (*bufferp == '/')
skipLineComment();
else if (*bufferp == '*')
skipBlockComment();
else
bufferOffset++;
break;
case '#':
languageDirective();
break;
default:
bufferOffset++;
}
}
void skipBlockComment(FILE *f)
{
char last = 'a';
char curr;
while((curr = fgetc(f)) != '/' && last != '*')
{
last = curr;
}
skipLineComment(f);
return;
}
/* Tokenizer::readToken
* Reads the next 'token' from the text & moves past it
*******************************************************************/
void Tokenizer::readToken(bool toeol)
{
token_current.clear();
bool ready = false;
qstring = false;
// Increment pointer to next token
while (!ready)
{
ready = true;
// Increment pointer until non-whitespace is found
while (isWhitespace(current[0]))
{
// Return if end of text found
if (!incrementCurrent())
return;
}
// Skip C-style comments
if (comments & CCOMMENTS)
{
// Check if we have a line comment
if (current + 1 < end && current[0] == '/' && current[1] == '/')
{
ready = false;
// DECORATE //$ handling
if (!decorate)
skipLineComment();
else if (current + 2 < end && current[2] != '$')
skipLineComment();
else
ready = true;
}
// Check if we have a multiline comment
if (current + 1 != end && current[0] == '/' && current[1] == '*')
{
skipMultilineComment(); // Skip it
ready = false;
}
}
// Skip '##' comments
if (comments & DCOMMENTS)
{
if (current + 1 != end && current[0] == '#' && current[1] == '#')
{
skipLineComment(); // Skip it
ready = false;
}
}
// Skip '#' comments
if (comments & HCOMMENTS)
{
if (current + 1 != end && current[0] == '#')
{
skipLineComment(); // Skip it
ready = false;
}
}
// Skip ';' comments
if (comments & SCOMMENTS)
{
if (current[0] == ';')
{
skipLineComment(); // Skip it
ready = false;
}
}
// Check for end of text
if (position == size)
return;
}
// Init token delimiters
t_start = position;
t_end = position;
// If we're at a special character, it's our token
if (isSpecialCharacter(current[0]))
{
token_current += current[0];
t_end = position + 1;
incrementCurrent();
return;
}
// Now read the token
if (current[0] == '\"') // If we have a literal string (enclosed with "")
{
qstring = true;
// Skip opening "
incrementCurrent();
// Read literal string (include whitespace)
while (current[0] != '\"')
{
//if (position < size - 1 && current[0] == '\\' && current[1] == '\"')
if (current[0] == '\\')
incrementCurrent();
token_current += current[0];
if (!incrementCurrent())
return;
}
// Skip closing "
incrementCurrent();
}
else
{
// Read token (don't include whitespace)
while (!((!toeol && isWhitespace(current[0])) || current[0] == '\n'))
{
// Return if special character found
if (!toeol && isSpecialCharacter(current[0]))
return;
// Add current character to the token
token_current += current[0];
// Return if end of text found
if (!incrementCurrent())
return;
}
}
// Write token to log if debug mode enabled
if (debug)
wxLogMessage(token_current);
// Return the token
return;
}
temp<Token> Lexer::readToken()
{
while (true)
{
if (isDone()) return Token::create(TOKEN_EOF, String::create(""));
start_ = pos_;
char c = advance();
switch (c)
{
case ' ':
case '\t':
case '\r':
// Skip whitespace.
while (isWhitespace(peek())) advance();
break;
case '(': return makeToken(TOKEN_LEFT_PAREN);
case ')': return makeToken(TOKEN_RIGHT_PAREN);
case '[': return makeToken(TOKEN_LEFT_BRACKET);
case ']': return makeToken(TOKEN_RIGHT_BRACKET);
case '{': return makeToken(TOKEN_LEFT_BRACE);
case '}': return makeToken(TOKEN_RIGHT_BRACE);
case '=': return makeToken(TOKEN_EQUALS);
case '+': return makeToken(TOKEN_PLUS);
case '-': return makeToken(TOKEN_MINUS);
case '*': return makeToken(TOKEN_STAR);
case '%': return makeToken(TOKEN_PERCENT);
case '<': return makeToken(TOKEN_LESS_THAN);
case '\n': return makeToken(TOKEN_LINE);
case '/':
if (peek() == '/')
{
skipLineComment();
/*}
else if (peek() == '*')
{
skipBlockComment();*/
}
else
{
return makeToken(TOKEN_SLASH);
}
break;
/*
case ',': return singleToken(TOKEN_LINE);
case '@': return singleToken(TOKEN_AT);
case '.': return singleToken(TOKEN_DOT);
case '#': return singleToken(TOKEN_HASH);
case ';': return singleToken(TOKEN_SEMICOLON);
case '\\': return singleToken(TOKEN_IGNORE_LINE);
case '|': return singleToken(TOKEN_PIPE);
case ':':
advance();
if (peek() == ':')
{
// "::".
advance();
return Ref<Token>(new Token(TOKEN_BIND));
}
// Just a ":" by itself.
return Ref<Token>(new Token(TOKEN_KEYWORD, ":"));
case '-':
advance();
if (isDigit(peek())) return readNumber();
return readOperator();
case '/':
advance();
if (peek() == '/')
{
// Line comment, so ignore the rest of the line and
// emit the line token.
mNeedsLine = true;
return Ref<Token>(new Token(TOKEN_LINE));
}
else if (peek() == '*')
{
skipBlockComment();
}
else
{
return readOperator();
}
break;
case '"': return readString();
default:
if (isDigit(c)) return readNumber();
if (isOperator(c)) return readOperator();
*/
default:
if (isNameStart(c)) return readName();
if (isDigit(c)) return readNumber();
// If we got here, we don't know what it is.
return makeToken(TOKEN_ERROR);
}
}
}