struct Token get_special(char c)
{
struct Token tokenOneChar, tokenTwoChar, tokenFinal;
TokenCode codeOneChar, codeTwoChar, codeFinal;
tokenOneChar.literalType = STRING_LIT;
tokenTwoChar.literalType = STRING_LIT;
tokenOneChar.literalValue.valString[0] = c;
tokenOneChar.literalValue.valString[1] = '\0';
tokenTwoChar.literalValue.valString[0] = c;
tokenTwoChar.literalValue.valString[1] = peek_char();
tokenTwoChar.literalValue.valString[2] = '\0';
codeOneChar = is_reserved_word(tokenOneChar.literalValue.valString);
codeTwoChar = is_reserved_word(tokenTwoChar.literalValue.valString);
if(codeTwoChar != NO_TOKEN) {
get_char();
codeFinal = codeTwoChar;
tokenFinal = tokenTwoChar;
} else {
codeFinal = codeOneChar;
tokenFinal = tokenOneChar;
}
tokenFinal.tokenCode = codeFinal;
return tokenFinal;
}
token create_string_token(char * s, int is_string_constant)
{ token t;
if (is_string_constant)
{ t.type = string_constant;
t.detail.name = s; }
else if (is_reserved_word(s))
{ t.type = reserved_word;
t.detail.id = is_reserved_word(s); }
else
{ t.type = identifier;
t.detail.name = s; }
return t; }
static char* get_word(Token* theToken, char token_string[MAX_TOKEN_STRING_LENGTH],char *token_ptr)
{
/*
Write some code to Extract the word
*/
int charCount = 0;
while((char_table[(*(token_ptr))] == LETTER || char_table[(*(token_ptr))] == DIGIT) && charCount < (MAX_TOKEN_STRING_LENGTH-1))
{
token_string[charCount] = *(token_ptr);
token_string[charCount+1] = '\0';
charCount++;
token_ptr++;
}
theToken->literal_value = downshift_word(token_string); //Downshift the word, to make it lower case
/*
Write some code to Check if the word is a reserved word.
if it is not a reserved word its an identifier.
*/
if(is_reserved_word(theToken))
{
theToken->literal_type = STRING_LIT;
}else
{
theToken->literal_type = STRING_LIT;
theToken->token_code = IDENTIFIER;
theToken->nextToken = NULL;
}
return token_ptr;
}
static Token *get_word(char token_string[], Token *token2) {
/*
Write some code to Extract the word
Downshift the word, to make it lower case
Write some code to Check if the word is a reserved word.
if it is not a reserved word its an identifier.
*/
downshift_word(token_string);
token2->nextptr = NULL;
token2->type = STRING_LIT;
strcpy(token2->token_string, token_string);
/*
>> Write some code to Check if the word is a reserved word. if it is not a reserved word its an identifier
>> use the arrays provided in the source code I gave out to find out if something is a reserved word or not. If not then this token is an identifier
*/
is_reserved_word(token_string, token2);
return token2;
}
bool Grammar::is_identifier(const std::string& str) {
if (str.size() == 0 || !str_utils::is_alpha(str[0]))
return false;
for (int i = 1; i < int(str.size()); i++)
if (!str_utils::is_alphanumeric(str[i]) && str[i] != '_')
return false;
return !is_reserved_word(str);
}
int read_next_token(TokenCacheManager *tcm, Token **token, pval *phplval)
{
/* YY_TLS_VARS;
TLS_VARS; */
if (GLOBAL(tc)->count == GLOBAL(tc)->pos || GLOBAL(php3_display_source)) {
/* we need to read from the lexical scanner */
Token next_token;
phplval->type = IS_LONG; /* the lex scanner doesn't always set phplval->type, make sure the type is not 'dirty' */
phplval->cs_data.switched = 0;
next_token.token_type = lex_scan(phplval);
/*printf("Read token: %c (%d)\n",next_token.token_type,next_token.token_type);*/
if (next_token.token_type == DONE_EVAL) {
return DONE_EVAL;
}
if (GLOBAL(tc)->count >= GLOBAL(tc)->max_tokens) { /* we need more memory */
GLOBAL(tc)->tokens = (Token *) erealloc(GLOBAL(tc)->tokens,
(sizeof(Token) * (GLOBAL(tc)->max_tokens += GLOBAL(tc)->block_size)));
if (!GLOBAL(tc)->tokens) {
return FAILURE;
}
}
next_token.phplval = *phplval;
next_token.lineno = GLOBAL(phplineno);
next_token.phplval.offset = (tcm->active<<TOKEN_BITS) + GLOBAL(tc)->count;
/* ugly hack to support $keyword */
if (last_token_suggests_variable_reference() && is_reserved_word(next_token.token_type)) {
next_token.phplval.value.str.val = estrndup(phptext,phpleng);
next_token.phplval.value.str.len = phpleng;
next_token.phplval.type = IS_STRING;
next_token.token_type = STRING;
}
/* end of ugly hack */
last_token_type=next_token.token_type;
if (GLOBAL(php3_display_source)) {
syntax_highlight(&next_token);
*token = &next_token;
return (*token)->token_type;
}
GLOBAL(tc)->tokens[GLOBAL(tc)->count] = next_token;
GLOBAL(tc)->count++;
}
*token = &GLOBAL(tc)->tokens[GLOBAL(tc)->pos++];
return (*token)->token_type;
}
char* get_word(char** original, char ch[], char* ch_ptr, Token* t)
{
/*
Write some code to Extract the word
*/
ch_ptr = ch;
int i = 1;
while(isalpha(get_char(original))){
ch[i] = get_char(original);
i++;
}
ch[i] = '\0';
//Downshift the word, to make it lower case
downshift_word(ch);
/*
Write some code to Check if the word is a reserved word.
if it is not a reserved word its an identifier.
*/
if(is_reserved_word(ch, t) == FALSE){
t -> token_code = IDENTIFIER;
}
return ch_ptr;
}
struct Token get_word(char c)
{
struct Token token;
TokenCode code;
int i = 1;
token.literalType = STRING_LIT;
token.literalValue.valString[0] = c;
while(char_table[peek_char()] == LETTER || char_table[peek_char()] == NUMBER) {
c = get_char();
token.literalValue.valString[i] = c;
i++;
}
while(i < MAX_TOKEN_STRING_LENGTH) {
token.literalValue.valString[i] = '\0';
i++;
}
downshift_word(token.literalValue.valString);
code = is_reserved_word(token.literalValue.valString);
token.tokenCode = (code == NO_TOKEN)? IDENTIFIER : code;
return token;
}
token_list tokenize(std::string source_) {
std::list<token> token_stream;
std::string token_str;
bool in_string = false;
bool is_comment_line = false;
bool is_comment_block = false;
std::string string_quote_type = "'";
std::regex replace_macros("[a-zA-Z0-9_]+?\\([a-zA-Z0-9_\\(\\), ]*?\\)");
source_ = std::regex_replace(source_, replace_macros, "MACRO_REPLACED_BY_TOKENIZE");
//std::cout << source_;
for (std::string::size_type stream_pos = 0; stream_pos < source_.size(); ++stream_pos) {
std::string cur_char = source_.substr(stream_pos, 1);
if (!in_string) {
if (cur_char == "\"" || cur_char == "'" || cur_char == " " || cur_char == "\t" ||
cur_char == "+" ||
cur_char == "-" ||
cur_char == "*" ||
cur_char == "/" ||
cur_char == ">" ||
cur_char == "<" ||
cur_char == "=" ||
cur_char == "]" ||
cur_char == "[" ||
cur_char == "(" ||
cur_char == ")" ||
cur_char == "}" ||
cur_char == "{" ||
cur_char == "," ||
cur_char == "!" ||
cur_char == "." ||
cur_char == ":" ||
cur_char == ";"
) {
if (token_str != "") {
if (!is_comment_line && !is_comment_block) {
token new_token;
new_token.val = token_str;
if (is_number(token_str)) {
new_token.type = type::SCALAR;
}
else {
new_token.type = type::LITERAL;
}
new_token.stream_pos = stream_pos - token_str.length();
if (token_str[0] == '_')
new_token.type = type::LOCALVARIABLE;
if (is_reserved_word(token_str))
new_token.type = type::SCRIPTCOMMAND;
token_stream.push_back(new_token);
}
}
if (cur_char == "+" ||
cur_char == "-" ||
cur_char == "*" ||
cur_char == "/" ||
cur_char == ">" ||
cur_char == "<" ||
cur_char == "=" ||
cur_char == "]" ||
cur_char == "[" ||
cur_char == "(" ||
cur_char == ")" ||
cur_char == "}" ||
cur_char == "{" ||
cur_char == "," ||
cur_char == "!" ||
cur_char == "." ||
cur_char == ":" ||
cur_char == ";"
) {
bool skip = false;
if (cur_char == ">" ||
cur_char == "<" ||
cur_char == "!" ||
cur_char == "="
) {
if (source_.substr(stream_pos + 1, 1) == "=") {
skip = true;
cur_char += "=";
}
}
if (cur_char == ":") {
if (source_.substr(stream_pos + 1, 1) == ":") {
skip = true;
cur_char += ":";
}
}
if (cur_char == "/") {
if (source_.substr(stream_pos + 1, 1) == "*") {
skip = true;
cur_char += "*";
is_comment_block = true;
}
}
if (cur_char == "/") {
if (source_.substr(stream_pos + 1, 1) == "/") {
skip = true;
cur_char += "/";
is_comment_line = true;
}
}
if (!is_comment_line && !is_comment_block) {
token operator_token;
operator_token.val = cur_char;
operator_token.stream_pos = stream_pos;
operator_token.type = type::OPERATOR;
if (cur_char == ";") {
operator_token.type = type::ENDOFSTATEMENT;
}
token_stream.push_back(operator_token);
}
if (cur_char == "*") {
if (source_.substr(stream_pos + 1, 1) == "/") {
skip = true;
cur_char += "/";
is_comment_block = false;
token_str = "";
}
}
if (skip) {
stream_pos++;
}
}
token_str = "";
if (cur_char == "\"") {
token_str = "\"" + token_str;
}
if (cur_char == "'") {
token_str = "'" + token_str;
}
}
else {
if (
token_str == "<=" ||
token_str == ">=" ||
token_str == "!=" ||
token_str == "::" ||
token_str == "=="
) {
if (!is_comment_line && !is_comment_block) {
token new_token;
new_token.val = token_str;
new_token.stream_pos = stream_pos - token_str.length();
new_token.type = type::OPERATOR;
token_stream.push_back(new_token);
}
token_str = "";
}
else {
if (cur_char != "\n") {
token_str += cur_char;
}
else {
if (!is_comment_line && !is_comment_block && token_str!= "") {
token new_token;
new_token.val = token_str;
if (is_number(token_str)) {
new_token.type = type::SCALAR;
}
else {
new_token.type = type::LITERAL;
}
new_token.stream_pos = stream_pos - token_str.length();
if (token_str[0] == '_')
new_token.type = type::LOCALVARIABLE;
if (is_reserved_word(token_str))
new_token.type = type::SCRIPTCOMMAND;
token_stream.push_back(new_token);
}
if (is_comment_line) {
is_comment_line = false;
token_str = "";
}
}
}
}
}
else {
token_str += cur_char;
}
if (!is_comment_line && !is_comment_block) {
if (cur_char == "\"" || cur_char == "'") {
if (!in_string) {
string_quote_type = cur_char;
in_string = true;
}
else if(in_string && cur_char == string_quote_type) {
if (source_.substr(stream_pos + 1, 1) == string_quote_type) {
token_str += string_quote_type;
stream_pos++;
}
else {
in_string = false;
}
}
}
}
}
return token_stream;
}
