view release on metacpan or  search on metacpan

src/compiler/lexer/Compiler_scanner.cpp  view on Meta::CPAN

#include <lexer.hpp>
using namespace std;
namespace TokenType = Enum::Token::Type;
namespace SyntaxType = Enum::Parser::Syntax;
namespace TokenKind = Enum::Token::Kind;

Scanner::Scanner() :
	isStringStarted(false), isRegexStarted(false), isPrototypeStarted(false), isFormatStarted(false),
	formatDeclaredToken(NULL), commentFlag(false), skipFlag(false),
	regex_delim(0), regex_middle_delim(0),
	brace_count_inner_regex(0), bracket_count_inner_regex(0), cury_brace_count_inner_regex(0)
{
	const char *regex_prefixes[] = {
		"q", "qq", "qw", "qx", "qr", "m", NULL
	};
	const char *regex_replaces[] = {
		"s", "y", "tr", NULL
	};
	const char *enable_regex_argument_funcs[] = {
		"map", "grep", "split", NULL
	};
	const char *operators[] = {
		"<=>", "**=", "//=", "||=", "&&=", "...", "$#{",
		"$^A", "$^D", "$^E", "$^F", "$^G", "$^H", "$^I",
		"$^L", "$^M", "$^O", "$^P", "$^R", "$^T", "$^W", "$^X",
		"<=",  ">=",  ".=",  "!=",  "==",  "+=",  "-=",
		"*=",  "%=",  "|=",  "&=",  "^=",  "<<",  ">>",
		"++",  "--",  "**",  "//",  "&&",  "||",  "::",
		"..",  "=>",  "->",  "@{",  "%{",  "${",  "@$",
		"%$",  "%-",  "%+",  "@-",  "@+",  "&$",  "$#",
		"<>",  "!~",  "~~",  "=~",
		"$0",  "$1",  "$2",  "$3",  "$4",  "$5",  "$6",
		"$7",  "$8",  "$9",
		"$&",  "$`",  "$'",  "$+",  "$.",  "$/",  "$|",
		"$,",  "$\\", "$\"", "$%",  "$=",  "$-",  "$~",
		"$^",  "$*",  "$:",  "$;",  "$?",  "$!",  "$@",
		/*"$$",*/  "$<",  "$>",  "$(",  "$)",  "$[",  "$]",
		NULL
	};
	const char *dereference_prefixes[] = {
		"@{", "%{", "${", "&{", "$#{", NULL
	};
	for (size_t i = 0; regex_prefixes[i] != NULL; i++) {
		regex_prefix_map.insert(StringMap::value_type(regex_prefixes[i], ""));
	}
	for (size_t i = 0; regex_replaces[i] != NULL; i++) {
		enable_regex_argument_func_map.insert(StringMap::value_type(enable_regex_argument_funcs[i], ""));
		regex_replace_map.insert(StringMap::value_type(regex_replaces[i], ""));
	}
	for (size_t i = 0; operators[i] != NULL; i++) {
		operator_map.insert(StringMap::value_type(operators[i], ""));
	}
	for (size_t i = 0; dereference_prefixes[i] != NULL; i++) {
		dereference_prefix_map.insert(StringMap::value_type(dereference_prefixes[i], ""));
	}
}

Token *Scanner::scanQuote(LexContext *ctx, char quote)
{
	TokenManager *tmgr = ctx->tmgr;
	ScriptManager *smgr = ctx->smgr;
	char prev_ch = smgr->previousChar();
	Token *prev_token = tmgr->lastToken();
	if (prev_token && prev_token->info.type == TokenType::RegExp) {
		return scanSymbol(ctx);
	}
	if (isalnum(prev_ch) || prev_ch == '_') {
		char *token = ctx->buffer();
		TokenInfo info = tmgr->getTokenInfo(token);
		char cur_ch = smgr->currentChar();
		if (cur_ch == '\'' && info.type == TokenType::Undefined) {
			Token *namespace_tk = tmgr->new_Token(token, ctx->finfo);
			namespace_tk->info = tmgr->getTokenInfo(TokenType::Namespace);
			tmgr->add(namespace_tk);
			ctx->clearBuffer();

			ctx->writeBuffer(cur_ch);
			Token *namespace_resolver = tmgr->new_Token(ctx->buffer(), ctx->finfo);
			namespace_resolver->info  = tmgr->getTokenInfo(TokenType::NamespaceResolver);
			ctx->clearBuffer();
			return namespace_resolver;
		} else if (info.kind == TokenKind::RegPrefix || info.kind == TokenKind::RegReplacePrefix) {
			Token *tk = tmgr->new_Token(token, ctx->finfo);
			tk->info = info;
			tmgr->add(tk);
			ctx->clearBuffer();
			return scanSymbol(ctx);
		} else {
			Token *tk = tmgr->new_Token(token, ctx->finfo);
			tk->info = info;
			tmgr->add(tk);
			ctx->clearBuffer();
		}
	}
	for (smgr->next(); !smgr->end(); smgr->next()) {
		char ch = smgr->currentChar();
		if (ch == '\n') {