#include "lexer.h"
#include "../../kerep/src/String/string.h"

typedef struct SharedLexerData{
    char* _source;
    char* _filename;
    Autoarr(Token)* _tokens;
    string _context;
    string _line;
    string _label;
    uint32 _linenum;
    uint32 _charnum;
} SharedLexerData;

#define source sld->_source
#define filename sld->_filename
#define tokens sld->_tokens
#define context sld->_context
#define line sld->_line
#define label sld->_label
#define linenum sld->_linenum
#define charnum sld->_charnum

Maybe _throw_wrongchar(SharedLexerData* sld){
    char* errline=string_extract(line);
    char* _context=string_extract(context);
    printf("\n\e[91mwrong char <%c> at [%s:%u:%u %s]\n >>> %s\n",
                source[charnum], filename,linenum,charnum,_context, errline);
    exit(96);
}
#define throw_wrongchar(freeMem) { freeMem; return _throw_wrongchar(sld); }

// adds <label> to <tokens> as tok_label or tok_number
void _tryAddLabel(SharedLexerData* sld){
    if(label.length==0) return;

    Unitype uni=ST_pullString(keywordsSearchTree, label);
    if(uni.VoidPtr!=NULL) // built-in keyword
        Autoarr_add(tokens, *(Token*)uni.VoidPtr);
    else {          // user-defined label
        Token ut;
        ut.value=string_extract(label);
        switch(*label.ptr){
            case '0': case '1': case '2': case '3': case '4':
            case '5': case '6': case '7': case '8': case '9':
                ut.id=tok_number;
                break;
            default:
                ut.id=tok_label;
                break;
        }
        Autoarr_add(tokens, ut);
    }
   
    label=(string){source, 0};
};
#define tryAddLabel() _tryAddLabel(sld)
   
#define addDefTok(id) tryAddLabel(); Autoarr_add(tokens, default_tokens[id])
   
void _addDefTok_ifnext(char next, TokenId yes, TokenId no, SharedLexerData* sld){
    if(*(source+1)==next){
        source++;
        addDefTok(yes);
    }
    else addDefTok(no);
}
#define addDefTok_ifnext(nextChar, yesTok, noTok) _addDefTok_ifnext(nextChar, yesTok, noTok, sld)

// returns text in quotes or error
Maybe _readString(char quotChar, SharedLexerData* sld){
    char c;
    bool prevIsBackslash=false;
    char* srcFirst=source;
    while ((c=*++source)){
        if(c==quotChar) {
            if(prevIsBackslash) // \"
                prevIsBackslash=false;
            else { // "
                string str={srcFirst, source-srcFirst+1};
                char* extracted=string_extract(str);
                return SUCCESS(UniHeapPtr(char, extracted));
            }
        }
        else prevIsBackslash= c=='\\' && !prevIsBackslash;
    }
    source=srcFirst;
    throw_wrongchar(;);
}
#define readString(quotChar) _readString(quotChar, sld)


Maybe _lexan(SharedLexerData* sld){
    char c;
    source--;
    while ((c=*++source)) {
        // tryAddLabel() resets label.length, but label.ptr still pointing to prev token beginning
        if(label.length==0)
            label.ptr=source;

        switch(c){
        case ' ': case '\t':
        case '\r': case '\n':
            tryAddLabel();
            break;
		
        case '(': addDefTok(tok_lbracket);    break;
        case '{': addDefTok(tok_lbracket_fi); break;
        case '[': addDefTok(tok_lbracket_sq); break;
        case ')': addDefTok(tok_rbracket);    break;
        case '}': addDefTok(tok_rbracket_fi); break;
        case ']': addDefTok(tok_rbracket_sq); break;
		
        case '\'':
            tryAddLabel();
            try(readString('\''), maybeC, ;){
                Token ctok={
                    .id=tok_character,
                    .value=(char*)maybeC.value.VoidPtr
                };
                Autoarr_add(tokens, ctok);
            }
            break;
        case '"':
            tryAddLabel();
            try(readString('"'), maybeS, ;){
                Token stok={
                    .id=tok_string,
                    .value=(char*)maybeS.value.VoidPtr
                };
                Autoarr_add(tokens, stok);
            }
            break;
		
        case '<': addDefTok(tok_less); break;
        case '>': addDefTok(tok_more); break;
        case '+': addDefTok(tok_plus); break;
        case '-': addDefTok(tok_minus); break;
        case '*': addDefTok(tok_asterisk); break;
		
        case '/':
            tryAddLabel();
            string commentStr={source, 1};
            c=*++source;
            if(c=='/') { // single-line comment
                commentStr.length++;
                while((c=*++source)){
                    if(c=='\n' || c=='\r') break;
                    else commentStr.length++;
                }
            }
            else if(c=='*') { // multi-line comment
                commentStr.length++;
                while((c=*++source)){
                    commentStr.length++;
                    if(c=='*' && *(++source)=='/') break;
                }
            }
            else { // not comment
                source--;
                addDefTok(tok_slash);
                break;
            }
            Token comTok={
                .value=string_extract(commentStr),
                .id=tok_comment
            };
            Autoarr_add(tokens, comTok);
            break;
		
        case '=': addDefTok_ifnext('=', tok_equals, tok_assign);      break;
        case '!': addDefTok_ifnext('=', tok_not_equals, tok_not);     break;
        case '&': addDefTok_ifnext('&', tok_and_d, tok_and);          break;
        case '|': addDefTok_ifnext('|', tok_or_d, tok_or);            break;
        case '?': addDefTok_ifnext('?', tok_question_d, tok_question);break;
        case ':': addDefTok(tok_colon);      break;
        case ';': addDefTok(tok_semicolon);  break;
        case '.': addDefTok(tok_point);      break;
        case ',': addDefTok(tok_comma);     break;
        case '~': addDefTok(tok_tilda);      break;
        case '\\': addDefTok(tok_backslash); break;
        case '%': addDefTok(tok_percent);    break;
        case '^': addDefTok(tok_xor);        break;
        case '$': addDefTok(tok_dollar);     break;
        case '@': addDefTok(tok_at);         break;
		
        default:
            label.length++;
            break;
        }
    }
    return SUCCESS(UniHeapPtr(Autoarr(Token), tokens));
}


Maybe lexan(char* _source, char* _filename){
    SharedLexerData sld={
        ._source=_source,
        ._filename=_filename,
        ._tokens=Autoarr_create(Token, 64, 1024),
        ._label={_source, 0},
        ._line={_source, 0},
        ._linenum=0,
        ._charnum=0
    };
    return _lexan(&sld);
}