/*
 * Copyright (C) 2016-2017 Redis Labs
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Affero General Public License as
 * published by the Free Software Foundation, either version 3 of the
 * License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU Affero General Public License for more details.
 *
 * You should have received a copy of the GNU Affero General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

#include "json_path.h"

int _tokenizePath(const char *json, size_t len, SearchPath *path, JSONSearchPathError_t *err) {
    tokenizerState st = S_NULL;
    size_t offset = 0;
    char *pos = (char *)json;
    token tok;
    tok.s = pos;
    tok.len = 0;
    char *jsperr = NULL;
    while (offset < len) {
        char c = *pos;
        switch (st) {
            // initial state
            case S_NULL: {
                switch (c) {
                    // . at the beginning means "root"
                    case '.':
                        tok.s++;
                        st = S_ROOT;
                        if (pos == json) {
                            path->hasLeadingDot = 1;
                        }
                        break;
                    // start of key/index specifier
                    case '[':
                        tok.s++;
                        st = S_BRACKET;
                        break;
                    default:
                        // only letters, dollar signs and underscores are allowed at the beginning
                        if (isalpha(c) || '$' == c || '_' == c) {
                            tok.len++;
                            st = S_IDENT;
                            break;
                        }

                        jsperr = JSON_PATH_IDENT_FIRST_CHAR_ERR;
                        goto syntaxerror;
                }
            } break;

            // we're after a square bracket opening
            case S_BRACKET:  // [
                // quotes after brackets means dict key
                if (c == '"') {
                    // skip to the beginnning of the key
                    tok.s++;
                    st = S_DKEY;
                } else if (c == '\'') {
                    // skip to the beginnning of the key
                    tok.s++;
                    st = S_SKEY;
                } else if (isdigit(c)) {
                    // digit after bracket means numeric index
                    tok.len++;
                    st = S_NUMBER;
                } else if ('-' == c) {
                    // this could be the beginning of a negative index
                    tok.len++;
                    st = S_MINUS;
                } else {
                    jsperr = JSON_PATH_BRACKET_FIRST_CHAR_ERR;
                    goto syntaxerror;
                }
                break;

            // we're after a dot
            case S_ROOT:
            case S_DOT:
                // start of ident token, can only be a letter, dollar sign or underscore
                if (isalpha(c) || '$' == c || '_' == c) {
                    tok.len++;
                    st = S_IDENT;
                } else {
                    jsperr = JSON_PATH_IDENT_FIRST_CHAR_ERR;
                    goto syntaxerror;
                }
                break;

            // we're within a number (array index)
            case S_NUMBER:
                if (isdigit(c)) {
                    tok.len++;
                    break;
                }
                if (c == ']') {
                    st = S_NULL;
                    tok.type = T_INDEX;
                    pos++;
                    offset++;
                    goto tokenend;
                }
                jsperr = JSON_PATH_NUMBER_ERR;
                goto syntaxerror;

            // we're within an ident string
            case S_IDENT:
                // end of ident
                if (c == '.' || c == '[') {
                    st = c == '.' ? S_DOT : S_BRACKET;
                    tok.type = T_KEY;
                    pos++;
                    offset++;
                    goto tokenend;
                }
                // we only allow letters, numbers, dollar signs and underscores in identifiers
                if (!isalnum(c) && '$' != c && '_' != c) {
                    jsperr = JSON_PATH_IDENT_ERR;
                    goto syntaxerror;
                }
                // advance one
                tok.len++;
                break;

            // we're within a bracketed string key
            case S_DKEY:
                // end of key
                if (c == '"') {
                    if (offset < len - 1 && *(pos + 1) == ']') {
                        tok.type = T_KEY;
                        pos += 2;
                        offset += 2;
                        st = S_NULL;
                        goto tokenend;
                    } else {
                        jsperr = JSON_PATH_MISSING_BRACKET_ERR;
                        goto syntaxerror;
                    }
                }
                tok.len++;
                break;
            case S_SKEY:
                // end of key
                if (c == '\'') {
                    if (offset < len - 1 && *(pos + 1) == ']') {
                        tok.type = T_KEY;
                        pos += 2;
                        offset += 2;
                        st = S_NULL;
                        goto tokenend;
                    } else {
                        jsperr = JSON_PATH_MISSING_BRACKET_ERR;
                        goto syntaxerror;
                    }
                }
                tok.len++;
                break;

            // we're within a negative index so we expect a digit now
            case S_MINUS:
                if (isdigit(c)) {
                    tok.len++;
                    st = S_NUMBER;
                } else {
                    jsperr = JSON_PATH_NEGATIVE_NUMBER_ERR;
                    goto syntaxerror;
                }
                break;
        }  // switch (st)
        offset++;
        pos++;

        // root or ident string must end if len reached
        if (len == offset && (S_IDENT == st || S_ROOT == st)) {
            st = S_NULL;
            tok.type = T_KEY;
            goto tokenend;
        }
        continue;

    tokenend : {
        if (T_INDEX == tok.type) {
            // convert the string to int. we can't use atoi because it expects
            // NULL termintated strings
            int64_t num = 0;
            for (int i = !isdigit(tok.s[0]); i < tok.len; i++) {
                int digit = tok.s[i] - '0';
                num = num * 10 + digit;
            }
            if ('-' == tok.s[0]) num = -num;
            SearchPath_AppendIndex(path, num);
        } else if (T_KEY == tok.type) {
            if ((1 == offset) && (1 == len) && '.' == c) {  // check for root
                SearchPath_AppendRoot(path);
            } else {
                SearchPath_AppendKey(path, tok.s, tok.len);
            }
        }
        tok.s = pos;
        tok.len = 0;
    }
    }  // while (offset < len)

    // these are the only legal states at the end of consuming the string
    if (st == S_NULL || st == S_IDENT || st == S_ROOT) {
        return PARSE_OK;
    }

syntaxerror:
    if (err) {
        err->errmsg = jsperr;
        err->offset = offset;
    }
    return PARSE_ERR;
}

int ParseJSONPath(const char *jsonPath, size_t len, SearchPath *path, JSONSearchPathError_t *err) {
    return _tokenizePath(jsonPath, len, path, err);
}