/*
    attocc - A minimal C compiler.
    Copyright (C) 2024 metamuffin

    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 <https://www.gnu.org/licenses/>.
*/
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>

const int NUM_KEYWORDS = 32;
static char *KEYWORDS[] = {
    "auto",     "break",  "case",    "char",   "const",    "continue",
    "default",  "do",     "double",  "else",   "enum",     "extern",
    "float",    "for",    "goto",    "if",     "int",      "long",
    "register", "return", "short",   "signed", "sizeof",   "static",
    "struct",   "switch", "typedef", "union",  "unsigned", "void",
    "volatile", "while"};

static char *SEPERATORS = "()[]{};,:";

const int NUM_OPERATORS = 36;
static char *OPERATORS[] = {
    "sizeof", "<<", ">>", "+=", "-=", "*=", "/=", "%=", "<<=",
    ">>=",    "&=", "|=", "^=", "++", "--", "==", "!=", "<=",
    ">=",     "||", "&&", "+",  "-",  "*",  "/",  "%",  "<",
    ">",      "~",  "&",  "|",  "^",  "!",  "&",  "*",  "=",
};
static char *OPERATOR_NAMES[] = {
    "OP_SIZEOF",
    "OP_SHIFT_LEFT",
    "OP_SHIFT_RIGHT",
    "OP_ADD_ASSIGN",
    "OP_SUB_ASSIGN",
    "OP_MUL_ASSIGN",
    "OP_DIV_ASSIGN",
    "OP_MOD_ASSIGN",
    "OP_SHIFT_LEFT_ASSIGN",
    "OP_SHIFT_RIGHT_ASSIGN",
    "OP_BITWISE_AND_ASSIGN",
    "OP_BITWISE_OR_ASSIGN",
    "OP_BITWISE_XOR_ASSIGN",
    "OP_INCREMENT",
    "OP_DECREMENT",
    "OP_EQUAL",
    "OP_NOT_EQUAL",
    "OP_LESS_EQUAL",
    "OP_GREATER_EQUAL",
    "OP_LOGICAL_OR",
    "OP_LOGICAL_AND",
    "OP_ADD",
    "OP_SUB",
    "OP_MUL",
    "OP_DIV",
    "OP_MOD",
    "OP_LESS",
    "OP_GREATER",
    "OP_BITWISE_NOT",
    "OP_BITWISE_AND",
    "OP_BITWISE_OR",
    "OP_BITWISE_XOR",
    "OP_LOGICAL_NOT",
    "OP_POINTER_REF",
    "OP_POINTER_DEREF",
    "OP_ASSIGN",
};
static char *KEYWORD_NAMES[] = {
    "KW_AUTO",     "KW_BREAK",   "KW_CASE",   "KW_CHAR",     "KW_CONST",
    "KW_CONTINUE", "KW_DEFAULT", "KW_DO",     "KW_DOUBLE",   "KW_ELSE",
    "KW_ENUM",     "KW_EXTERN",  "KW_FLOAT",  "KW_FOR",      "KW_GOTO",
    "KW_IF",       "KW_INT",     "KW_LONG",   "KW_REGISTER", "KW_RETURN",
    "KW_SHORT",    "KW_SIGNED",  "KW_SIZEOF", "KW_STATIC",   "KW_STRUCT",
    "KW_SWITCH",   "KW_TYPEDEF", "KW_UNION",  "KW_UNSIGNED", "KW_VOID",
    "KW_VOLATILE", "KW_WHILE",
};
static char *SEPERATOR_NAMES[] = {
    "SEP_LPAREN", "SEP_RPAREN",    "SEP_LSQUARE", "SEP_RSQUARE", "SEP_LCURLY",
    "SEP_RCURLY", "SEP_SEMICOLON", "SEP_COMMA",   "SEP_DOT",     "SEP_COLOR",
};

enum keyword {
  KW_AUTO,
  KW_BREAK,
  KW_CASE,
  KW_CHAR,
  KW_CONST,
  KW_CONTINUE,
  KW_DEFAULT,
  KW_DO,
  KW_DOUBLE,
  KW_ELSE,
  KW_ENUM,
  KW_EXTERN,
  KW_FLOAT,
  KW_FOR,
  KW_GOTO,
  KW_IF,
  KW_INT,
  KW_LONG,
  KW_REGISTER,
  KW_RETURN,
  KW_SHORT,
  KW_SIGNED,
  KW_SIZEOF,
  KW_STATIC,
  KW_STRUCT,
  KW_SWITCH,
  KW_TYPEDEF,
  KW_UNION,
  KW_UNSIGNED,
  KW_VOID,
  KW_VOLATILE,
  KW_WHILE,
};

enum operator{
  OP_SIZEOF,     // len=6
  OP_SHIFT_LEFT, // len=2
  OP_SHIFT_RIGHT,
  OP_ADD_ASSIGN,
  OP_SUB_ASSIGN,
  OP_MUL_ASSIGN,
  OP_DIV_ASSIGN,
  OP_MOD_ASSIGN,
  OP_SHIFT_LEFT_ASSIGN,
  OP_SHIFT_RIGHT_ASSIGN,
  OP_BITWISE_AND_ASSIGN,
  OP_BITWISE_OR_ASSIGN,
  OP_BITWISE_XOR_ASSIGN,
  OP_INCREMENT,
  OP_DECREMENT,
  OP_EQUAL,
  OP_NOT_EQUAL,
  OP_LESS_EQUAL,
  OP_GREATER_EQUAL,
  OP_LOGICAL_OR,
  OP_LOGICAL_AND,
  OP_ADD, // len=1
  OP_SUB,
  OP_MUL,
  OP_DIV,
  OP_MOD,
  OP_LESS,
  OP_GREATER,
  OP_BITWISE_NOT,
  OP_BITWISE_AND,
  OP_BITWISE_OR,
  OP_BITWISE_XOR,
  OP_LOGICAL_NOT,
  OP_POINTER_REF,
  OP_POINTER_DEREF,
  OP_ASSIGN,
};

enum seperator {
  SEP_LPAREN,
  SEP_RPAREN,
  SEP_LSQUARE,
  SEP_RSQUARE,
  SEP_LCURLY,
  SEP_RCURLY,
  SEP_SEMICOLON,
  SEP_COMMA,
  SEP_DOT,
  SEP_COLOR,
};

enum token_kind {
  TOK_IDENTIFIER,
  TOK_KEYWORD,
  TOK_INT_CONSTANT,
  TOK_SEPERATOR,
  TOK_OPERATOR,
  TOK_END,
};

union token_data {
  char *identifier;
  enum keyword keyword;
  int constant_value;
  enum seperator seperator;
  enum operator operator;
};

struct token {
  enum token_kind kind;
  union token_data data;
};

char is_numeric(char c) { return c >= '0' && c <= '9'; }
char is_octal(char c) { return c >= '0' && c <= '7'; }
char is_hexadecial(char c) {
  return (c >= '0' && c <= '9') || (c >= 'A' && c <= 'F') ||
         (c >= 'a' && c <= 'f');
}
char is_alpha(char c) {
  return (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z');
}
char is_alphanumeric(char c) { return is_alpha(c) || is_numeric(c); }

struct token *tokenize(char *source) {
  char *end = source + strlen(source);
  char *p = source;
  unsigned long num_tokens = 0;
  struct token *tokens = NULL;
  while (*p) {
    unsigned long remaining = end - p;

    //* whitespace
    if (p[0] == ' ' || p[0] == '\t' || p[0] == '\n') {
      p++;
      continue;
    }

    //* comments
    if (remaining >= 2) {
      if (p[0] == '/' && p[1] == '/') {
        p += 2;
        for (char c; (c = *p++) && c != '\n';)
          ;
        continue;
      } else if (p[0] == '/' && p[1] == '*') {
        p += 2;
        for (char c, d = '\0'; (d = c, c = *p++) && c == '/' && d == '*';)
          ;
        continue;
      } else if (p[0] == '#') {
        p += 1;
        for (char c; (c = *p++) && c != '\n';)
          ;
        continue;
      }
    }

    //* seperators
    char match = 0;
    for (int i = 0; SEPERATORS[i]; i++) {
      if (p[0] == SEPERATORS[i]) {
        p++;
        num_tokens += 1;
        tokens = realloc(tokens, sizeof(struct token) * num_tokens);
        if (!tokens) {
          fprintf(stderr, "realloc failed\n");
          return NULL;
        }
        struct token *new_token = &tokens[num_tokens - 1];
        new_token->kind = TOK_SEPERATOR;
        new_token->data.seperator = i;
        match = 1;
        break;
      }
    }
    if (match)
      continue;

    //* operators
    match = 0;
    for (int i = 0; i < NUM_OPERATORS; i++) {
      char *op = OPERATORS[i];
      if (remaining >= strlen(op)) {
        if (strncmp(op, p, strlen(op)) == 0) {
          num_tokens += 1;
          tokens = realloc(tokens, sizeof(struct token) * num_tokens);
          if (!tokens) {
            fprintf(stderr, "realloc failed\n");
            return NULL;
          }
          struct token *new_token = &tokens[num_tokens - 1];
          new_token->kind = TOK_OPERATOR;
          new_token->data.operator= i;

          p += strlen(op);
          match = 1;
          break;
        }
      }
    }
    if (match)
      continue;

    //* keyword
    match = 0;
    for (int i = 0; i < NUM_KEYWORDS; i++) {
      char *kw = KEYWORDS[i];
      if (remaining >= strlen(kw)) {
        if (strncmp(kw, p, strlen(kw)) == 0) {
          num_tokens += 1;
          tokens = realloc(tokens, sizeof(struct token) * num_tokens);
          if (!tokens) {
            fprintf(stderr, "realloc failed\n");
            return NULL;
          }
          struct token *new_token = &tokens[num_tokens - 1];
          new_token->kind = TOK_KEYWORD;
          new_token->data.keyword = i;

          // printf("kw match %i %li\n", i, strlen(kw));
          p += strlen(kw);
          match = 1;
          break;
        }
      }
    }
    if (match)
      continue;

    if (is_numeric(p[0])) {
      int value = 0;
      if (remaining >= 2 && p[1] == 'x') {
        p += 2;
        for (char c; (c = *p++) && is_hexadecial(c);) {
          value *= 0x10;
          value += c <= '9' ? c - '0' : 10 + (c <= 'F' ? c - 'A' : c - 'a');
        }
      } else if (p[0] == '0') {
        p += 1;
        for (char c; (c = *p++) && is_octal(c);) {
          value *= 010;
          value += c - '0';
        }
      } else {
        for (char c; (c = *p++) && is_numeric(c);) {
          value *= 10;
          value += c - '0';
        }
      }
      p--;

      num_tokens += 1;
      tokens = realloc(tokens, sizeof(struct token) * num_tokens);
      if (!tokens) {
        fprintf(stderr, "realloc failed\n");
        return NULL;
      }
      struct token *new_token = &tokens[num_tokens - 1];
      new_token->kind = TOK_INT_CONSTANT;
      new_token->data.constant_value = value;
      continue;
    }

    if (is_alpha(p[0])) {
      char *ident_start = p;
      p++;
      for (char c; (c = *p++) && is_alphanumeric(c);)
        ;
      p--;
      int ident_len = p - ident_start - 1;
      char *ident_str = malloc(ident_len + 1);
      if (!ident_str) {
        fprintf(stderr, "malloc failed\n");
        return NULL;
      }
      for (int i = 0; i < ident_len; i++)
        ident_str[i] = ident_start[i];
      ident_str[ident_len] = '\0';

      num_tokens += 1;
      tokens = realloc(tokens, sizeof(struct token) * num_tokens);
      if (!tokens) {
        fprintf(stderr, "realloc failed\n");
        return NULL;
      }
      struct token *new_token = &tokens[num_tokens - 1];
      new_token->kind = TOK_IDENTIFIER;
      new_token->data.identifier = ident_str;

      continue;
    }

    fprintf(stderr, "unknown token at %li\n", p - source);
    return NULL;
  }

  num_tokens += 1;
  tokens = realloc(tokens, sizeof(struct token) * num_tokens);
  if (!tokens) {
    fprintf(stderr, "realloc failed\n");
    return NULL;
  }
  struct token *new_token = &tokens[num_tokens - 1];
  new_token->kind = TOK_END;

  return tokens;
}

void debug_tokens(struct token *tokens) {
  for (int i = 0; tokens[i].kind != TOK_END; i++) {
    switch (tokens[i].kind) {
    case TOK_IDENTIFIER:
      printf("TOK_IDENTIFIER:%s, ", tokens[i].data.identifier);
      break;
    case TOK_KEYWORD:
      printf("TOK_KEYWORD:%s, ", KEYWORD_NAMES[tokens[i].data.keyword]);
      break;
    case TOK_INT_CONSTANT:
      printf("TOK_CONSTANT:%i, ", tokens[i].data.constant_value);
      break;
    case TOK_OPERATOR:
      printf("TOK_OPERATOR:%s, ", OPERATOR_NAMES[tokens[i].data.operator] );
      break;
    case TOK_SEPERATOR:
      printf("TOK_SEPERATOR:%s, ", SEPERATOR_NAMES[tokens[i].data.seperator]);
      break;
    case TOK_END:
      break;
    }
  }
  printf("TOK_END\n");
}

int main(int argc, char **argv) {
  if (argc < 3) {
    fprintf(stderr, "USAGE:\n\tattocc <input> <output>\n");
    return 1;
  }
  char *input = argv[1];
  char *output = argv[2];

  int input_fd = open(input, O_RDONLY | O_CLOEXEC);
  if (input_fd < 0) {
    perror("cannot open input");
    return 1;
  }

  int output_fd = open(output, O_WRONLY | O_TRUNC | O_CREAT | O_CLOEXEC, 0640);
  if (output_fd < 0) {
    perror("cannot open input");
    return 1;
  }

  int source_len = 0;
  char *source = NULL;
  int size;
  char buffer[4096];

  while ((size = read(input_fd, &buffer, 4096))) {
    if (size < 0) {
      perror("cannot read source");
      return 1;
    }

    source_len += size;
    source = realloc(source, source_len + 1);
    if (!source) {
      fprintf(stderr, "malloc failed\n");
      return 1;
    }

    for (int i = 0; i < size; i++)
      source[source_len - size + i] = buffer[i];
  }
  source[source_len] = '\0';

  struct token *tokens = tokenize(source);
  if (!tokens)
    return 1;
  debug_tokens(tokens);
  free(tokens);

  return 0;
}