diff options
| -rw-r--r-- | .gitignore | 1 | ||||
| -rw-r--r-- | Makefile | 8 | ||||
| -rw-r--r-- | main.c | 685 |
3 files changed, 634 insertions, 60 deletions
@@ -3,3 +3,4 @@ main m68k-trasm cmake[-_]build*/ build*/ +compile_commands.json @@ -1,12 +1,12 @@ # SPDX-License-Identifier: Unlicense WARNFLAGS = -Wall -Wextra -pedantic -Wlogical-op -OPTFLAGS = -O2 INCLUDES = lib -_CFLAGS = $(CFLAGS) $(WARNFLAGS) $(addprefix -I,$(INCLUDES)) $(OPTFLAGS) -pipe -g -_CXXFLAGS = $(CXXFLAGS) $(WARNFLAGS) $(addprefix -I,$(INCLUDES)) $(OPTFLAGS) -pipe -g +_FLAGS = -O2 -fsanitize=address +_CFLAGS = $(CFLAGS) $(WARNFLAGS) $(addprefix -I,$(INCLUDES)) $(_FLAGS) -pipe -g +_CXXFLAGS = $(CXXFLAGS) $(WARNFLAGS) $(addprefix -I,$(INCLUDES)) $(_FLAGS) -pipe -g LDSCRIPTS = -_LDFLAGS = $(LDFLAGS) $(OPTFLAGS) $(addprefix -T,$(LDSCRIPTS)) +_LDFLAGS = $(LDFLAGS) $(_FLAGS) $(addprefix -T,$(LDSCRIPTS)) OBJECTS=main.o @@ -12,55 +12,90 @@ #include <stdlib.h> #include <stdint.h> -#define OK 0 -#define ERR 2 - -enum tok_kind { - TK_NONE = 0, - TK_SPACE, - TK_TAB, - TK_CR, - TK_LF, - TK_DOT, - TK_COMMA, - TK_PLUS, - TK_MINUS, - TK_EQ, - TK_COLON, - TK_ASTERISK, - TK_STRING, - TK_ALNUM, - TK_NUMDEC, - TK_NUMOCT, - TK_NUMHEX, - TK_PARENL, - TK_PARENR, - TK_COMMENT_ASTERISK, - TK_COMMENT_SEMICOLON, +#ifndef TRACE_LEX +#define TRACE_LEX 1 +#endif + +#define ERR 0 +#define OK 1 +#define CONTINUE 2 + +enum token_type { + TT_NONE = 0, + TT_SPACE, + TT_TAB, + TT_NEWLINE, + TT_DOT, + TT_COMMA, + TT_PLUS, + TT_MINUS, + TT_EQ, + TT_COLON, + TT_PERCENT, + TT_HASH, + TT_ASTERISK, + TT_STRING, + TT_IDENTIFIER, + TT_NUMDEC, + TT_NUMOCT, + TT_NUMHEX, + TT_LPAREN, + TT_RPAREN, + TT_COMMENT_ASTERISK, + TT_COMMENT_SEMICOLON, }; -struct tok { - enum tok_kind kind; +struct token { + enum token_type type; size_t offset; size_t length; }; enum lex_error { LE_NONE = 0, + LE_SOME, +}; + +enum lex_state { + LS_FREE = 0, + LS_CR, + LS_SPACE, + LS_TAB, + LS_IDENTIFIER, + LS_NUMOCTHEX, + LS_NUMOCT, + LS_NUMHEX, + LS_NUMDEC, + LS_STRING, + LS_STRING_ESC, + LS_COMMENT_ASTERISK, + LS_COMMENT_SEMICOLON, + LS_ERROR, + LS_EOF, +}; + +struct line_pos_info { + unsigned long line_num; + unsigned long column_num; + unsigned long line_offset; }; struct lex { + // State variables + enum lex_state state; + enum lex_error error; + size_t cursor; + size_t tok_offset; + size_t tokens_count; + bool inside_line; // Input data buffer - FILE *parsed_input_stream; - char *parsed_input; - size_t parsed_input_size; + FILE *input_stream; + char *input; + size_t input_size; // Tokens table FILE *tokbuf_stream; - struct tok *tokbuf; + struct token *tokbuf; size_t tokbuf_size; - // State - size_t offset; - enum lex_error error; }; enum stmt_kind { @@ -131,7 +166,7 @@ struct def_endef { }; struct stmt { - enum stmt_kind kind; + enum stmt_kind type; union { struct instruction instruction; int32_t align; @@ -149,8 +184,17 @@ struct symbol { uint32_t hash; }; +enum pars_error { + PE_NONE = 0, + PE_LEX, + PE_SOME, +}; + struct pars { - struct lex *lex; + const struct lex *lex; + // State + size_t cur_tok_id; + enum pars_error error; // Statement table FILE *stmttab_stream; struct stmt *stmttab; @@ -166,49 +210,567 @@ struct pars { }; struct assem { - struct pars *pars; + const struct pars *pars; }; +const char *const g_escape_table[256] = { + "\\x00", "\\x01", "\\x02", "\\x03", "\\x04", "\\x05", "\\x06", "\\x07", + "\\x08", "\\t", "\\n", "\\x0b", "\\x0c", "\\r", "\\x0e", "\\x0f", "\\x10", + "\\x11", "\\x12", "\\x13", "\\x14", "\\x15", "\\x16", "\\x17", "\\x18", + "\\x19", "\\x1a", "\\x1b", "\\x1c", "\\x1d", "\\x1e", "\\x1f", " ", "!", + "\\\"", "#", "$", "%", "&", "'", "(", ")", "*", "+", ",", "-", ".", "/", "0", + "1", "2", "3", "4", "5", "6", "7", "8", "9", ":", ";", "\\<", "=", "\\>", "?", + "@", "A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", + "O", "P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z", "[", "\\\\", + "]", "^", "_", "`", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", + "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", + "{", "|", "}", "~", "\\x7f", "\\x80", "\\x81", "\\x82", "\\x83", "\\x84", + "\\x85", "\\x86", "\\x87", "\\x88", "\\x89", "\\x8a", "\\x8b", "\\x8c", + "\\x8d", "\\x8e", "\\x8f", "\\x90", "\\x91", "\\x92", "\\x93", "\\x94", + "\\x95", "\\x96", "\\x97", "\\x98", "\\x99", "\\x9a", "\\x9b", "\\x9c", + "\\x9d", "\\x9e", "\\x9f", "\\xa0", "\\xa1", "\\xa2", "\\xa3", "\\xa4", + "\\xa5", "\\xa6", "\\xa7", "\\xa8", "\\xa9", "\\xaa", "\\xab", "\\xac", + "\\xad", "\\xae", "\\xaf", "\\xb0", "\\xb1", "\\xb2", "\\xb3", "\\xb4", + "\\xb5", "\\xb6", "\\xb7", "\\xb8", "\\xb9", "\\xba", "\\xbb", "\\xbc", + "\\xbd", "\\xbe", "\\xbf", "\\xc0", "\\xc1", "\\xc2", "\\xc3", "\\xc4", + "\\xc5", "\\xc6", "\\xc7", "\\xc8", "\\xc9", "\\xca", "\\xcb", "\\xcc", + "\\xcd", "\\xce", "\\xcf", "\\xd0", "\\xd1", "\\xd2", "\\xd3", "\\xd4", + "\\xd5", "\\xd6", "\\xd7", "\\xd8", "\\xd9", "\\xda", "\\xdb", "\\xdc", + "\\xdd", "\\xde", "\\xdf", "\\xe0", "\\xe1", "\\xe2", "\\xe3", "\\xe4", + "\\xe5", "\\xe6", "\\xe7", "\\xe8", "\\xe9", "\\xea", "\\xeb", "\\xec", + "\\xed", "\\xee", "\\xef", "\\xf0", "\\xf1", "\\xf2", "\\xf3", "\\xf4", + "\\xf5", "\\xf6", "\\xf7", "\\xf8", "\\xf9", "\\xfa", "\\xfb", "\\xfc", + "\\xfd", "\\xfe", +}; + +static bool should_be_escaped(const int c) +{ + return c < ' ' || c == '"' || c == '\\' || c == '<' || c == '>' || c > '~'; +} + +static bool is_oct(const int c) +{ + return c >= '0' && c <= '7'; +} + +static bool is_dec(const int c) +{ + return c >= '0' && c <= '9'; +} + +static bool is_hex(const int c) +{ + return is_dec(c) || (c >= 'A' && c <= 'F') || (c >= 'a' && c <= 'f'); +} + +static bool is_alphabetic(const int c) +{ + return (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z'); +} + +static bool is_alphanum(const int c) +{ + return is_dec(c) || is_alphabetic(c); +} + +static int printed_size(const char c) +{ + if (c < ' ' || c > '~') { + return sizeof("\\x00")-1; + } + if (c == '"' || c == '\\') { + return sizeof("\\\\")-1; + } + return 1; +} + +static int fprint_string_escaped( + const char *const str, const size_t length, FILE *const stream) +{ + int written = 0; + for (size_t i = 0; i < length; i++, written += printed_size(str[i])) { + if (should_be_escaped(str[i])) { + fputs(g_escape_table[(unsigned char)str[i]], stream); + } else { + fputc(str[i], stream); + } + } + return written; +} + +static const char *tok_kind_to_string(const enum token_type type) +{ + switch (type) { + case TT_NONE: return "NONE"; + case TT_SPACE: return "SPACE"; + case TT_TAB: return "TAB"; + case TT_NEWLINE: return "NEWLINE"; + case TT_DOT: return "DOT"; + case TT_COMMA: return "COMMA"; + case TT_PLUS: return "PLUS"; + case TT_MINUS: return "MINUS"; + case TT_EQ: return "EQ"; + case TT_COLON: return "COLON"; + case TT_PERCENT: return "PERCENT"; + case TT_HASH: return "HASH"; + case TT_STRING: return "STRING"; + case TT_ASTERISK: return "ASTERISK"; + case TT_IDENTIFIER: return "IDENTIFIER"; + case TT_NUMDEC: return "NUMDEC"; + case TT_NUMOCT: return "NUMOCT"; + case TT_NUMHEX: return "NUMHEX"; + case TT_LPAREN: return "PARENL"; + case TT_RPAREN: return "PARENR"; + case TT_COMMENT_ASTERISK: return "COMMENT"; + case TT_COMMENT_SEMICOLON: return "COMMENT"; + } + assert(0); + return "UNKNOWN"; +} + +static int fprint_tok(const char *const input, struct token *token, FILE *const stream) +{ + int res = fprintf(stream, "%s<", tok_kind_to_string(token->type)); + if (res == -1) { + return -1; + } + int written = res; + res = fprint_string_escaped(input + token->offset, token->length, stream); + if (res == -1) { + return -1; + } + written += res; + res = fputs(">\n", stream); + if (res == -1) { + return -1; + } + written += res; + return written; +} + static int lex_init(struct lex *const self) { - (void) self; + *self = (struct lex){ + .input_stream = open_memstream(&self->input, &self->input_size), + .tokbuf_stream = open_memstream( + (char **)&self->tokbuf, &self->tokbuf_size), + }; + assert(self->input_stream != NULL); + assert(self->tokbuf_stream != NULL); return OK; } -static int lex_next(struct lex *const self) +static int fwrite_token(const struct token *const token, FILE *const stream) { - (void) self; + const int res = fwrite(token, sizeof *token, 1, stream); + assert(res == 1); + return res; +} + +static void lex_yield_token(struct lex *const self, const struct token *const token) +{ + self->inside_line = token->type != TT_NEWLINE; + fwrite_token(token, self->tokbuf_stream); + self->tokens_count++; +} + +static const char *lex_state_error_string( + const enum lex_state state, const bool inside_line) +{ + if (!inside_line) { + assert(state == LS_FREE); + return "'*', ';', '0', '[1-9]', '[a-zA-Z_]', ',', '.', '(', ')', '+', " + "'-', '=', ':', '%', '#', ' ', '\\t', '\\r', '\\n', '\\r\\n' " + "or EOF"; + } + switch (state) { + case LS_FREE: + return "';', '0', '[1-9]', '[a-zA-Z_]', ',', '.', '(', ')', '+', " + "'-', '=', ':', '%', '#', ' ', '\\t', '\\r', '\\n', '\\r\\n' " + "or EOF"; + case LS_NUMOCTHEX: + return "';', '[0-7]', [xX], ',', '.', '(', ')', '+', " + "'-', '=', ':', '%', '#', ' ', '\\t', '\\r', '\\n', '\\r\\n' " + "or EOF"; + case LS_NUMOCT: + return "';', '[0-7]' , ',', '.', '(', ')', '+', " + "'-', '=', ':', '%', '#', ' ', '\\t', '\\r', '\\n', '\\r\\n' " + "or EOF"; + case LS_NUMHEX: + return "';', '[0-9a-zA-Z]' , ',', '.', '(', ')', '+', " + "'-', '=', ':', '%', '#', ' ', '\\t', '\\r', '\\n', '\\r\\n' " + "or EOF"; + case LS_NUMDEC: + return "';', '[0-9]' , ',', '.', '(', ')', '+', " + "'-', '=', ':', '%', '#', ' ', '\\t', '\\r', '\\n', '\\r\\n' " + "or EOF"; + case LS_CR: + case LS_SPACE: + case LS_TAB: + case LS_IDENTIFIER: + case LS_STRING: + case LS_STRING_ESC: + case LS_COMMENT_ASTERISK: + case LS_COMMENT_SEMICOLON: + case LS_ERROR: + case LS_EOF: + assert(0); + break; + } + return "???"; +} + +static struct line_pos_info lex_get_line_pos_info(const struct lex *const self) +{ + struct line_pos_info l = {0, 0, 0}; + bool cr = false; + // `input` is null terminated, that's why we subtract 1 here + for (size_t i = 0; i < self->input_size - 1; i++) { + const char c = self->input[i]; + if (c == '\r') { + cr = true; + l.line_offset = i + 1; + l.line_num++; + l.column_num = 0; + } else if (c == '\n') { + cr = false; + l.line_offset = i + 1; + if (!cr) { + l.line_num++; + } + l.column_num = 0; + } else { + cr = false; + l.column_num++; + } + } + return l; +} + +static int lex_yield_error(struct lex *const self, const int c) +{ + fflush(self->input_stream); + const struct line_pos_info l = lex_get_line_pos_info(self); + { + // Read out the rest of the line + int c; + do { + c = getc(stdin); + const char c_char = (c == EOF) ? 0 : c; + fwrite(&c_char, sizeof c_char, 1, self->input_stream); + } while (c != EOF && c != '\n' && c != '\r'); + fflush(self->input_stream); + } + fprintf( + stderr, + "<stdin>:%lu:%lu: lexing error: expected %s, found '%c'\n", + l.line_num + 1, + l.column_num + 1, + lex_state_error_string(self->state, self->inside_line), + c); + fprintf(stderr, "%5lu | %s\n", l.line_num + 1, self->input + l.line_offset); + fputs(" | ", stderr); + for (size_t i = 0; i < l.column_num; i++) { + if (self->input[l.line_offset + i] == '\t') { + fputc('\t', stderr); + } else { + fputc(' ', stderr); + } + } + fputs("^\n", stderr); + self->state = LS_ERROR; + return ERR; +} + +static int lex_handle_next(struct lex *const self, const int c) +{ + switch (self->state) { + case LS_FREE: + if (is_alphabetic(c) || c == '_') { + self->inside_line = false; + self->tok_offset = self->cursor; + self->state = LS_IDENTIFIER; + } else if (c == '0') { + self->tok_offset = self->cursor; + self->state = LS_NUMOCTHEX; + } else if (is_dec(c)) { + self->tok_offset = self->cursor; + self->state = LS_NUMDEC; + } else if (c == '"') { + self->tok_offset = self->cursor; + self->state = LS_STRING; + } else if (c == ';') { + self->tok_offset = self->cursor; + self->state = LS_COMMENT_SEMICOLON; + } else if (c == '*') { + if (self->inside_line) { + return lex_yield_error(self, c); + } + self->tok_offset = self->cursor; + self->state = LS_COMMENT_ASTERISK; + } else if (c == ',') { + lex_yield_token(self, &(struct token){TT_COMMA, self->cursor, 1}); + } else if (c == '.') { + lex_yield_token(self, &(struct token){TT_DOT, self->cursor, 1}); + } else if (c == '(') { + lex_yield_token(self, &(struct token){TT_LPAREN, self->cursor, 1}); + } else if (c == ')') { + lex_yield_token(self, &(struct token){TT_RPAREN, self->cursor, 1}); + } else if (c == '+') { + lex_yield_token(self, &(struct token){TT_PLUS, self->cursor, 1}); + } else if (c == '-') { + lex_yield_token(self, &(struct token){TT_MINUS, self->cursor, 1}); + } else if (c == '=') { + lex_yield_token(self, &(struct token){TT_EQ, self->cursor, 1}); + } else if (c == ':') { + lex_yield_token(self, &(struct token){TT_COLON, self->cursor, 1}); + } else if (c == '%') { + lex_yield_token(self, &(struct token){TT_PERCENT, self->cursor, 1}); + } else if (c == '#') { + lex_yield_token(self, &(struct token){TT_HASH, self->cursor, 1}); + } else if (c == '\r') { + self->tok_offset = self->cursor; + self->state = LS_CR; + } else if (c == '\n') { + lex_yield_token(self, &(struct token){TT_NEWLINE, self->cursor, 1}); + } else if (c == ' ') { + self->tok_offset = self->cursor; + self->state = LS_SPACE; + } else if (c == '\t') { + self->tok_offset = self->cursor; + self->state = LS_TAB; + } else if (c == EOF) { + self->state = LS_EOF; + } else { + return lex_yield_error(self, c); + } + break; + case LS_CR: // Accumulate CRLF into single token + { + const size_t size = c == '\n' ? 2 : 1; // 2 for CRLF, 1 for just CR + const struct token token = {TT_NEWLINE, self->tok_offset, size}; + lex_yield_token(self, &token); + self->state = LS_FREE; + if (c != '\n') { + // It is just CR, handle this char in LS_FREE state then + return lex_handle_next(self, c); + } + } + break; + case LS_SPACE: // Accumulate multiple spaces into single token + if (c != ' ') { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_SPACE, self->tok_offset, length}; + lex_yield_token(self, &token); + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_TAB: // Accumulate multiple tabs into single token + if (c != '\t') { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_TAB, self->tok_offset, length}; + lex_yield_token(self, &token); + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_IDENTIFIER: + if (!is_alphanum(c) && c != '_') { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_IDENTIFIER, self->tok_offset, length}; + lex_yield_token(self, &token); + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_NUMOCTHEX: + if (c == 'x' || c == 'X') { + self->state = LS_NUMHEX; + } else if (is_oct(c)) { + self->state = LS_NUMOCT; + } else if (is_alphabetic(c) || c == '_') { + return lex_yield_error(self, c); + } else { + assert((self->cursor - self->tok_offset) == 1); + const struct token token = {TT_NUMDEC, self->tok_offset, 1}; + lex_yield_token(self, &token); + // It was just zero, handle this char in LS_FREE state then + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_NUMOCT: + if (is_alphabetic(c) || c == '_') { + return lex_yield_error(self, c); + } else if (!is_oct(c)) { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_NUMOCT, self->tok_offset, length}; + lex_yield_token(self, &token); + // This token is finished, handle this char in LS_FREE state + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_NUMHEX: + if (is_hex(c)) { + // Keep calm + } else if (is_alphabetic(c) || c == '_') { + // Panik! + return lex_yield_error(self, c); + } else { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_NUMHEX, self->tok_offset, length}; + lex_yield_token(self, &token); + // This token is finished, handle this char in LS_FREE state + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_NUMDEC: + if (is_alphabetic(c) || c == '_') { + return lex_yield_error(self, c); + } else if (!is_dec(c)) { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_NUMDEC, self->tok_offset, length}; + lex_yield_token(self, &token); + // This token is finished, handle this char in LS_FREE state + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_STRING: + if (c == '\\') { + self->state = LS_STRING_ESC; + } else if (c == '"') { + const size_t length = self->cursor - self->tok_offset + 1; + const struct token token = {TT_STRING, self->tok_offset, length}; + lex_yield_token(self, &token); + // This token is finished + self->state = LS_FREE; + } + break; + case LS_STRING_ESC: + self->state = LS_STRING; + break; + case LS_COMMENT_ASTERISK: + if (c == '\r' || c == '\n') { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_COMMENT_ASTERISK, self->tok_offset, length}; + lex_yield_token(self, &token); + // This token is finished, handle this char in LS_FREE state + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_COMMENT_SEMICOLON: + if (c == '\r' || c == '\n') { + const size_t length = self->cursor - self->tok_offset; + const struct token token = {TT_COMMENT_SEMICOLON, self->tok_offset, length}; + lex_yield_token(self, &token); + // This token is finished, handle this char in LS_FREE state + self->state = LS_FREE; + return lex_handle_next(self, c); + } + break; + case LS_ERROR: + return ERR; + case LS_EOF: + assert(0); + } + return CONTINUE; +} + +/** Advance lexer to produce new token. + * \returns EOF if end of file reached. + * \returns ERR if error encountered and lexing cannot continue. + * \returns OK has one or more new tokens parsed. + */ +static int lex_next(struct lex *const self, FILE *const stream) +{ + for (;; self->cursor++) { + const int c = fgetc(stream); + const char c_char = (c == EOF) ? 0 : c; + fwrite(&c_char, sizeof c_char, 1, self->input_stream); + const int ret = lex_handle_next(self, c); + if (OK == ret) { + return OK; + } else if (ERR == ret) { + // TODO handle errors + return ERR; + } + if (c == EOF) { + break; + } + } + return EOF; +} + +/** Run lexer until the end of the input reached + * \returns OK if lexing finished successfully + * \returns ERR if error encountered and lexing cannot continue. + */ +static int lex_run(struct lex *const self, FILE *const stream) +{ + int res; + do { + res = lex_next(self, stream); + if (res == OK) { + res = 0; + } else if (res == ERR) { + return ERR; + } + } while (res != EOF); + fflush(self->input_stream); + fflush(self->tokbuf_stream); return OK; } static void lex_destroy(struct lex *const self) { - (void) self; + fclose(self->input_stream); + free(self->input); + fclose(self->tokbuf_stream); + free(self->tokbuf); } -static int pars_init(struct pars *const self, struct lex *const lex) +static int pars_init(struct pars *const self, const struct lex *const lex) { - (void) self; - (void) lex; + *self = (struct pars){ + .lex = lex, + .stmttab_stream = open_memstream( + (char **)&self->stmttab, &self->stmttab_size), + .symbuf_stream = open_memstream(&self->symbuf, &self->symbuf_size), + }; + assert(self->stmttab_stream != NULL); + assert(self->symbuf_stream != NULL); return OK; } -static int pars_run(struct pars *const self, FILE *const stream) +/** Run parser until the end of the input reached + * \returns OK if parsing finished successfully + * \returns ERR if error encountered and parsing cannot continue. + */ +static int pars_run(struct pars *const self) { (void) self; - (void) stream; + // TODO return OK; } static void pars_destroy(struct pars *const self) { - (void) self; + fclose(self->stmttab_stream); + free(self->stmttab); + fclose(self->symbuf_stream); + free(self->symbuf); } -static int assem_init(struct assem *const self, struct pars *const pars) +static int assem_init(struct assem *const self, const struct pars *const pars) { - (void) self; - (void) pars; + *self = (struct assem){ + .pars = pars, + }; return OK; } @@ -220,8 +782,12 @@ static int assem_resolve(struct assem *const self) static int assem_emit(struct assem *const self, FILE *const stream) { - (void) self; - (void) stream; + if (TRACE_LEX) { + const struct lex *const lex = self->pars->lex; + for (size_t i = 1; i < lex->tokbuf_size / (sizeof *lex->tokbuf); i++) { + fprint_tok(lex->input, &lex->tokbuf[i], stream); + } + } return OK; } @@ -240,19 +806,26 @@ int main(const int argc, char *const argv[]) if (OK != lex_init(&lex)) { return EXIT_FAILURE; } - // Chain lexer and parser + // Tokenize assembly program text + if (OK != lex_run(&lex, stdin)) { + lex_destroy(&lex); + return EXIT_FAILURE; + } + // Parser needs final lexer state to access parsed tokens and input data if (OK != pars_init(&pars, &lex)) { lex_destroy(&lex); return EXIT_FAILURE; } - // Parse assembly program text (tokenize + parse) - if (OK != pars_run(&pars, stdin)) { + // Parse assembly program text + if (OK != pars_run(&pars)) { pars_destroy(&pars); lex_destroy(&lex); return EXIT_FAILURE; } struct assem assem; - // Allocate adn populate code table and metadata table from parsed data + // Allocate and populate code table and metadata table from parsed data. + // Assembler needs parser's and lexer's final state to access parsed + // structure, tokens and input. if (OK != assem_init(&assem, &pars)) { pars_destroy(&pars); lex_destroy(&lex); |