micropython/py/lexer.h

179 wiersze
5.7 KiB
C

/*
* This file is part of the Micro Python project, http://micropython.org/
*
* The MIT License (MIT)
*
* Copyright (c) 2013, 2014 Damien P. George
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*/
/* lexer.h -- simple tokeniser for Micro Python
*
* Uses (byte) length instead of null termination.
* Tokens are the same - UTF-8 with (byte) length.
*/
typedef enum _mp_token_kind_t {
MP_TOKEN_END, // 0
MP_TOKEN_INVALID,
MP_TOKEN_DEDENT_MISMATCH,
MP_TOKEN_LONELY_STRING_OPEN,
MP_TOKEN_BAD_LINE_CONTINUATION,
MP_TOKEN_NEWLINE, // 5
MP_TOKEN_INDENT, // 6
MP_TOKEN_DEDENT, // 7
MP_TOKEN_NAME, // 8
MP_TOKEN_NUMBER,
MP_TOKEN_STRING,
MP_TOKEN_BYTES,
MP_TOKEN_ELLIPSIS,
MP_TOKEN_KW_FALSE, // 13
MP_TOKEN_KW_NONE,
MP_TOKEN_KW_TRUE,
MP_TOKEN_KW_AND,
MP_TOKEN_KW_AS,
MP_TOKEN_KW_ASSERT,
MP_TOKEN_KW_BREAK,
MP_TOKEN_KW_CLASS,
MP_TOKEN_KW_CONTINUE,
MP_TOKEN_KW_DEF, // 22
MP_TOKEN_KW_DEL,
MP_TOKEN_KW_ELIF,
MP_TOKEN_KW_ELSE,
MP_TOKEN_KW_EXCEPT,
MP_TOKEN_KW_FINALLY,
MP_TOKEN_KW_FOR,
MP_TOKEN_KW_FROM,
MP_TOKEN_KW_GLOBAL,
MP_TOKEN_KW_IF,
MP_TOKEN_KW_IMPORT, // 32
MP_TOKEN_KW_IN,
MP_TOKEN_KW_IS,
MP_TOKEN_KW_LAMBDA,
MP_TOKEN_KW_NONLOCAL,
MP_TOKEN_KW_NOT,
MP_TOKEN_KW_OR,
MP_TOKEN_KW_PASS,
MP_TOKEN_KW_RAISE,
MP_TOKEN_KW_RETURN,
MP_TOKEN_KW_TRY, // 42
MP_TOKEN_KW_WHILE,
MP_TOKEN_KW_WITH,
MP_TOKEN_KW_YIELD,
MP_TOKEN_OP_PLUS, // 46
MP_TOKEN_OP_MINUS,
MP_TOKEN_OP_STAR,
MP_TOKEN_OP_DBL_STAR,
MP_TOKEN_OP_SLASH,
MP_TOKEN_OP_DBL_SLASH,
MP_TOKEN_OP_PERCENT,
MP_TOKEN_OP_LESS,
MP_TOKEN_OP_DBL_LESS,
MP_TOKEN_OP_MORE,
MP_TOKEN_OP_DBL_MORE, // 56
MP_TOKEN_OP_AMPERSAND,
MP_TOKEN_OP_PIPE,
MP_TOKEN_OP_CARET,
MP_TOKEN_OP_TILDE,
MP_TOKEN_OP_LESS_EQUAL,
MP_TOKEN_OP_MORE_EQUAL,
MP_TOKEN_OP_DBL_EQUAL,
MP_TOKEN_OP_NOT_EQUAL,
MP_TOKEN_DEL_PAREN_OPEN, // 65
MP_TOKEN_DEL_PAREN_CLOSE,
MP_TOKEN_DEL_BRACKET_OPEN,
MP_TOKEN_DEL_BRACKET_CLOSE,
MP_TOKEN_DEL_BRACE_OPEN,
MP_TOKEN_DEL_BRACE_CLOSE,
MP_TOKEN_DEL_COMMA,
MP_TOKEN_DEL_COLON,
MP_TOKEN_DEL_PERIOD,
MP_TOKEN_DEL_SEMICOLON,
MP_TOKEN_DEL_AT, // 75
MP_TOKEN_DEL_EQUAL,
MP_TOKEN_DEL_PLUS_EQUAL,
MP_TOKEN_DEL_MINUS_EQUAL,
MP_TOKEN_DEL_STAR_EQUAL,
MP_TOKEN_DEL_SLASH_EQUAL,
MP_TOKEN_DEL_DBL_SLASH_EQUAL,
MP_TOKEN_DEL_PERCENT_EQUAL,
MP_TOKEN_DEL_AMPERSAND_EQUAL,
MP_TOKEN_DEL_PIPE_EQUAL,
MP_TOKEN_DEL_CARET_EQUAL, // 85
MP_TOKEN_DEL_DBL_MORE_EQUAL,
MP_TOKEN_DEL_DBL_LESS_EQUAL,
MP_TOKEN_DEL_DBL_STAR_EQUAL,
MP_TOKEN_DEL_MINUS_MORE,
} mp_token_kind_t;
typedef struct _mp_token_t {
uint src_line; // source line
uint src_column; // source column
mp_token_kind_t kind; // kind of token
const char *str; // string of token (valid only while this token is current token)
uint len; // (byte) length of string of token
} mp_token_t;
// the next-char function must return the next character in the stream
// it must return MP_LEXER_CHAR_EOF if end of stream
// it can be called again after returning MP_LEXER_CHAR_EOF, and in that case must return MP_LEXER_CHAR_EOF
#define MP_LEXER_CHAR_EOF (-1)
typedef unichar (*mp_lexer_stream_next_char_t)(void*);
typedef void (*mp_lexer_stream_close_t)(void*);
typedef struct _mp_lexer_t mp_lexer_t;
void mp_token_show(const mp_token_t *tok);
mp_lexer_t *mp_lexer_new(qstr src_name, void *stream_data, mp_lexer_stream_next_char_t stream_next_char, mp_lexer_stream_close_t stream_close);
mp_lexer_t *mp_lexer_new_from_str_len(qstr src_name, const char *str, uint len, uint free_len);
void mp_lexer_free(mp_lexer_t *lex);
qstr mp_lexer_source_name(mp_lexer_t *lex);
void mp_lexer_to_next(mp_lexer_t *lex);
const mp_token_t *mp_lexer_cur(const mp_lexer_t *lex);
bool mp_lexer_is_kind(mp_lexer_t *lex, mp_token_kind_t kind);
bool mp_lexer_show_error_pythonic_prefix(mp_lexer_t *lex);
bool mp_lexer_show_error_pythonic(mp_lexer_t *lex, const char *msg);
/******************************************************************/
// platform specific import function; must be implemented for a specific port
// TODO tidy up, rename, or put elsewhere
//mp_lexer_t *mp_import_open_file(qstr mod_name);
typedef enum {
MP_IMPORT_STAT_NO_EXIST,
MP_IMPORT_STAT_DIR,
MP_IMPORT_STAT_FILE,
} mp_import_stat_t;
mp_import_stat_t mp_import_stat(const char *path);
mp_lexer_t *mp_lexer_new_from_file(const char *filename);