wireshark/epan/protobuf_lang_scanner.l
João Valverde 79c3a77752 Add macros to control lemon diagnostics
Rename flex macros using parenthesis (mostly a style issue):

DIAG_OFF_FLEX -> DIAG_OFF_FLEX()
DIAG_ON_FLEX  -> DIAG_ON_FLEX()

Use the same kind of construct with lemon generated code using
DIAG_OFF_LEMON() and DIAG_ON_LEMON(). Use %include and %code
directives to enforce the desired order with generated code
in the middle in between pragmas.

Fix a clang-specific pragma to use DIAG_OFF_CLANG().

DIAG_OFF(unreachable-code) -> DIAG_OFF_CLANG(unreachable-code).

Apparently GCC is ignoring the -Wunreachable flag, that's why
it did not trigger an unknown pragma warning. From [1}:

  The -Wunreachable-code has been removed, because it was unstable: it
  relied on the optimizer, and so different versions of gcc would warn
  about different code.  The compiler still accepts and ignores the
  command line option so that existing Makefiles are not broken.  In some
  future release the option will be removed entirely. - Ian

[1] https://gcc.gnu.org/legacy-ml/gcc-help/2011-05/msg00360.html
2022-11-20 10:11:27 +00:00

195 lines
7.4 KiB
Text

%top {
/* Include this before everything else, for various large-file definitions */
#include "config.h"
#include <wireshark.h>
}
/*
* We want a reentrant scanner.
*/
%option reentrant
/*
* We don't read interactively from the terminal.
*/
%option never-interactive
/*
* We want to stop processing when we get to the end of the input.
*/
%option noyywrap
/*
* The type for the state we keep for the scanner (and parser).
*/
%option extra-type="protobuf_lang_state_t *"
/*
* Prefix scanner routines with "protobuf_lang_" rather than "yy", so this scanner
* can coexist with other scanners.
*/
%option prefix="protobuf_lang_"
/*
* We have to override the memory allocators so that we don't get
* "unused argument" warnings from the yyscanner argument (which
* we don't use, as we have a global memory allocator).
*
* We provide, as macros, our own versions of the routines generated by Flex,
* which just call malloc()/realloc()/free() (as the Flex versions do),
* discarding the extra argument.
*/
%option noyyalloc
%option noyyrealloc
%option noyyfree
%option yylineno
%option noinput
%option nounput
%{
/* protobuf_lang_scanner.l
*
* C Protocol Buffers Language Lexer (for *.proto files)
* Copyright 2020, Huang Qiangxiong <qiangxiong.huang@qq.com>
*
* SPDX-License-Identifier: GPL-2.0-or-later
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "protobuf_lang_tree.h"
#include "protobuf_lang_parser.h"
/*
* Disable diagnostics in the code generated by Flex.
*/
DIAG_OFF_FLEX()
/*
* Sleazy hack to suppress compiler warnings in yy_fatal_error().
*/
#define YY_EXIT_FAILURE ((void)yyscanner, 2)
/*
* Macros for the allocators, to discard the extra argument.
*/
#define protobuf_lang_alloc(size, yyscanner) (void *)malloc(size)
#define protobuf_lang_realloc(ptr, size, yyscanner) (void *)realloc((char *)(ptr), (size))
#define protobuf_lang_free(ptr, yyscanner) free((char *)ptr)
int old_status;
/* Extended error handling function defined in protobuf_lang_grammar.lemon */
void pbl_parser_error(protobuf_lang_state_t *state, const char *fmt, ...);
/* duplicate the text and keep the pointer in parser state for freeing later automatically */
static gchar*
strdup_and_store(void* yyscanner, const char* text);
#define PROTOBUF_LANG_PARSE(token_type) \
protobuf_lang_get_extra(yyscanner)->tmp_token = g_new0(protobuf_lang_token_t, 1); \
pbl_store_struct_token(protobuf_lang_get_extra(yyscanner), protobuf_lang_get_extra(yyscanner)->tmp_token); \
protobuf_lang_get_extra(yyscanner)->tmp_token->v = strdup_and_store(yyscanner, yytext); \
protobuf_lang_get_extra(yyscanner)->tmp_token->ln = protobuf_lang_get_lineno(yyscanner); \
return (token_type);
%}
%x COMMENT
%%
/* operations or symbols (PT_ means PBL Token) */
"(" PROTOBUF_LANG_PARSE(PT_LPAREN);
")" PROTOBUF_LANG_PARSE(PT_RPAREN);
"[" PROTOBUF_LANG_PARSE(PT_LBRACKET);
"]" PROTOBUF_LANG_PARSE(PT_RBRACKET);
"{" PROTOBUF_LANG_PARSE(PT_LCURLY);
"}" PROTOBUF_LANG_PARSE(PT_RCURLY);
"==" PROTOBUF_LANG_PARSE(PT_EQUAL);
"!=" PROTOBUF_LANG_PARSE(PT_NOTEQUAL);
"<>" PROTOBUF_LANG_PARSE(PT_NOTEQUAL2);
">=" PROTOBUF_LANG_PARSE(PT_GEQUAL);
"<=" PROTOBUF_LANG_PARSE(PT_LEQUAL);
"+=" PROTOBUF_LANG_PARSE(PT_ASSIGN_PLUS);
"=" PROTOBUF_LANG_PARSE(PT_ASSIGN);
"+" PROTOBUF_LANG_PARSE(PT_PLUS);
"-" PROTOBUF_LANG_PARSE(PT_MINUS);
"*" PROTOBUF_LANG_PARSE(PT_MULTIPLY);
"/" PROTOBUF_LANG_PARSE(PT_DIV);
"||" PROTOBUF_LANG_PARSE(PT_LOGIC_OR);
"|" PROTOBUF_LANG_PARSE(PT_OR);
"&&" PROTOBUF_LANG_PARSE(PT_LOGIC_AND);
"&" PROTOBUF_LANG_PARSE(PT_AND);
"!" PROTOBUF_LANG_PARSE(PT_NOT);
"~" PROTOBUF_LANG_PARSE(PT_NEG);
"^" PROTOBUF_LANG_PARSE(PT_XOR);
"<<" PROTOBUF_LANG_PARSE(PT_SHL);
">>" PROTOBUF_LANG_PARSE(PT_SHR);
"%" PROTOBUF_LANG_PARSE(PT_PERCENT);
"$" PROTOBUF_LANG_PARSE(PT_DOLLAR);
"?" PROTOBUF_LANG_PARSE(PT_COND);
";" PROTOBUF_LANG_PARSE(PT_SEMICOLON);
"." PROTOBUF_LANG_PARSE(PT_DOT);
"," PROTOBUF_LANG_PARSE(PT_COMMA);
":" PROTOBUF_LANG_PARSE(PT_COLON);
"<" PROTOBUF_LANG_PARSE(PT_LESS);
">" PROTOBUF_LANG_PARSE(PT_GREATER);
/* key words */
syntax PROTOBUF_LANG_PARSE(PT_SYNTAX);
import PROTOBUF_LANG_PARSE(PT_IMPORT);
weak PROTOBUF_LANG_PARSE(PT_WEAK);
public PROTOBUF_LANG_PARSE(PT_PUBLIC);
package PROTOBUF_LANG_PARSE(PT_PACKAGE);
option PROTOBUF_LANG_PARSE(PT_OPTION);
required PROTOBUF_LANG_PARSE(PT_REQUIRED);
optional PROTOBUF_LANG_PARSE(PT_OPTIONAL);
repeated PROTOBUF_LANG_PARSE(PT_REPEATED);
oneof PROTOBUF_LANG_PARSE(PT_ONEOF);
map PROTOBUF_LANG_PARSE(PT_MAP);
reserved PROTOBUF_LANG_PARSE(PT_RESERVED);
enum PROTOBUF_LANG_PARSE(PT_ENUM);
group PROTOBUF_LANG_PARSE(PT_GROUP);
extend PROTOBUF_LANG_PARSE(PT_EXTEND);
extensions PROTOBUF_LANG_PARSE(PT_EXTENSIONS);
message PROTOBUF_LANG_PARSE(PT_MESSAGE);
service PROTOBUF_LANG_PARSE(PT_SERVICE);
rpc PROTOBUF_LANG_PARSE(PT_RPC);
stream PROTOBUF_LANG_PARSE(PT_STREAM);
returns PROTOBUF_LANG_PARSE(PT_RETURNS);
to PROTOBUF_LANG_PARSE(PT_TO);
/* intLit values */
0|[1-9][0-9]* PROTOBUF_LANG_PARSE(PT_DECIMALLIT);
0[0-7]* PROTOBUF_LANG_PARSE(PT_OCTALLIT);
0[xX][0-9a-fA-F]+ PROTOBUF_LANG_PARSE(PT_HEXLIT);
/* Using extended identifier because we care only about position */
[a-zA-Z0-9_.][a-zA-Z0-9_.+-]* PROTOBUF_LANG_PARSE(PT_IDENT);
\"(\'|\\.|\"\"|[^"\n"])*\" PROTOBUF_LANG_PARSE(PT_STRLIT);
\'(\"|\\.|\'\'|[^"\n"])*\' PROTOBUF_LANG_PARSE(PT_STRLIT);
/* comments */
"//"[^\r\n]*
"/*" { old_status = YY_START; BEGIN COMMENT; }
<COMMENT>"*/" { BEGIN old_status; }
<COMMENT>([^*]|\n)+|.
/* space & tab */
[ \t\r\n]
/* prevent flex jam */
. { pbl_parser_error(protobuf_lang_get_extra(yyscanner), "unexpected token in proto file!\n"); }
%%
static gchar*
strdup_and_store(void* yyscanner, const char* text) {
return pbl_store_string_token(protobuf_lang_get_extra(yyscanner), g_strdup(text));
}
/*
* Turn diagnostics back on, so we check the code that we've written.
*/
DIAG_ON_FLEX()