freebsd-nq/usr.bin/lex/scan.l

573 lines
12 KiB
Plaintext
Raw Normal View History

/* scan.l - scanner for flex input */
%{
/*-
* Copyright (c) 1990 The Regents of the University of California.
* All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Vern Paxson.
*
* The United States Government has rights in this work pursuant
* to contract no. DE-AC03-76SF00098 between the United States
* Department of Energy and the University of California.
*
* Redistribution and use in source and binary forms are permitted provided
* that: (1) source distributions retain this entire copyright notice and
* comment, and (2) distributions including binaries display the following
* acknowledgement: ``This product includes software developed by the
* University of California, Berkeley and its contributors'' in the
* documentation or other materials provided with the distribution and in
* all advertising materials mentioning features or use of this software.
* Neither the name of the University nor the names of its contributors may
* be used to endorse or promote products derived from this software without
* specific prior written permission.
* THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
* WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
*/
/* $Header: scan.l,v 1.2 94/01/04 14:33:09 vern Exp $ */
#include "flexdef.h"
#include "parse.h"
#define ACTION_ECHO add_action( yytext )
#define MARK_END_OF_PROLOG mark_prolog();
#define YY_DECL \
int flexscan()
#define RETURNCHAR \
yylval = (unsigned char) yytext[0]; \
return CHAR;
#define RETURNNAME \
strcpy( nmstr, yytext ); \
return NAME;
#define PUT_BACK_STRING(str, start) \
for ( i = strlen( str ) - 1; i >= start; --i ) \
unput((str)[i])
#define CHECK_REJECT(str) \
if ( all_upper( str ) ) \
reject = true;
#define CHECK_YYMORE(str) \
if ( all_lower( str ) ) \
yymore_used = true;
%}
%x SECT2 SECT2PROLOG SECT3 CODEBLOCK PICKUPDEF SC CARETISBOL NUM QUOTE
%x FIRSTCCL CCL ACTION RECOVER BRACEERROR C_COMMENT ACTION_COMMENT
%x ACTION_STRING PERCENT_BRACE_ACTION USED_LIST CODEBLOCK_2
WS [ \t]+
OPTWS [ \t]*
NOT_WS [^ \t\n]
NL (\n|\r\n|\n\r)
NAME ([a-z_][a-z_0-9-]*)
NOT_NAME [^a-z_*\n]+
SCNAME {NAME}
ESCSEQ (\\([^\n]|[0-9]{1,3}|x[0-9a-f]{1,2}))
FIRST_CCL_CHAR ([^\\\n]|{ESCSEQ})
CCL_CHAR ([^\\\n\]]|{ESCSEQ})
%%
static int bracelevel, didadef, indented_code, checking_used;
int doing_codeblock = false;
int i;
Char nmdef[MAXLINE], myesc();
^{WS} indented_code = true; BEGIN(CODEBLOCK);
^"/*" ACTION_ECHO; BEGIN(C_COMMENT);
^"%s"{NAME}? return SCDECL;
^"%x"{NAME}? return XSCDECL;
^"%{".*{NL} {
++linenum;
line_directive_out( (FILE *) 0 );
indented_code = false;
BEGIN(CODEBLOCK);
}
{WS} return WHITESPACE;
^"%%".* {
sectnum = 2;
bracelevel = 0;
mark_defs1();
line_directive_out( (FILE *) 0 );
BEGIN(SECT2PROLOG);
return SECTEND;
}
^"%pointer".*{NL} {
if ( lex_compat )
warn( "%pointer incompatible with -l option" );
else
yytext_is_array = false;
++linenum;
}
^"%array".*{NL} {
if ( C_plus_plus )
warn( "%array incompatible with -+ option" );
else
yytext_is_array = true;
++linenum;
}
^"%used" {
warn( "%used/%unused have been deprecated" );
checking_used = REALLY_USED; BEGIN(USED_LIST);
}
^"%unused" {
warn( "%used/%unused have been deprecated" );
checking_used = REALLY_NOT_USED; BEGIN(USED_LIST);
}
^"%"[aceknopr]{OPTWS}[0-9]*{OPTWS}{NL} ++linenum; /* ignore */
^"%"[^sxanpekotcru{}].* synerr( "unrecognized '%' directive" );
^{NAME} {
strcpy( nmstr, yytext );
didadef = false;
BEGIN(PICKUPDEF);
}
{SCNAME} RETURNNAME;
^{OPTWS}{NL} ++linenum; /* allows blank lines in section 1 */
{OPTWS}{NL} ++linenum; return '\n';
<C_COMMENT>"*/" ACTION_ECHO; BEGIN(INITIAL);
<C_COMMENT>"*/".*{NL} ++linenum; ACTION_ECHO; BEGIN(INITIAL);
<C_COMMENT>[^*\n]+ ACTION_ECHO;
<C_COMMENT>"*" ACTION_ECHO;
<C_COMMENT>{NL} ++linenum; ACTION_ECHO;
<CODEBLOCK>^"%}".*{NL} ++linenum; BEGIN(INITIAL);
<CODEBLOCK>"reject" ACTION_ECHO; CHECK_REJECT(yytext);
<CODEBLOCK>"yymore" ACTION_ECHO; CHECK_YYMORE(yytext);
<CODEBLOCK>{NAME}|{NOT_NAME}|. ACTION_ECHO;
<CODEBLOCK>{NL} {
++linenum;
ACTION_ECHO;
if ( indented_code )
BEGIN(INITIAL);
}
<PICKUPDEF>{WS} /* separates name and definition */
<PICKUPDEF>{NOT_WS}.* {
strcpy( (char *) nmdef, yytext );
/* Skip trailing whitespace. */
for ( i = strlen( (char *) nmdef ) - 1;
i >= 0 && (nmdef[i] == ' ' || nmdef[i] == '\t');
--i )
;
nmdef[i + 1] = '\0';
ndinstal( nmstr, nmdef );
didadef = true;
}
<PICKUPDEF>{NL} {
if ( ! didadef )
synerr( "incomplete name definition" );
BEGIN(INITIAL);
++linenum;
}
<RECOVER>.*{NL} ++linenum; BEGIN(INITIAL); RETURNNAME;
<USED_LIST>{NL} ++linenum; BEGIN(INITIAL);
<USED_LIST>{WS}
<USED_LIST>"reject" {
if ( all_upper( yytext ) )
reject_really_used = checking_used;
else
synerr(
"unrecognized %used/%unused construct" );
}
<USED_LIST>"yymore" {
if ( all_lower( yytext ) )
yymore_really_used = checking_used;
else
synerr(
"unrecognized %used/%unused construct" );
}
<USED_LIST>{NOT_WS}+ synerr( "unrecognized %used/%unused construct" );
<SECT2PROLOG>^"%{".* ++bracelevel; yyless( 2 ); /* eat only %{ */
<SECT2PROLOG>^"%}".* --bracelevel; yyless( 2 ); /* eat only %} */
<SECT2PROLOG>^{WS}.* ACTION_ECHO; /* indented code in prolog */
<SECT2PROLOG>^{NOT_WS}.* { /* non-indented code */
if ( bracelevel <= 0 )
{ /* not in %{ ... %} */
yyless( 0 ); /* put it all back */
mark_prolog();
BEGIN(SECT2);
}
else
ACTION_ECHO;
}
<SECT2PROLOG>.* ACTION_ECHO;
<SECT2PROLOG>{NL} ++linenum; ACTION_ECHO;
<SECT2PROLOG><<EOF>> {
mark_prolog();
sectnum = 0;
yyterminate(); /* to stop the parser */
}
<SECT2>^{OPTWS}{NL} ++linenum; /* allow blank lines in section 2 */
<SECT2>^({WS}|"%{") {
indented_code = (yytext[0] != '%');
doing_codeblock = true;
bracelevel = 1;
if ( indented_code )
ACTION_ECHO;
BEGIN(CODEBLOCK_2);
}
<SECT2>^"<" BEGIN(SC); return '<';
<SECT2>^"^" return '^';
<SECT2>\" BEGIN(QUOTE); return '"';
<SECT2>"{"/[0-9] BEGIN(NUM); return '{';
<SECT2>"{"[^0-9\n][^}\n]* BEGIN(BRACEERROR);
<SECT2>"$"/([ \t]|{NL}) return '$';
<SECT2>{WS}"%{" {
bracelevel = 1;
BEGIN(PERCENT_BRACE_ACTION);
return '\n';
}
<SECT2>{WS}"|".*{NL} continued_action = true; ++linenum; return '\n';
<SECT2>{WS} {
/* This rule is separate from the one below because
* otherwise we get variable trailing context, so
* we can't build the scanner using -{f,F}.
*/
bracelevel = 0;
continued_action = false;
BEGIN(ACTION);
return '\n';
}
<SECT2>{OPTWS}{NL} {
bracelevel = 0;
continued_action = false;
BEGIN(ACTION);
unput( '\n' ); /* so <ACTION> sees it */
return '\n';
}
<SECT2>"<<EOF>>" return EOF_OP;
<SECT2>^"%%".* {
sectnum = 3;
BEGIN(SECT3);
yyterminate(); /* to stop the parser */
}
<SECT2>"["{FIRST_CCL_CHAR}{CCL_CHAR}* {
int cclval;
strcpy( nmstr, yytext );
/* Check to see if we've already encountered this
* ccl.
*/
if ( (cclval = ccllookup( (Char *) nmstr )) )
{
if ( input() != ']' )
synerr( "bad character class" );
yylval = cclval;
++cclreuse;
return PREVCCL;
}
else
{
/* We fudge a bit. We know that this ccl will
* soon be numbered as lastccl + 1 by cclinit.
*/
cclinstal( (Char *) nmstr, lastccl + 1 );
/* Push back everything but the leading bracket
* so the ccl can be rescanned.
*/
yyless( 1 );
BEGIN(FIRSTCCL);
return '[';
}
}
<SECT2>"{"{NAME}"}" {
register Char *nmdefptr;
Char *ndlookup();
strcpy( nmstr, yytext + 1 );
nmstr[yyleng - 2] = '\0'; /* chop trailing brace */
if ( ! (nmdefptr = ndlookup( nmstr )) )
format_synerr( "undefined definition {%s}",
nmstr );
else
{ /* push back name surrounded by ()'s */
int len = strlen( (char *) nmdefptr );
if ( lex_compat || nmdefptr[0] == '^' ||
(len > 0 && nmdefptr[len - 1] == '$') )
{ /* don't use ()'s after all */
PUT_BACK_STRING((char *) nmdefptr, 0);
if ( nmdefptr[0] == '^' )
BEGIN(CARETISBOL);
}
else
{
unput(')');
PUT_BACK_STRING((char *) nmdefptr, 0);
unput('(');
}
}
}
<SECT2>[/|*+?.()] return (unsigned char) yytext[0];
<SECT2>. RETURNCHAR;
<SC>[,*] return (unsigned char) yytext[0];
<SC>">" BEGIN(SECT2); return '>';
<SC>">"/^ BEGIN(CARETISBOL); return '>';
<SC>{SCNAME} RETURNNAME;
<SC>. {
format_synerr( "bad <start condition>: %s", yytext );
}
<CARETISBOL>"^" BEGIN(SECT2); return '^';
<QUOTE>[^"\n] RETURNCHAR;
<QUOTE>\" BEGIN(SECT2); return '"';
<QUOTE>{NL} {
synerr( "missing quote" );
BEGIN(SECT2);
++linenum;
return '"';
}
<FIRSTCCL>"^"/[^-\]\n] BEGIN(CCL); return '^';
<FIRSTCCL>"^"/("-"|"]") return '^';
<FIRSTCCL>. BEGIN(CCL); RETURNCHAR;
<CCL>-/[^\]\n] return '-';
<CCL>[^\]\n] RETURNCHAR;
<CCL>"]" BEGIN(SECT2); return ']';
<CCL>.|{NL} {
synerr( "bad character class" );
BEGIN(SECT2);
return ']';
}
<NUM>[0-9]+ {
yylval = myctoi( yytext );
return NUMBER;
}
<NUM>"," return ',';
<NUM>"}" BEGIN(SECT2); return '}';
<NUM>. {
synerr( "bad character inside {}'s" );
BEGIN(SECT2);
return '}';
}
<NUM>{NL} {
synerr( "missing }" );
BEGIN(SECT2);
++linenum;
return '}';
}
<BRACEERROR>"}" synerr( "bad name in {}'s" ); BEGIN(SECT2);
<BRACEERROR>{NL} synerr( "missing }" ); ++linenum; BEGIN(SECT2);
<CODEBLOCK_2>"/*" ACTION_ECHO; BEGIN(ACTION_COMMENT);
<PERCENT_BRACE_ACTION,CODEBLOCK_2>{OPTWS}"%}".* bracelevel = 0;
<PERCENT_BRACE_ACTION,CODEBLOCK_2,ACTION>"reject" {
ACTION_ECHO;
CHECK_REJECT(yytext);
}
<PERCENT_BRACE_ACTION,CODEBLOCK_2,ACTION>"yymore" {
ACTION_ECHO;
CHECK_YYMORE(yytext);
}
<PERCENT_BRACE_ACTION,CODEBLOCK_2>{NAME}|{NOT_NAME}|. ACTION_ECHO;
<PERCENT_BRACE_ACTION,CODEBLOCK_2>{NL} {
++linenum;
ACTION_ECHO;
if ( bracelevel == 0 ||
(doing_codeblock && indented_code) )
{
if ( ! doing_codeblock )
add_action( "\tYY_BREAK\n" );
doing_codeblock = false;
BEGIN(SECT2);
}
}
/* Reject and YYmore() are checked for above, in PERCENT_BRACE_ACTION */
<ACTION>"{" ACTION_ECHO; ++bracelevel;
<ACTION>"}" ACTION_ECHO; --bracelevel;
<ACTION>[^a-z_{}"'/\n]+ ACTION_ECHO;
<ACTION>{NAME} ACTION_ECHO;
<ACTION>"/*" ACTION_ECHO; BEGIN(ACTION_COMMENT);
<ACTION>"'"([^'\\\n]|\\.)*"'" ACTION_ECHO; /* character constant */
<ACTION>\" ACTION_ECHO; BEGIN(ACTION_STRING);
<ACTION>{NL} {
++linenum;
ACTION_ECHO;
if ( bracelevel == 0 )
{
add_action( "\tYY_BREAK\n" );
BEGIN(SECT2);
}
}
<ACTION>. ACTION_ECHO;
<ACTION_COMMENT>"*/" {
ACTION_ECHO;
if ( doing_codeblock )
BEGIN(CODEBLOCK_2);
else
BEGIN(ACTION);
}
<ACTION_COMMENT>"*" ACTION_ECHO;
<ACTION_COMMENT>[^*\n]+ ACTION_ECHO;
<ACTION_COMMENT>[^*\n]*{NL} ++linenum; ACTION_ECHO;
<ACTION_STRING>[^"\\\n]+ ACTION_ECHO;
<ACTION_STRING>\\. ACTION_ECHO;
<ACTION_STRING>{NL} ++linenum; ACTION_ECHO;
<ACTION_STRING>\" ACTION_ECHO; BEGIN(ACTION);
<ACTION_STRING>. ACTION_ECHO;
<ACTION,ACTION_COMMENT,ACTION_STRING><<EOF>> {
synerr( "EOF encountered inside an action" );
yyterminate();
}
<SECT2,QUOTE,CCL>{ESCSEQ} {
yylval = myesc( (Char *) yytext );
return CHAR;
}
<FIRSTCCL>{ESCSEQ} {
yylval = myesc( (Char *) yytext );
BEGIN(CCL);
return CHAR;
}
<SECT3>.*(\n?) ECHO;
<SECT3><<EOF>> sectnum = 0; yyterminate();
<*>.|\n format_synerr( "bad character: %s", yytext );
%%
int yywrap()
{
if ( --num_input_files > 0 )
{
set_input_file( *++input_files );
return 0;
}
else
return 1;
}
/* set_input_file - open the given file (if NULL, stdin) for scanning */
void set_input_file( file )
char *file;
{
if ( file )
{
infilename = file;
yyin = fopen( infilename, "r" );
if ( yyin == NULL )
lerrsf( "can't open %s", file );
}
else
{
yyin = stdin;
infilename = "<stdin>";
}
}
/* Wrapper routines for accessing the scanner's malloc routines. */
void *flex_alloc( size )
unsigned int size;
{
return yy_flex_alloc( size );
}
void *flex_realloc( ptr, size )
void *ptr;
unsigned int size;
{
return yy_flex_realloc( ptr, size );
}
void flex_free( ptr )
void *ptr;
{
yy_flex_free( ptr );
}