summaryrefslogtreecommitdiff
path: root/usr.bin/lex/scan.l
diff options
context:
space:
mode:
authorTheo de Raadt <deraadt@cvs.openbsd.org>1995-10-18 08:53:40 +0000
committerTheo de Raadt <deraadt@cvs.openbsd.org>1995-10-18 08:53:40 +0000
commitd6583bb2a13f329cf0332ef2570eb8bb8fc0e39c (patch)
treeece253b876159b39c620e62b6c9b1174642e070e /usr.bin/lex/scan.l
initial import of NetBSD tree
Diffstat (limited to 'usr.bin/lex/scan.l')
-rw-r--r--usr.bin/lex/scan.l710
1 files changed, 710 insertions, 0 deletions
diff --git a/usr.bin/lex/scan.l b/usr.bin/lex/scan.l
new file mode 100644
index 00000000000..2aada16b478
--- /dev/null
+++ b/usr.bin/lex/scan.l
@@ -0,0 +1,710 @@
+/* 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: /cvs/OpenBSD/src/usr.bin/lex/scan.l,v 1.1 1995/10/18 08:45:32 deraadt Exp $ */
+
+#include "flexdef.h"
+#include "parse.h"
+
+#define ACTION_ECHO add_action( yytext )
+#define ACTION_IFDEF(def, should_define) \
+ { \
+ if ( should_define ) \
+ action_define( def, 1 ); \
+ }
+
+#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;
+%}
+
+%option caseless nodefault outfile="scan.c" stack noyy_top_state
+%option nostdinit
+
+%x SECT2 SECT2PROLOG SECT3 CODEBLOCK PICKUPDEF SC CARETISBOL NUM QUOTE
+%x FIRSTCCL CCL ACTION RECOVER COMMENT ACTION_STRING PERCENT_BRACE_ACTION
+%x OPTION LINEDIR
+
+WS [[:blank:]]+
+OPTWS [[:blank:]]*
+NOT_WS [^[:blank:]\n]
+
+NL \r?\n
+
+NAME ([[:alpha:]_][[:alnum:]_-]*)
+NOT_NAME [^[:alpha:]_*\n]+
+
+SCNAME {NAME}
+
+ESCSEQ (\\([^\n]|[0-7]{1,3}|x[[:xdigit:]]{1,2}))
+
+FIRST_CCL_CHAR ([^\\\n]|{ESCSEQ})
+CCL_CHAR ([^\\\n\]]|{ESCSEQ})
+CCL_EXPR ("[:"[[:alpha:]]+":]")
+
+LEXOPT [aceknopr]
+
+%%
+ static int bracelevel, didadef, indented_code;
+ static int doing_rule_action = false;
+ static int option_sense;
+
+ int doing_codeblock = false;
+ int i;
+ Char nmdef[MAXLINE], myesc();
+
+
+<INITIAL>{
+ ^{WS} indented_code = true; BEGIN(CODEBLOCK);
+ ^"/*" ACTION_ECHO; yy_push_state( COMMENT );
+ ^#{OPTWS}line{WS} yy_push_state( LINEDIR );
+ ^"%s"{NAME}? return SCDECL;
+ ^"%x"{NAME}? return XSCDECL;
+ ^"%{".*{NL} {
+ ++linenum;
+ line_directive_out( (FILE *) 0, 1 );
+ indented_code = false;
+ BEGIN(CODEBLOCK);
+ }
+
+ {WS} /* discard */
+
+ ^"%%".* {
+ sectnum = 2;
+ bracelevel = 0;
+ mark_defs1();
+ line_directive_out( (FILE *) 0, 1 );
+ BEGIN(SECT2PROLOG);
+ return SECTEND;
+ }
+
+ ^"%pointer".*{NL} yytext_is_array = false; ++linenum;
+ ^"%array".*{NL} yytext_is_array = true; ++linenum;
+
+ ^"%option" BEGIN(OPTION); return OPTION_OP;
+
+ ^"%"{LEXOPT}{OPTWS}[[:digit:]]*{OPTWS}{NL} ++linenum; /* ignore */
+ ^"%"{LEXOPT}{WS}.*{NL} ++linenum; /* ignore */
+
+ ^"%"[^sxaceknopr{}].* synerr( _( "unrecognized '%' directive" ) );
+
+ ^{NAME} {
+ strcpy( nmstr, yytext );
+ didadef = false;
+ BEGIN(PICKUPDEF);
+ }
+
+ {SCNAME} RETURNNAME;
+ ^{OPTWS}{NL} ++linenum; /* allows blank lines in section 1 */
+ {OPTWS}{NL} ACTION_ECHO; ++linenum; /* maybe end of comment line */
+}
+
+
+<COMMENT>{
+ "*/" ACTION_ECHO; yy_pop_state();
+ "*" ACTION_ECHO;
+ [^*\n]+ ACTION_ECHO;
+ [^*\n]*{NL} ++linenum; ACTION_ECHO;
+}
+
+<LINEDIR>{
+ \n yy_pop_state();
+ [[:digit:]]+ linenum = myctoi( yytext );
+
+ \"[^"\n]*\" {
+ flex_free( (void *) infilename );
+ infilename = copy_string( yytext + 1 );
+ infilename[strlen( infilename ) - 1] = '\0';
+ }
+ . /* ignore spurious characters */
+}
+
+<CODEBLOCK>{
+ ^"%}".*{NL} ++linenum; BEGIN(INITIAL);
+
+ {NAME}|{NOT_NAME}|. ACTION_ECHO;
+
+ {NL} {
+ ++linenum;
+ ACTION_ECHO;
+ if ( indented_code )
+ BEGIN(INITIAL);
+ }
+}
+
+
+<PICKUPDEF>{
+ {WS} /* separates name and definition */
+
+ {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;
+ }
+
+ {NL} {
+ if ( ! didadef )
+ synerr( _( "incomplete name definition" ) );
+ BEGIN(INITIAL);
+ ++linenum;
+ }
+}
+
+
+<OPTION>{
+ {NL} ++linenum; BEGIN(INITIAL);
+ {WS} option_sense = true;
+
+ "=" return '=';
+
+ no option_sense = ! option_sense;
+
+ 7bit csize = option_sense ? 128 : 256;
+ 8bit csize = option_sense ? 256 : 128;
+
+ align long_align = option_sense;
+ always-interactive {
+ action_define( "YY_ALWAYS_INTERACTIVE", option_sense );
+ }
+ array yytext_is_array = option_sense;
+ backup backing_up_report = option_sense;
+ batch interactive = ! option_sense;
+ "c++" C_plus_plus = option_sense;
+ caseful|case-sensitive caseins = ! option_sense;
+ caseless|case-insensitive caseins = option_sense;
+ debug ddebug = option_sense;
+ default spprdflt = ! option_sense;
+ ecs useecs = option_sense;
+ fast {
+ useecs = usemecs = false;
+ use_read = fullspd = true;
+ }
+ full {
+ useecs = usemecs = false;
+ use_read = fulltbl = true;
+ }
+ input ACTION_IFDEF("YY_NO_INPUT", ! option_sense);
+ interactive interactive = option_sense;
+ lex-compat lex_compat = option_sense;
+ main {
+ action_define( "YY_MAIN", option_sense );
+ do_yywrap = ! option_sense;
+ }
+ meta-ecs usemecs = option_sense;
+ never-interactive {
+ action_define( "YY_NEVER_INTERACTIVE", option_sense );
+ }
+ perf-report performance_report += option_sense ? 1 : -1;
+ pointer yytext_is_array = ! option_sense;
+ read use_read = option_sense;
+ reject reject_really_used = option_sense;
+ stack action_define( "YY_STACK_USED", option_sense );
+ stdinit do_stdinit = option_sense;
+ stdout use_stdout = option_sense;
+ unput ACTION_IFDEF("YY_NO_UNPUT", ! option_sense);
+ verbose printstats = option_sense;
+ warn nowarn = ! option_sense;
+ yylineno do_yylineno = option_sense;
+ yymore yymore_really_used = option_sense;
+ yywrap do_yywrap = option_sense;
+
+ yy_push_state ACTION_IFDEF("YY_NO_PUSH_STATE", ! option_sense);
+ yy_pop_state ACTION_IFDEF("YY_NO_POP_STATE", ! option_sense);
+ yy_top_state ACTION_IFDEF("YY_NO_TOP_STATE", ! option_sense);
+
+ yy_scan_buffer ACTION_IFDEF("YY_NO_SCAN_BUFFER", ! option_sense);
+ yy_scan_bytes ACTION_IFDEF("YY_NO_SCAN_BYTES", ! option_sense);
+ yy_scan_string ACTION_IFDEF("YY_NO_SCAN_STRING", ! option_sense);
+
+ outfile return OPT_OUTFILE;
+ prefix return OPT_PREFIX;
+ yyclass return OPT_YYCLASS;
+
+ \"[^"\n]*\" {
+ strcpy( nmstr, yytext + 1 );
+ nmstr[strlen( nmstr ) - 1] = '\0';
+ return NAME;
+ }
+
+ (([a-mo-z]|n[a-np-z])[[:alpha:]\-+]*)|. {
+ format_synerr( _( "unrecognized %%option: %s" ),
+ yytext );
+ BEGIN(RECOVER);
+ }
+}
+
+<RECOVER>.*{NL} ++linenum; BEGIN(INITIAL);
+
+
+<SECT2PROLOG>{
+ ^"%{".* ++bracelevel; yyless( 2 ); /* eat only %{ */
+ ^"%}".* --bracelevel; yyless( 2 ); /* eat only %} */
+
+ ^{WS}.* ACTION_ECHO; /* indented code in prolog */
+
+ ^{NOT_WS}.* { /* non-indented code */
+ if ( bracelevel <= 0 )
+ { /* not in %{ ... %} */
+ yyless( 0 ); /* put it all back */
+ yy_set_bol( 1 );
+ mark_prolog();
+ BEGIN(SECT2);
+ }
+ else
+ ACTION_ECHO;
+ }
+
+ .* ACTION_ECHO;
+ {NL} ++linenum; ACTION_ECHO;
+
+ <<EOF>> {
+ mark_prolog();
+ sectnum = 0;
+ yyterminate(); /* to stop the parser */
+ }
+}
+
+<SECT2>{
+ ^{OPTWS}{NL} ++linenum; /* allow blank lines in section 2 */
+
+ ^{OPTWS}"%{" {
+ indented_code = false;
+ doing_codeblock = true;
+ bracelevel = 1;
+ BEGIN(PERCENT_BRACE_ACTION);
+ }
+
+ ^{OPTWS}"<" BEGIN(SC); return '<';
+ ^{OPTWS}"^" return '^';
+ \" BEGIN(QUOTE); return '"';
+ "{"/[[:digit:]] BEGIN(NUM); return '{';
+ "$"/([[:blank:]]|{NL}) return '$';
+
+ {WS}"%{" {
+ bracelevel = 1;
+ BEGIN(PERCENT_BRACE_ACTION);
+
+ if ( in_rule )
+ {
+ doing_rule_action = true;
+ in_rule = false;
+ return '\n';
+ }
+ }
+ {WS}"|".*{NL} continued_action = true; ++linenum; return '\n';
+
+ ^{WS}"/*" {
+ yyless( yyleng - 2 ); /* put back '/', '*' */
+ bracelevel = 0;
+ continued_action = false;
+ BEGIN(ACTION);
+ }
+
+ ^{WS} /* allow indented rules */
+
+ {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);
+
+ if ( in_rule )
+ {
+ doing_rule_action = true;
+ in_rule = false;
+ return '\n';
+ }
+ }
+
+ {OPTWS}{NL} {
+ bracelevel = 0;
+ continued_action = false;
+ BEGIN(ACTION);
+ unput( '\n' ); /* so <ACTION> sees it */
+
+ if ( in_rule )
+ {
+ doing_rule_action = true;
+ in_rule = false;
+ return '\n';
+ }
+ }
+
+ ^{OPTWS}"<<EOF>>" |
+ "<<EOF>>" return EOF_OP;
+
+ ^"%%".* {
+ sectnum = 3;
+ BEGIN(SECT3);
+ yyterminate(); /* to stop the parser */
+ }
+
+ "["({FIRST_CCL_CHAR}|{CCL_EXPR})({CCL_CHAR}|{CCL_EXPR})* {
+ int cclval;
+
+ strcpy( nmstr, yytext );
+
+ /* Check to see if we've already encountered this
+ * ccl.
+ */
+ if ( (cclval = ccllookup( (Char *) nmstr )) != 0 )
+ {
+ 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 '[';
+ }
+ }
+
+ "{"{NAME}"}" {
+ register Char *nmdefptr;
+ Char *ndlookup();
+
+ strcpy( nmstr, yytext + 1 );
+ nmstr[yyleng - 2] = '\0'; /* chop trailing brace */
+
+ if ( (nmdefptr = ndlookup( nmstr )) == 0 )
+ 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('(');
+ }
+ }
+ }
+
+ [/|*+?.(){}] return (unsigned char) yytext[0];
+ . RETURNCHAR;
+}
+
+
+<SC>{
+ [,*] return (unsigned char) yytext[0];
+ ">" BEGIN(SECT2); return '>';
+ ">"/^ BEGIN(CARETISBOL); return '>';
+ {SCNAME} RETURNNAME;
+ . {
+ format_synerr( _( "bad <start condition>: %s" ),
+ yytext );
+ }
+}
+
+<CARETISBOL>"^" BEGIN(SECT2); return '^';
+
+
+<QUOTE>{
+ [^"\n] RETURNCHAR;
+ \" BEGIN(SECT2); return '"';
+
+ {NL} {
+ synerr( _( "missing quote" ) );
+ BEGIN(SECT2);
+ ++linenum;
+ return '"';
+ }
+}
+
+
+<FIRSTCCL>{
+ "^"/[^-\]\n] BEGIN(CCL); return '^';
+ "^"/("-"|"]") return '^';
+ . BEGIN(CCL); RETURNCHAR;
+}
+
+<CCL>{
+ -/[^\]\n] return '-';
+ [^\]\n] RETURNCHAR;
+ "]" BEGIN(SECT2); return ']';
+ .|{NL} {
+ synerr( _( "bad character class" ) );
+ BEGIN(SECT2);
+ return ']';
+ }
+}
+
+<FIRSTCCL,CCL>{
+ "[:alnum:]" BEGIN(CCL); return CCE_ALNUM;
+ "[:alpha:]" BEGIN(CCL); return CCE_ALPHA;
+ "[:blank:]" BEGIN(CCL); return CCE_BLANK;
+ "[:cntrl:]" BEGIN(CCL); return CCE_CNTRL;
+ "[:digit:]" BEGIN(CCL); return CCE_DIGIT;
+ "[:graph:]" BEGIN(CCL); return CCE_GRAPH;
+ "[:lower:]" BEGIN(CCL); return CCE_LOWER;
+ "[:print:]" BEGIN(CCL); return CCE_PRINT;
+ "[:punct:]" BEGIN(CCL); return CCE_PUNCT;
+ "[:space:]" BEGIN(CCL); return CCE_SPACE;
+ "[:upper:]" BEGIN(CCL); return CCE_UPPER;
+ "[:xdigit:]" BEGIN(CCL); return CCE_XDIGIT;
+ {CCL_EXPR} {
+ format_synerr(
+ _( "bad character class expression: %s" ),
+ yytext );
+ BEGIN(CCL); return CCE_ALNUM;
+ }
+}
+
+<NUM>{
+ [[:digit:]]+ {
+ yylval = myctoi( yytext );
+ return NUMBER;
+ }
+
+ "," return ',';
+ "}" BEGIN(SECT2); return '}';
+
+ . {
+ synerr( _( "bad character inside {}'s" ) );
+ BEGIN(SECT2);
+ return '}';
+ }
+
+ {NL} {
+ synerr( _( "missing }" ) );
+ BEGIN(SECT2);
+ ++linenum;
+ return '}';
+ }
+}
+
+
+<PERCENT_BRACE_ACTION>{
+ {OPTWS}"%}".* bracelevel = 0;
+
+ <ACTION>"/*" ACTION_ECHO; yy_push_state( COMMENT );
+
+ <CODEBLOCK,ACTION>{
+ "reject" {
+ ACTION_ECHO;
+ CHECK_REJECT(yytext);
+ }
+ "yymore" {
+ ACTION_ECHO;
+ CHECK_YYMORE(yytext);
+ }
+ }
+
+ {NAME}|{NOT_NAME}|. ACTION_ECHO;
+ {NL} {
+ ++linenum;
+ ACTION_ECHO;
+ if ( bracelevel == 0 ||
+ (doing_codeblock && indented_code) )
+ {
+ if ( doing_rule_action )
+ add_action( "\tYY_BREAK\n" );
+
+ doing_rule_action = doing_codeblock = false;
+ BEGIN(SECT2);
+ }
+ }
+}
+
+
+ /* Reject and YYmore() are checked for above, in PERCENT_BRACE_ACTION */
+<ACTION>{
+ "{" ACTION_ECHO; ++bracelevel;
+ "}" ACTION_ECHO; --bracelevel;
+ [^[:alpha:]_{}"'/\n]+ ACTION_ECHO;
+ {NAME} ACTION_ECHO;
+ "'"([^'\\\n]|\\.)*"'" ACTION_ECHO; /* character constant */
+ \" ACTION_ECHO; BEGIN(ACTION_STRING);
+ {NL} {
+ ++linenum;
+ ACTION_ECHO;
+ if ( bracelevel == 0 )
+ {
+ if ( doing_rule_action )
+ add_action( "\tYY_BREAK\n" );
+
+ doing_rule_action = false;
+ BEGIN(SECT2);
+ }
+ }
+ . ACTION_ECHO;
+}
+
+<ACTION_STRING>{
+ [^"\\\n]+ ACTION_ECHO;
+ \\. ACTION_ECHO;
+ {NL} ++linenum; ACTION_ECHO;
+ \" ACTION_ECHO; BEGIN(ACTION);
+ . ACTION_ECHO;
+}
+
+<COMMENT,ACTION,ACTION_STRING><<EOF>> {
+ synerr( _( "EOF encountered inside an action" ) );
+ yyterminate();
+ }
+
+
+<SECT2,QUOTE,FIRSTCCL,CCL>{ESCSEQ} {
+ yylval = myesc( (Char *) yytext );
+
+ if ( YY_START == FIRSTCCL )
+ BEGIN(CCL);
+
+ return CHAR;
+ }
+
+
+<SECT3>{
+ .*(\n?) ECHO;
+ <<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 && strcmp( file, "-" ) )
+ {
+ infilename = copy_string( file );
+ yyin = fopen( infilename, "r" );
+
+ if ( yyin == NULL )
+ lerrsf( _( "can't open %s" ), file );
+ }
+
+ else
+ {
+ yyin = stdin;
+ infilename = copy_string( "<stdin>" );
+ }
+
+ linenum = 1;
+ }
+
+
+/* Wrapper routines for accessing the scanner's malloc routines. */
+
+void *flex_alloc( size )
+size_t size;
+ {
+ return (void *) malloc( size );
+ }
+
+void *flex_realloc( ptr, size )
+void *ptr;
+size_t size;
+ {
+ return (void *) realloc( ptr, size );
+ }
+
+void flex_free( ptr )
+void *ptr;
+ {
+ if ( ptr )
+ free( ptr );
+ }