pgc.l

/* This is a modified version of src/backend/parser/scan.l */
%{
#include "config.h"

#include <ctype.h>
#include <sys/types.h>
#include <limits.h>

#ifndef PATH_MAX
#include <sys/param.h>
#define PATH_MAX MAXPATHLEN
#endif

#if defined(HAVE_STRING_H)
#include <string.h>
#else
#include <strings.h>
#endif
#include <errno.h>

#include "postgres.h"
#include "miscadmin.h"
#include "nodes/pg_list.h"
#include "nodes/parsenodes.h"
#include "parser/gramparse.h"
#include "parser/scansup.h"
#include "type.h"
#include "y.tab.h"
#include "utils/builtins.h"

#include "extern.h"

/* some versions of lex define this as a macro */
#if defined(yywrap)
#undef yywrap
#endif /* yywrap */

int debugging = 0;
extern YYSTYPE yylval;
int llen;
char literal[MAX_PARSE_BUFFER];
int before_comment;

struct _yy_buffer { YY_BUFFER_STATE 	buffer;
		    long		lineno;
		    char	      * filename;
		    struct _yy_buffer * next;
		  } *yy_buffer = NULL;

%}
%option yylineno
%s C SQL incl
/* OK, here is a short description of lex/flex rules behavior.
 * The longest pattern which matches an input string is always chosen.
 * For equal-length patterns, the first occurring in the rules list is chosen.
 * INITIAL is the starting condition, to which all non-conditional rules apply.
 * When in an exclusive condition, only those rules defined for that condition apply.
 *
 * Exclusive states change parsing rules while the state is active.
 * There are exclusive states for quoted strings, extended comments,
 *  and to eliminate parsing troubles for numeric strings.
 * Exclusive states:
 *  <xb> binary numeric string - thomas 1997-11-16
 *  <xc> extended C-style comments - tgl 1997-07-12
 *  <xd> delimited identifiers (double-quoted identifiers) - tgl 1997-10-27
 *  <xh> hexadecimal numeric string - thomas 1997-11-16
 *  <xm> numeric strings with embedded minus sign - tgl 1997-09-05
 *  <xq> quoted strings - tgl 1997-07-30
 *
 * The "extended comment" syntax closely resembles allowable operator syntax.
 * So, when in condition <xc>, only strings which would terminate the
 *  "extended comment" trigger any action other than "ignore".
 * Be sure to match _any_ candidate comment, including those with appended
 *	operator-like symbols. - thomas 1997-07-14
 */

%x xb
%x xc
%x xd
%x xdc
%x xh
%x xm
%x xq

/* Binary number
 */
xbstart			[bB]{quote}
xbstop			{quote}
xbinside		[^']*
xbcat			{quote}{space}*\n{space}*{quote}

/* Hexadecimal number
 */
xhstart			[xX]{quote}
xhstop			{quote}
xhinside		[^']*
xhcat			{quote}{space}*\n{space}*{quote}

/* Extended quote
 * xqdouble implements SQL92 embedded quote
 * xqcat allows strings to cross input lines
 */
quote			'
xqstart			{quote}
xqstop			{quote}
xqdouble		{quote}{quote}
xqinside		[^\\']*
xqembedded		"\\'"
xqliteral		[\\](.|\n)
xqcat			{quote}{space}*\n{space}*{quote}

/* Delimited quote
 * Allows embedded spaces and other special characters into identifiers.
 */
dquote			\"
xdstart			{dquote}
xdstop			{dquote}
xdinside		[^"]*

/* Comments
 * Ignored by the scanner and parser.
 */
xcline			[\/][\*].*[\*][\/]{space}*\n*
xcstart			[\/][\*]{op_and_self}*
xcstop			{op_and_self}*[\*][\/]({space}*|\n)
xcinside		[^*]*
xcstar			[^/]

digit			[0-9]
number			[-+.0-9Ee]
letter			[\200-\377_A-Za-z]
letter_or_digit	[\200-\377_A-Za-z0-9]

identifier		{letter}{letter_or_digit}*

typecast		"::"

self			[,()\[\].$\:\+\-\*\/\<\>\=\|]
op_and_self		[\~\!\@\#\%\^\&\|\`\?\$\:\+\-\*\/\<\>\=]
operator		{op_and_self}+

xminteger              {integer}/-
xmreal                 {real}/{space}*-{digit}
xmstop			-

integer			-?{digit}+
real			-?{digit}+\.{digit}+([Ee][-+]?{digit}+)?

param			\${integer}

comment			("--"|"//").*\n

space			[ \t\n\f]
other			.

/* some stuff needed for ecpg */
ccomment	"//".*\n
exec    [eE][xX][eE][cC]
include [iI][nN][cC][lL][uU][dD][eE]
sql     [sS][qQ][lL]

/* DO NOT PUT ANY COMMENTS IN THE FOLLOWING SECTION.
 * AT&T lex does not properly handle C-style comments in this second lex block.
 * So, put comments here. tgl - 1997-09-08
 *
 * Quoted strings must allow some special characters such as single-quote
 *  and newline.
 * Embedded single-quotes are implemented both in the SQL/92-standard
 *  style of two adjacent single quotes "''" and in the Postgres/Java style
 *  of escaped-quote "\'".
 * Other embedded escaped characters are matched explicitly and the leading
 *  backslash is dropped from the string. - thomas 1997-09-24
 */

%%
<SQL>{comment}		{ /* ignore */ }

{xcline}		{ /* ignore */ }

<xc>{xcstar}		{ /* ignore */ }
{xcstart}		{
				fprintf(stderr,"ys = %d %d\n", YYSTATE, before_comment);
				before_comment = YYSTATE;
	 			BEGIN(xc);
				fprintf(stderr,"ys = %d %d\n", YYSTATE,
before_comment);
			}

<xc>{xcstop}	{ fprintf(stderr,"ys = %d %d\n", YYSTATE, before_comment);BEGIN(before_comment); }

<xc>{xcinside}	{ /* ignore */ }

<SQL>{xbstart}		{
					BEGIN(xb);
					llen = 0;
					*literal = '\0';
				}
<xb>{xbstop}	{
					char* endptr;

					BEGIN(SQL);
					errno = 0;
					yylval.ival = strtol((char *)literal,&endptr,2);
					if (*endptr != '\0' || errno == ERANGE)
						yyerror("ERROR: Bad binary integer input!");
					return (ICONST);
				}
<xh>{xhinside}	|
<xb>{xbinside}	{
					if ((llen+yyleng) > (MAX_PARSE_BUFFER - 1))
						yyerror("ERROR: quoted string parse buffer exceeded");
					memcpy(literal+llen, yytext, yyleng+1);
					llen += yyleng;
				}
<xh>{xhcat}		|
<xb>{xbcat}		{
				}

<SQL>{xhstart}		{
					BEGIN(xh);
					llen = 0;
					*literal = '\0';
				}
<xh>{xhstop}	{
					char* endptr;

					BEGIN(SQL);
					errno = 0;
					yylval.ival = strtol((char *)literal,&endptr,16);
					if (*endptr != '\0' || errno == ERANGE)
						yyerror("ERROR: Bad hexadecimal integer input");
					return (ICONST);
				}

<SQL>{xqstart}		{
					BEGIN(xq);
					llen = 0;
					*literal = '\0';
				}
<xq>{xqstop}	{
					BEGIN(SQL);
					yylval.str = strdup(scanstr(literal));
					return (SCONST);
				}
<xq>{xqdouble}	|
<xq>{xqinside}	{
					if ((llen+yyleng) > (MAX_PARSE_BUFFER - 1))
						yyerror("ERROR: quoted string parse buffer exceeded");
					memcpy(literal+llen, yytext, yyleng+1);
					llen += yyleng;
				}
<xq>{xqembedded} {
					if ((llen+yyleng-1) > (MAX_PARSE_BUFFER - 1))
						yyerror("ERROR: quoted string parse buffer exceeded");
					memcpy(literal+llen, yytext, yyleng+1);
					*(literal+llen) = '\'';
					llen += yyleng;
				}

<xq>{xqliteral} {
					if ((llen+yyleng-1) > (MAX_PARSE_BUFFER - 1))
						yyerror("ERROR: quoted string parse buffer exceeded");
					memcpy(literal+llen, yytext, yyleng+1);
					llen += yyleng;
				}
<xq>{xqcat}		{
				}


<SQL>{xdstart}		{
					BEGIN(xd);
					llen = 0;
					*literal = '\0';
				}
<xd>{xdstop}	{
					BEGIN(SQL);
					yylval.str = strdup(literal);
					return (CSTRING);
				}
<xd>{xdinside}	{
					if ((llen+yyleng) > (MAX_PARSE_BUFFER - 1))
						yyerror("ERROR: quoted string parse buffer exceeded");
					memcpy(literal+llen, yytext, yyleng+1);
					llen += yyleng;
				}
<C>{xdstart}		{
					BEGIN(xdc);
					llen = 0;
					*literal = '\0';
				}
<xdc>{xdstop}	{
					BEGIN(C);
					yylval.str = strdup(literal);
					return (CSTRING);
				}
<xdc>{xdinside}	{
					if ((llen+yyleng) > (MAX_PARSE_BUFFER - 1))
						yyerror("ERROR: quoted string parse buffer exceeded");
					memcpy(literal+llen, yytext, yyleng+1);
					llen += yyleng;
				}

<xm>{space}*	{ /* ignore */ }
<xm>{xmstop}	{
					BEGIN(SQL);
					return (yytext[0]);
				}


<SQL>{typecast}			{ 	return TYPECAST; }

<SQL>{self}/-[\.0-9]		{
					return (yytext[0]);
				}
<SQL>{self}				{ 	return (yytext[0]); }
<SQL>{operator}/-[\.0-9]	{
					yylval.str = strdup((char*)yytext);
					return (Op);
				}
<SQL>{operator}		{
					if (strcmp((char*)yytext,"!=") == 0)
						yylval.str = strdup("<>"); /* compatability */
					else
						yylval.str = strdup((char*)yytext);
					return (Op);
				}
<SQL>{param}			{
					yylval.ival = atoi((char*)&yytext[1]);
					return (PARAM);
				}

<SQL>{identifier}/{space}*-{number}	{
					int i;
					ScanKeyword		*keyword;

					BEGIN(xm);
					for(i = 0; yytext[i]; i++)
						if (isupper(yytext[i]))
							yytext[i] = tolower(yytext[i]);

					keyword = ScanKeywordLookup((char*)yytext);
					if (keyword != NULL) {
						return (keyword->value);
					}
					else
					{
						keyword = ScanECPGKeywordLookup((char*)yytext);
						if (keyword != NULL) {
							return (keyword->value);
						}
						else
						{
							yylval.str = strdup((char*)yytext);
							return (IDENT);
						}
					}
				}
{integer}/{space}*-{number}	{
					char* endptr;

					BEGIN(xm);
					errno = 0;
					yylval.ival = strtol((char *)yytext,&endptr,10);
					if (*endptr != '\0' || errno == ERANGE)
					{
						errno = 0;
						yylval.dval = strtod(((char *)yytext),&endptr);
						if (*endptr != '\0' || errno == ERANGE)
							yyerror("ERROR: Bad integer input");
						yyerror("WARNING: Integer input is out of range; promoted to float");
						return (FCONST);
					}
					return (ICONST);
				}
{real}/{space}*-{number} {
					char* endptr;

					BEGIN(xm);
					errno = 0;
					yylval.dval = strtod(((char *)yytext),&endptr);
					if (*endptr != '\0' || errno == ERANGE)
						yyerror("ERROR: Bad float8 input");
					return (FCONST);
				}
{integer}		{
					char* endptr;

					errno = 0;
					yylval.ival = strtol((char *)yytext,&endptr,10);
					if (*endptr != '\0' || errno == ERANGE)
					{
						errno = 0;
						yylval.dval = strtod(((char *)yytext),&endptr);
						if (*endptr != '\0' || errno == ERANGE)
							yyerror("ERROR: Bad integer input");
						yyerror("WARNING: Integer input is out of range; promoted to float");
						return (FCONST);
					}
					return (ICONST);
				}
{real}			{
					char* endptr;

					errno = 0;
					yylval.dval = strtod((char *)yytext,&endptr);
					if (*endptr != '\0' || errno == ERANGE)
						yyerror("ERROR: Bad float input");
					return (FCONST);
				}

<SQL>:{identifier}(("->"|\.){identifier})*	{
					yylval.str = strdup((char*)yytext+1);
					return(CVARIABLE);
			}
<SQL>{identifier}	{
					int i;
					ScanKeyword		*keyword;

					for(i = 0; yytext[i]; i++)
						if (isupper(yytext[i]))
							yytext[i] = tolower(yytext[i]);

					keyword = ScanKeywordLookup((char*)yytext);
					if (keyword != NULL) {
						return (keyword->value);
					}
					else
					{
						keyword = ScanECPGKeywordLookup((char*)yytext);
						if (keyword != NULL) {
							return (keyword->value);
						}
						else
						{
							yylval.str = strdup((char*)yytext);
							return (IDENT);
						}
					}
				}
<SQL>{space}			{ /* ignore */ }
<SQL>";"	                { BEGIN C; return SQL_SEMI; }
<SQL>{other}			{ return (yytext[0]); }

<C>{exec}{space}{sql}		{ BEGIN SQL; return SQL_START; }
<C>{ccomment}			{ /* ignore */ } 
<C>{identifier}	{
					ScanKeyword		*keyword;

					keyword = ScanCKeywordLookup((char*)yytext);
					if (keyword != NULL) {
						return (keyword->value);
					}
					else
					{
						yylval.str = strdup((char*)yytext);
						return (IDENT);
					}
				}
<C>";"	      	        { return(';'); }
<C>","	      	        { return(','); }
<C>"*"	      	        { return('*'); }
<C>{space}		{ ECHO; }
<C>\{			{ return('{'); }
<C>\}			{ return('}'); }
<C>\[			{ return('['); }
<C>\]			{ return(']'); }
<C>\=			{ return('='); }
<C>{other}			{ return (S_ANYTHING); }
<C>{exec}{space}{sql}{space}{include}	{ BEGIN(incl); }
<incl>{space}		/* eat the whitespace */
<incl>[^ \t\n]+ 	{ /* got the include file name */
			  struct _yy_buffer *yb;
			  struct _include_path *ip;
			  char inc_file[PATH_MAX];

			  yb = mm_alloc(sizeof(struct _yy_buffer));

			  yb->buffer =  YY_CURRENT_BUFFER;
			  yb->lineno = yylineno;
			  yb->filename = input_filename;
			  yb->next = yy_buffer;

			  yy_buffer = yb;

			  if (yytext[strlen(yytext) - 1] == ';')
				yytext[strlen(yytext) - 1] = '\0';

			  yyin = NULL;
			  for (ip = include_paths; yyin == NULL && ip != NULL; ip = ip->next)
			  {
				if (strlen(ip->path) + strlen(yytext) + 3 > PATH_MAX)
				{
					fprintf(stderr, "Error: Path %s/%s is too long in line %d, skipping.\n", ip->path, yytext, yylineno);
					continue;
				}
			  	sprintf (inc_file, "%s/%s", ip->path, yytext);
		          	yyin = fopen( inc_file, "r" );
			  	if (!yyin)
				{
					if (strcmp(inc_file + strlen(inc_file) - 2, ".h"))
					{
						strcat(inc_file, ".h");
						yyin = fopen( inc_file, "r" );
					}

				}
			  }
			  if (!yyin)
			  {
				fprintf(stderr, "Error: Cannot open include file %s in line %d\n", yytext, yylineno);
				exit(NO_INCLUDE_FILE); 
			  }

			  input_filename = strdup(inc_file);
			  yy_switch_to_buffer(yy_create_buffer(yyin,YY_BUF_SIZE ));
			  yylineno = 0;

			  BEGIN C;
			}
<incl>";"		{ BEGIN C; }
<<EOF>>			{ if (yy_buffer == NULL)
				yyterminate();
			  else
			  {
				struct _yy_buffer *yb = yy_buffer;

				if (yyin != NULL)
					fclose(yyin);

				yy_delete_buffer( YY_CURRENT_BUFFER );
				yy_switch_to_buffer(yy_buffer->buffer);

				yylineno = yy_buffer->lineno;

				free(input_filename);
				input_filename = yy_buffer->filename;

				yy_buffer = yy_buffer->next;
				free(yb);
			  }
			}

%%
void
lex_init(void)
{
    braces_open = 0;
    BEGIN C;
}

int yywrap(void) 
{ 
    return 1;
}