pgc.l

postgresql8.3.4源码,开源数据库

<xe>{xehexesc}		{ addlit(yytext, yyleng); }
<xq,xqc,xe,xn>{quotecontinue}	{ /* ignore */ }
<xe>.                   {
			   /* This is only needed for \ just before EOF */
			   addlitchar(yytext[0]);
			}
<xq,xqc,xe,xn><<EOF>>	{ mmerror(PARSE_ERROR, ET_FATAL, "Unterminated quoted string"); }
<SQL>{dolqfailed}	{
				/* throw back all but the initial "$" */
				yyless(1);
				/* and treat it as {other} */
				return yytext[0];	
			}
<SQL>{dolqdelim}        {
				token_start = yytext;
				dolqstart = mm_strdup(yytext);
				BEGIN(xdolq);
				startlit();
				addlit(yytext, yyleng);
			}
<xdolq>{dolqdelim}      {
				if (strcmp(yytext, dolqstart) == 0)
				{
					addlit(yytext, yyleng);
					free(dolqstart);
					BEGIN(SQL);
					yylval.str = mm_strdup(literalbuf);
					return DOLCONST;
				}
				else
				{
			        /*
			         * When we fail to match $...$ to dolqstart, transfer
			         * the $... part to the output, but put back the final
			         * $ for rescanning.  Consider $delim$...$junk$delim$
			         */
			        addlit(yytext, yyleng-1);
			        yyless(yyleng-1);
				}
			}
<xdolq>{dolqinside} 	{ addlit(yytext, yyleng); }
<xdolq>{dolqfailed}	{ addlit(yytext, yyleng); }
<xdolq>{other}		{
				/* single quote or dollar sign */
				addlitchar(yytext[0]);
			}
<xdolq><<EOF>> 		{ base_yyerror("unterminated dollar-quoted string"); }
<SQL>{xdstart}		{
						state_before = YYSTATE;
						BEGIN(xd);
						startlit();
					}
<xd>{xdstop}		{
						BEGIN(state_before);
						if (literallen == 0)
							mmerror(PARSE_ERROR, ET_ERROR, "zero-length delimited identifier");
						/* The backend will truncate the idnetifier here. We do not as it does not change the result. */
						yylval.str = mm_strdup(literalbuf);
						return CSTRING;
					}
<xdc>{xdstop}		{
						BEGIN(state_before);
						yylval.str = mm_strdup(literalbuf);
						return CSTRING;
					}
<xd>{xddouble}		{ addlitchar('"'); }
<xd>{xdinside}		{ addlit(yytext, yyleng); }
<xd,xdc><<EOF>>		{ mmerror(PARSE_ERROR, ET_FATAL, "Unterminated quoted identifier"); }
<C,SQL>{xdstart}	{
						state_before = YYSTATE;
						BEGIN(xdc);
						startlit();
					}
<xdc>{xdcinside}	{ addlit(yytext, yyleng); }
<SQL>{typecast}		{ return TYPECAST; }
<SQL>{informix_special}	{
			  /* are we simulating Informix? */
				if (INFORMIX_MODE)
				{
					unput(':');
				}
				else
					return yytext[0];
				}
<SQL>{self}			{ /*
					   * We may find a ';' inside a structure
					   * definition in a TYPE or VAR statement.
					   * This is not an EOL marker.
					   */
					  if (yytext[0] == ';' && struct_level == 0)
						 BEGIN(C);
					  return yytext[0];
					}
<SQL>{operator}		{
						/*
						 * Check for embedded slash-star or dash-dash; those
						 * are comment starts, so operator must stop there.
						 * Note that slash-star or dash-dash at the first
						 * character will match a prior rule, not this one.
						 */
						int		nchars = yyleng;
						char   *slashstar = strstr(yytext, "/*");
						char   *dashdash = strstr(yytext, "--");

						if (slashstar && dashdash)
						{
							/* if both appear, take the first one */
							if (slashstar > dashdash)
								slashstar = dashdash;
						}
						else if (!slashstar)
							slashstar = dashdash;
						if (slashstar)
							nchars = slashstar - yytext;

						/*
						 * For SQL compatibility, '+' and '-' cannot be the
						 * last char of a multi-char operator unless the operator
						 * contains chars that are not in SQL operators.
						 * The idea is to lex '=-' as two operators, but not
						 * to forbid operator names like '?-' that could not be
						 * sequences of SQL operators.
						 */
						while (nchars > 1 &&
							   (yytext[nchars-1] == '+' ||
								yytext[nchars-1] == '-'))
						{
							int		ic;

							for (ic = nchars-2; ic >= 0; ic--)
							{
								if (strchr("~!@#^&|`?%", yytext[ic]))
									break;
							}
							if (ic >= 0)
								break; /* found a char that makes it OK */
							nchars--; /* else remove the +/-, and check again */
						}

						if (nchars < yyleng)
						{
							/* Strip the unwanted chars from the token */
							yyless(nchars);
							/*
							 * If what we have left is only one char, and it's
							 * one of the characters matching "self", then
							 * return it as a character token the same way
							 * that the "self" rule would have.
							 */
							if (nchars == 1 &&
								strchr(",()[].;:+-*/%^<>=", yytext[0]))
								return yytext[0];
						}

						/* Convert "!=" operator to "<>" for compatibility */
						if (strcmp(yytext, "!=") == 0)
							yylval.str = mm_strdup("<>");
						else
							yylval.str = mm_strdup(yytext);
						return Op;
					}
<SQL>{param}		{
						yylval.ival = atol(yytext+1);
						return PARAM;
					}
<C,SQL>{integer}	{
						long val;
						char* endptr;

						errno = 0;
						val = strtol((char *)yytext, &endptr,10);
						if (*endptr != '\0' || errno == ERANGE
#ifdef HAVE_LONG_INT_64
							/* if long > 32 bits, check for overflow of int4 */
							|| val != (long) ((int32) val)
#endif
							)
						{
							errno = 0;
							yylval.str = mm_strdup(yytext);
							return FCONST;
						}
						yylval.ival = val;
						return ICONST;
					}
<SQL>{ip}			{
						yylval.str = mm_strdup(yytext);
						return IP;
				}
<C,SQL>{decimal}	{
						yylval.str = mm_strdup(yytext);
						return FCONST;
			}
<C,SQL>{real}		{
						yylval.str = mm_strdup(yytext);
						return FCONST;
			}
<SQL>{realfail1}	{
						yyless(yyleng-1);
						yylval.str = mm_strdup(yytext);
						return FCONST;
					}
<SQL>{realfail2}	{
						yyless(yyleng-2);
						yylval.str = mm_strdup(yytext);
						return FCONST;
					}
<SQL>:{identifier}((("->"|\.){identifier})|(\[{array}\]))*	{
						yylval.str = mm_strdup(yytext+1);
						return(CVARIABLE);
					}
<SQL>{identifier}	{
						const ScanKeyword    *keyword;
						
						if (!isdefine())
						{
							/* Is it an SQL/ECPG keyword? */
							keyword = ScanKeywordLookup(yytext);
							if (keyword != NULL)
								return keyword->value;

							/* Is it a C keyword? */
							keyword = ScanCKeywordLookup(yytext);
							if (keyword != NULL)
								return keyword->value;

							/*
							 * None of the above.  Return it as an identifier.
							 *
							 * The backend will attempt to truncate and case-fold
							 * the identifier, but I see no good reason for ecpg
							 * to do so; that's just another way that ecpg could get
							 * out of step with the backend.
							 */
							yylval.str = mm_strdup(yytext);
							return IDENT;
						}
					}
<SQL>{other}		{ return yytext[0]; }
<C>{exec_sql}		{ BEGIN(SQL); return SQL_START; }
<C>{informix_special}	{ 
						/* are we simulating Informix? */
						if (INFORMIX_MODE)
						{
							BEGIN(SQL);
							return SQL_START;
						}
						else
							return S_ANYTHING;
					 }
<C>{ccomment}		{ ECHO; }
<C>{xch}			{
						char* endptr;

						errno = 0;
						yylval.ival = strtoul((char *)yytext,&endptr,16);
						if (*endptr != '\0' || errno == ERANGE)
						{
							errno = 0;
							yylval.str = mm_strdup(yytext);
							return SCONST;
						}
						return ICONST;
					}
<C>{cppinclude}		{
						if (system_includes)
						{
							BEGIN(incl);
						}
						else
					 	{
							yylval.str = mm_strdup(yytext);
							return(CPP_LINE);
					  	}
					}
<C,SQL>{cppline}	{
						yylval.str = mm_strdup(yytext);
						return(CPP_LINE);
					}
<C>{identifier} 	{
						const ScanKeyword		*keyword;
							
						/* Informix uses SQL defines only in SQL space */
						/* however, some defines have to be taken care of for compatibility */
						if ((!INFORMIX_MODE || !isinformixdefine()) && !isdefine())
						{
							keyword = ScanCKeywordLookup(yytext);
							if (keyword != NULL)
								return keyword->value;
							else
							{
								yylval.str = mm_strdup(yytext);
								return IDENT;
							}
						}
					}
<C>":"				{ return(':'); }
<C>";"				{ return(';'); }
<C>","				{ return(','); }
<C>"*"				{ return('*'); }
<C>"%"				{ return('%'); }
<C>"/"				{ return('/'); }
<C>"+"				{ return('+'); }
<C>"-"				{ return('-'); }
<C>"("				{ return('('); }
<C>")"				{ return(')'); }
<C,xskip>{space}		{ ECHO; }
<C>\{				{ return('{'); }
<C>\}				{ return('}'); }
<C>\[				{ return('['); }
<C>\]				{ return(']'); }
<C>\=				{ return('='); }
<C>"->"				{ return(S_MEMBER); }
<C>">>"				{ return(S_RSHIFT); }
<C>"<<"				{ return(S_LSHIFT); }
<C>"||"				{ return(S_OR); }
<C>"&&"				{ return(S_AND); }
<C>"++"				{ return(S_INC); }
<C>"--"				{ return(S_DEC); }
<C>"=="				{ return(S_EQUAL); }
<C>"!="				{ return(S_NEQUAL); }
<C>"+="				{ return(S_ADD); }
<C>"-="				{ return(S_SUB); }
<C>"*="				{ return(S_MUL); }
<C>"/="				{ return(S_DIV); }
<C>"%="				{ return(S_MOD); }
<C>"->*"			{ return(S_MEMPOINT); }
<C>".*"				{ return(S_DOTPOINT); }
<C>{other}			{ return S_ANYTHING; }
<C>{exec_sql}{define}{space}*	{ BEGIN(def_ident); }
<C>{informix_special}{define}{space}*	{
						/* are we simulating Informix? */
						if (INFORMIX_MODE)
						{
							BEGIN(def_ident);
						}
						else
						{
							yyless(1);
							return (S_ANYTHING);
						}
					}
<C>{exec_sql}{undef}{space}*		{ BEGIN(undef); }
<C>{informix_special}{undef}{space}* 	{
						/* are we simulating Informix? */
						if (INFORMIX_MODE)
						{
							BEGIN(undef);
						}
						else
						{
							yyless(1);
							return (S_ANYTHING);
						}
					}
<undef>{identifier}{space}*";" {
					struct _defines *ptr, *ptr2 = NULL;
					int i;

					/*
					 *	Skip the ";" and trailing whitespace. Note that yytext
					 *	contains at least one non-space character plus the ";"
					 */
					for (i = strlen(yytext)-2;
						 i > 0 && ecpg_isspace(yytext[i]);
						 i-- )
						;
					yytext[i+1] = '\0';


					for (ptr = defines; ptr != NULL; ptr2 = ptr, ptr = ptr->next)
					{
						if (strcmp(yytext, ptr->old) == 0)
						{
							if (ptr2 == NULL)
								defines = ptr->next;
							else
								ptr2->next = ptr->next;
							free(ptr->new);
							free(ptr->old);
							free(ptr);
							break;
						}
					}

					BEGIN(C);
				}
<undef>{other}|\n  		{
		                	mmerror(PARSE_ERROR, ET_FATAL, "Missing identifier in 'EXEC SQL UNDEF' command");
			                yyterminate();
				}
<C>{exec_sql}{include}{space}*	{ BEGIN(incl); }
<C>{informix_special}{include}{space}* { 
					  /* are we simulating Informix? */
					  if (INFORMIX_MODE)
					  {
						  BEGIN(incl);
					  }
					  else
					  {
						  yyless(1);
						  return (S_ANYTHING);
					  }
					}
<C,xskip>{exec_sql}{ifdef}{space}*	{ ifcond = TRUE; BEGIN(xcond); }
<C,xskip>{informix_special}{ifdef}{space}* { 
					  /* are we simulating Informix? */
					  if (INFORMIX_MODE)
					  {
						  ifcond = TRUE;
						  BEGIN(xcond);
					  }
					  else
					  {
						  yyless(1);
						  return (S_ANYTHING);
					  }
					}
<C,xskip>{exec_sql}{ifndef}{space}* { ifcond = FALSE; BEGIN(xcond); }
<C,xskip>{informix_special}{ifndef}{space}* { 
					  /* are we simulating Informix? */
					  if (INFORMIX_MODE)
					  {
						  ifcond = FALSE;
						  BEGIN(xcond);
					  }
					  else
					  {
						  yyless(1);
						  return (S_ANYTHING);
					  }
					}
<C,xskip>{exec_sql}{elif}{space}*	{	/* pop stack */
						if ( preproc_tos == 0 ) {
							mmerror(PARSE_ERROR, ET_FATAL, "Missing matching 'EXEC SQL IFDEF / EXEC SQL IFNDEF'");
						}
						else if ( stacked_if_value[preproc_tos].else_branch )
							mmerror(PARSE_ERROR, ET_FATAL, "Missing 'EXEC SQL ENDIF;'");
						else
							preproc_tos--;

						ifcond = TRUE; BEGIN(xcond);
					}
<C,xskip>{informix_special}{elif}{space}* { 
					/* are we simulating Informix? */
					if (INFORMIX_MODE)
					{
						if (preproc_tos == 0)
							mmerror(PARSE_ERROR, ET_FATAL, "Missing matching 'EXEC SQL IFDEF / EXEC SQL IFNDEF'");
						else if (stacked_if_value[preproc_tos].else_branch)
							mmerror(PARSE_ERROR, ET_FATAL, "Missing 'EXEC SQL ENDIF;'");
						else
							preproc_tos--;

						ifcond = TRUE;
						BEGIN(xcond);
					}
					else
					{
						yyless(1);
						return (S_ANYTHING);
					}
				}