diff options
Diffstat (limited to 'idlc/source/scanner.ll')
| -rw-r--r-- | idlc/source/scanner.ll | 523 |
1 files changed, 523 insertions, 0 deletions
diff --git a/idlc/source/scanner.ll b/idlc/source/scanner.ll new file mode 100644 index 000000000000..4125195f5916 --- /dev/null +++ b/idlc/source/scanner.ll @@ -0,0 +1,523 @@ +/************************************************************************* + * + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. + * + * Copyright 2000, 2010 Oracle and/or its affiliates. + * + * OpenOffice.org - a multi-platform office productivity suite + * + * This file is part of OpenOffice.org. + * + * OpenOffice.org is free software: you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License version 3 + * only, as published by the Free Software Foundation. + * + * OpenOffice.org is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License version 3 for more details + * (a copy is included in the LICENSE file that accompanied this code). + * + * You should have received a copy of the GNU Lesser General Public License + * version 3 along with OpenOffice.org. If not, see + * <http://www.openoffice.org/license.html> + * for a copy of the LGPLv3 License. + * + ************************************************************************/ + +%{ +/* + * scanner.ll - Lexical scanner for IDLC 1.0 + */ + +#include <ctype.h> +#include <stdlib.h> +#include <string.h> + +#ifndef _IDLC_IDLC_HXX_ +#include <idlc/idlc.hxx> +#endif +#ifndef _IDLC_ERRORHANDLER_HXX_ +#include <idlc/errorhandler.hxx> +#endif +#ifndef _IDLC_FEHELPER_HXX_ +#include <idlc/fehelper.hxx> +#endif + +#include "attributeexceptions.hxx" + +class AstExpression; +class AstArray; +class AstMember; + +#include <parser.hxx> + +sal_Int32 beginLine = 0; +::rtl::OString docu; + +static int asciiToInteger(char const * s, sal_Int64 * sval, sal_uInt64 * uval) { + bool neg = false; + if (*s == '-') { + neg = true; + ++s; + } + unsigned int base = 10; + if (*s == '0') { + base = 8; + ++s; + if (*s == 'X' || *s == 'x') { + base = 16; + ++s; + } + } + sal_uInt64 val = 0; + for (; *s != 0; ++s) { + unsigned int n; + if (*s >= '0' && *s <= '9') { + n = *s - '0'; + } else { + switch (*s) { + case 'A': + case 'a': + n = 10; + break; + case 'B': + case 'b': + n = 11; + break; + case 'C': + case 'c': + n = 12; + break; + case 'D': + case 'd': + n = 13; + break; + case 'E': + case 'e': + n = 14; + break; + case 'F': + case 'f': + n = 15; + break; + default: + goto done; + } + } + // The following guarantees the invariant val <= SAL_MAX_UINT64 (because + // base and n are sufficiently small), *if* + // std::numeric_limits<sal_uInt64>::max() == SAL_MAX_UINT64: + sal_uInt64 nval = val * base + n; + if (nval < val) { + idlc()->error()->syntaxError( + PS_NoState, idlc()->getLineNumber(), + "integral constant too large"); + val = 0; + break; + } + val = nval; + } + done: + if (neg) { + if (val < SAL_CONST_UINT64(0x8000000000000000)) { + *sval = -static_cast< sal_Int64 >(val); + } else if (val == SAL_CONST_UINT64(0x8000000000000000)) { + *sval = SAL_MIN_INT64; + } else { + idlc()->error()->syntaxError( + PS_NoState, idlc()->getLineNumber(), + "negative integral constant too large"); + *sval = 0; + } + return IDL_INTEGER_LITERAL; + } else if (val <= static_cast< sal_uInt64 >(SAL_MAX_INT64)) { + *sval = static_cast< sal_Int64 >(val); + return IDL_INTEGER_LITERAL; + } else { + *uval = val; + return IDL_INTEGER_ULITERAL; + } +} + +static double asciiToFloat(const sal_Char *s) +{ + double d = 0.0; + double e, k; + sal_Int32 neg = 0, negexp = 0; + + if (*s == '-') + { + neg = 1; + s++; + } + while (*s >= '0' && *s <= '9') + { + d = (d * 10) + *s - '0'; + s++; + } + if (*s == '.') + { + s++; + e = 10; + while (*s >= '0' && *s <= '9') + { + d += (*s - '0') / (e * 1.0); + e *= 10; + s++; + } + } + if (*s == 'e' || *s == 'E') + { + s++; + if (*s == '-') + { + negexp = 1; + s++; + } else + { + if (*s == '+') + s++; + e = 0; + while (*s >= '0' && *s <= '9') + { + e = (e * 10) + *s - '0'; + s++; + } + if (e > 0) + { + for (k = 1; e > 0; k *= 10, e--) ; + if (negexp) + d /= k; + else + d *= k; + } + } + } + if (neg) d *= -1.0; + return d; +} + +static void idlParsePragma(sal_Char* pPragma) +{ + ::rtl::OString pragma(pPragma); + sal_Int32 index = pragma.indexOf("include"); + sal_Char* begin = pPragma + index + 8; + sal_Char* offset = begin; + while (*offset != ',') offset++; + //::rtl::OString include = pragma.copy(index + 8, offset - begin); + idlc()->insertInclude(pragma.copy(index + 8, (sal_Int32)(offset - begin))); +} + +static void parseLineAndFile(sal_Char* pBuf) +{ + sal_Char *r = pBuf; + sal_Char *h; + sal_Bool bIsInMain = sal_False; + + /* Skip initial '#' */ + if (*r != '#') + return; + + /* Find line number */ + for (r++; *r == ' ' || *r == '\t' || isalpha(*r); r++) ; + h = r; + for (; *r != '\0' && *r != ' ' && *r != '\t'; r++) ; + *r++ = 0; + idlc()->setLineNumber((sal_uInt32)atol(h)); + + /* Find file name, if present */ + for (; *r != '"'; r++) + { + if (*r == '\n' || *r == '\0') + return; + } + h = ++r; + for (; *r != '"'; r++) ; + *r = 0; + if (*h == '\0') + idlc()->setFileName(::rtl::OString("standard input")); + else + idlc()->setFileName(::rtl::OString(h)); + + bIsInMain = (idlc()->getFileName() == idlc()->getRealFileName()) ? sal_True : sal_False; + idlc()->setInMainfile(bIsInMain); +} + +// Suppress any warnings from generated code: +#if defined __GNUC__ +#pragma GCC system_header +#elif defined __SUNPRO_CC +#pragma disable_warn +#elif defined _MSC_VER +#pragma warning(push, 1) +/**/ +#ifdef yywrap +#undef yywrap +#define yywrap() 1 +#endif +/**/ +#endif +%} + +%option noyywrap +%option never-interactive + +%x DOCU +%x COMMENT + +DIGIT [0-9] +OCT_DIGIT [0-7] +HEX_DIGIT [a-fA-F0-9] +CAPITAL [A-Z] +ALPHA [a-zA-Z] +INT_LITERAL [1-9][0-9]* +OCT_LITERAL 0{OCT_DIGIT}* +HEX_LITERAL (0x|0X){HEX_DIGIT}* + +IDENTIFIER_NEW ({ALPHA}({ALPHA}|{DIGIT})*)|({CAPITAL}("_"?({ALPHA}|{DIGIT})+)*) +IDENTIFIER ("_"?({ALPHA}|{DIGIT})+)* + +%% + +[ \t\r]+ ; /* eat up whitespace */ +[\n] { + idlc()->incLineNumber(); +} + +attribute return IDL_ATTRIBUTE; +bound return IDL_BOUND; +case return IDL_CASE; +const return IDL_CONST; +constants return IDL_CONSTANTS; +constrained return IDL_CONSTRAINED; +default return IDL_DEFAULT; +enum return IDL_ENUM; +exception return IDL_EXCEPTION; +interface return IDL_INTERFACE; +maybeambiguous return IDL_MAYBEAMBIGUOUS; +maybedefault return IDL_MAYBEDEFAULT; +maybevoid return IDL_MAYBEVOID; +module return IDL_MODULE; +needs return IDL_NEEDS; +observes return IDL_OBSERVES; +optional return IDL_OPTIONAL; +property return IDL_PROPERTY; +raises return IDL_RAISES; +readonly return IDL_READONLY; +removable return IDL_REMOVEABLE; +service return IDL_SERVICE; +sequence return IDL_SEQUENCE; +singleton return IDL_SINGLETON; +struct return IDL_STRUCT; +switch return IDL_SWITCH; +transient return IDL_TRANSIENT; +typedef return IDL_TYPEDEF; +union return IDL_UNION; + +any return IDL_ANY; +boolean return IDL_BOOLEAN; +byte return IDL_BYTE; +char return IDL_CHAR; +double return IDL_DOUBLE; +float return IDL_FLOAT; +hyper return IDL_HYPER; +long return IDL_LONG; +short return IDL_SHORT; +string return IDL_STRING; +type return IDL_TYPE; +unsigned return IDL_UNSIGNED; +void return IDL_VOID; + +TRUE return IDL_TRUE; +True return IDL_TRUE; +FALSE return IDL_FALSE; +False return IDL_FALSE; + +in return IDL_IN; +out return IDL_OUT; +inout return IDL_INOUT; +oneway return IDL_ONEWAY; + +get return IDL_GET; +set return IDL_SET; + +published return IDL_PUBLISHED; + +"..." return IDL_ELLIPSIS; + +("-")?{INT_LITERAL}+(l|L|u|U)? { + return asciiToInteger(yytext, &yylval.ival, &yylval.uval); + } + +("-")?{OCT_LITERAL}+(l|L|u|U)? { + return asciiToInteger(yytext, &yylval.ival, &yylval.uval); + } + +("-")?{HEX_LITERAL}+(l|L|u|U)? { + return asciiToInteger(yytext, &yylval.ival, &yylval.uval); + } + +("-")?{DIGIT}+(e|E){1}(("+"|"-")?{DIGIT}+)+(f|F)? | +("-")?"."{DIGIT}+((e|E)("+"|"-")?{DIGIT}+)?(f|F)? | +("-")?{DIGIT}*"."{DIGIT}+((e|E)("+"|"-")?{DIGIT}+)?(f|F)? { + yylval.dval = asciiToFloat( yytext ); + return IDL_FLOATING_PT_LITERAL; + } + +{IDENTIFIER} { + yylval.sval = new ::rtl::OString(yytext); + return IDL_IDENTIFIER; + } + +\<\< { + yylval.strval = yytext; + return IDL_LEFTSHIFT; + } +\>\> { + yylval.strval = yytext; + return IDL_RIGHTSHIFT; + } +\:\: { + yylval.strval = yytext; + return IDL_SCOPESEPARATOR; + } + +"/*" { + BEGIN( COMMENT ); + docu = ::rtl::OString(); + beginLine = idlc()->getLineNumber(); + } + +"/***" { + BEGIN( COMMENT ); + docu = ::rtl::OString(); + beginLine = idlc()->getLineNumber(); + } + +<COMMENT>[^*]+ { + docu += ::rtl::OString(yytext); + } + +<COMMENT>"*"[^*/]+ { + docu += ::rtl::OString(yytext); + } + +<COMMENT>"**" { + docu += ::rtl::OString(yytext); + } + +<COMMENT>[*]+"/" { + docu = docu.trim(); + sal_Int32 nIndex = 0; + int count = 0; + do { docu.getToken( 0, '\n', nIndex ); count++; } while( nIndex != -1 ); + idlc()->setLineNumber( beginLine + count - 1); + BEGIN( INITIAL ); + } + +"/**" { + BEGIN( DOCU ); + docu = ::rtl::OString(); + beginLine = idlc()->getLineNumber(); + } + +<DOCU>[^*\n]+ { + docu += ::rtl::OString(yytext); + } + +<DOCU>"\n"[ \t]*"*"{1} { + idlc()->setLineNumber( idlc()->getLineNumber() + 1); + docu += ::rtl::OString("\n"); + } + +<DOCU>"\n" { + idlc()->setLineNumber( idlc()->getLineNumber() + 1); + docu += ::rtl::OString(yytext); + } + +<DOCU>"*"[^*^/\n]* { + docu += ::rtl::OString(yytext); + } + +<DOCU>"\n"[ \t]*"*/" { + docu = docu.trim(); + sal_Int32 nIndex = 0; + int count = 0; + do { docu.getToken( 0, '\n', nIndex ); count++; } while( nIndex != -1 ); + idlc()->setLineNumber( beginLine + count - 1); + if ( (nIndex = docu.indexOf("/*")) >= 0 || (nIndex = docu.indexOf("///")) >= 0 ) + { + if ( 0 != nIndex && + (docu.getStr()[nIndex - 1] != '"' && docu.getStr()[nIndex - 1] != ':') ) + idlc()->error()->syntaxError(PS_NoState, idlc()->getLineNumber(), + "nested documentation strings are not allowed!"); + } + idlc()->setDocumentation(docu); + BEGIN( INITIAL ); + } + +<DOCU>"*/" { + docu = docu.trim(); + sal_Int32 nIndex = 0; + int count = 0; + do { docu.getToken( 0, '\n', nIndex ); count++; } while( nIndex != -1 ); + idlc()->setLineNumber( beginLine + count - 1); + if ( docu.indexOf("/*") >= 0 || docu.indexOf("//") >= 0 ) + { + if ( 0 != nIndex && + (docu.getStr()[nIndex - 1] != '"' && docu.getStr()[nIndex - 1] != ':') ) + idlc()->error()->syntaxError(PS_NoState, idlc()->getLineNumber(), + "nested documentation strings are not allowed!"); + } + idlc()->setDocumentation(docu); + BEGIN( INITIAL ); + } + +"//"[^/]{1}.*"\n" { + /* only a comment */ + ::rtl::OString docStr(yytext); + docStr = docStr.copy( 0, docStr.lastIndexOf('\n') ); + docStr = docStr.copy( docStr.lastIndexOf('/')+1 ); + docStr = docStr.trim(); + idlc()->incLineNumber(); + } + +"///".*"\n" { + ::rtl::OString docStr(yytext); + docStr = docStr.copy( 0, docStr.lastIndexOf('\n') ); + docStr = docStr.copy( docStr.lastIndexOf('/')+1 ); + docStr = docStr.trim(); + idlc()->incLineNumber(); + idlc()->setDocumentation(docStr); + } + +. return yytext[0]; + +^#[ \t]*line[ \t]*[0-9]*" ""\""[^\"]*"\""\n { + parseLineAndFile(yytext); +} + +^#[ \t]*[0-9]*" ""\""[^\"]*"\""" "[0-9]*\n { + parseLineAndFile(yytext); +} + +^#[ \t]*[0-9]*" ""\""[^\"]*"\""\n { + parseLineAndFile(yytext); +} + +^#[ \t]*[0-9]*\n { + parseLineAndFile(yytext); +} + +^#[ \t]*ident.*\n { + /* ignore cpp ident */ + idlc()->incLineNumber(); +} + +^#[ \t]*pragma[ \t].*\n { /* remember pragma */ + idlParsePragma(yytext); + idlc()->incLineNumber(); +} + +%% |