/** rtfn.c -- RTF (rich text format) file normalizer. * @author Eric Laroche * @version @(#)$Id: rtfn.c,v 1.7 1999/05/23 08:05:58 laroche Exp $ */ /* rtfn -- RTF (rich text format) file normalizer. * Copyright (C) 1998 Eric Laroche. * * This program is free software; * you can redistribute it and/or modify it. * * This program 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. */ static char const rcsid[] = "@(#)$Id: rtfn.c,v 1.7 1999/05/23 08:05:58 laroche Exp $"; #include #include #include #include #include #ifdef unix #include #include #else #include #endif static char const app[] = "rtfn"; static void quit( char const* fmt, ... ); static void normalizefile( char const* infile ); static void normalize( FILE* fpin, FILE* fpout ); /** main -- program entry point. */ int main( int argc, char** argv ) { int i; /* usage */ if ( argc > 1 && ( ! strcmp( argv[ 1 ], "-?" ) || argv[ 1 ][ 0 ] == '-' ) ) { printf( "Usage: %s [files]\n", app ); printf( "\n" ); printf( "%s normalizes RTF files (rich text format files).\n", app ); printf( "%s enables the reasonable use of line based file difference tools\n" "(diff, RCS, etc.).\n", app ); printf( "Output files will have the last character of the file name replaced\n" "by 'n' (or 'm' if they end on 'n').\n" ); printf( "\n" "This program is free software;\n" "you can redistribute it and/or modify it.\n" "\n" "This program is distributed in the hope that it will be useful,\n" "but WITHOUT ANY WARRANTY; without even the implied warranty of\n" "MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.\n" ); return 0; } for ( i = 1; i < argc; i++ ) normalizefile( argv[ i ] ); if ( i == 1 ) normalize( stdin, stdout ); return 0; } /** quit -- display an error message and terminate. */ static void quit( char const* fmt, ... ) { va_list p; fprintf( stderr, "%s: ", app ); va_start( p, fmt ); vfprintf( stderr, fmt, p ); va_end( p ); fprintf( stderr, "\n" ); fflush( stderr ); exit( 1 ); } /** normalizefile -- normalize a rtf file. */ static void normalizefile( char const* infile ) { char* outfile; FILE *fpin, *fpout; fpin = fopen( infile, "r" ); if ( fpin == NULL ) quit( "can't open %s: %s", infile, strerror( errno ) ); outfile = malloc( strlen( infile ) + sizeof ( "" ) ); if ( outfile == NULL ) quit( "out of memory: %s", strerror( errno ) ); strcpy( outfile, infile ); if ( strlen( outfile ) > 0 ) outfile[ strlen( outfile ) - 1 ] = (char)( outfile[ strlen( outfile ) - 1 ] == 'n' ? 'm' : 'n' ); if ( access( outfile, 00 ) != -1 ) quit( "output file %s already exists", outfile ); fpout= fopen( outfile, "w" ); if ( fpout == NULL ) quit( "can't write %s: %s", outfile, strerror( errno ) ); normalize( fpin, fpout ); if ( ferror( fpin ) ) quit( "%s: %s", infile, strerror( errno ) ); if ( ferror( fpout ) ) quit( "%s: %s", outfile, strerror( errno ) ); fclose( fpout ); free( outfile ); fclose( fpin ); } /** normalize -- normalize rtf file contents. */ static void normalize( FILE* fpin, FILE* fpout ) { int curr = EOF; int prev = EOF; int col = 0; int i; /* helper macros 1: * -quit if error on output */ #define get( ) ( getc( fpin ) ) #define put( c ) { if ( putc( ( c ), fpout ) == EOF ) return; col++; } /* helper macros 2: * -ignore input line breaks * -quit if end-of-file */ #define current( ) ( curr ) #define previous( ) ( prev ) #define freshline( ) { if ( col > 0 ) { newline( ); } } #define newline( ) { put( '\n' ); col = 0; } #define column( ) ( col ) #define save( ) { prev = current( ); } #ifdef unix #define skip( ) { while ( ( curr = get( ) ) == '\n' || curr == '\r' ) ; } #else #define skip( ) { while ( ( curr = get( ) ) == '\n' ) ; } #endif #define testeof( ) { if ( curr == EOF ) { freshline( ); return; } } #define advance( ) { save( ); skip( ); testeof( ); } /* initial character */ advance( ); /* copy magic header "{\rtfx\x"..} (e.g. "{\rtf1\ansi"..}) * that is needed without line breaks */ for ( i = 0; i < 8; i++ ) { put( current( ) ); advance( ); } /* body */ for ( ; ; ) { /* wrap at column 72 */ if ( column( ) >= 72 ) freshline( ); switch ( current( ) ) { /* check for rtf keywords (\[a-z][a-z]*) and control symbols (\[^a-z]) * -for simplicity and robustness we could have decided not to scan * for the end of rtf keywords * -we don't embed unknown control symbols for robustness (are there * control symbols even larger than 4 bytes?) */ case '\\' : /* note that we will lose this \ if it's the last in a file, * which is not the case in a correct rtf document */ advance( ); /* check for control symbols */ if ( ! isalpha( current( ) ) ) { switch ( current( ) ) { /* escaped \\, \{, \} */ case '\\' : case '{' : case '}' : /* \~ (non-breaking space), * \_ (non-breaking hyphen), * \- (not-required hyphen) */ case '~' : case '_' : case '-' : /* embed in text * note that this can enlarge the lines until column 73 */ put( previous( ) ); put( current( ) ); advance( ); break; /* \* */ case '*' : /* isolate on a line of its own */ freshline( ); put( previous( ) ); put( current( ) ); advance( ); freshline( ); break; /* \'xx (non-ascii character) */ case '\'' : /* embed in text * note that this can enlarge the lines until column 75 */ put( previous( ) ); put( current( ) ); advance( ); put( current( ) ); advance( ); put( current( ) ); advance( ); break; /* other control symbols */ default : /* put them on a new line to avoid breaking them * by line wrapping */ freshline( ); put( previous( ) ); /* at least one character is part of the control symbol */ put( current( ) ); advance( ); break; } } else { /* keyword */ freshline( ); put( previous( ) ); while ( ! isspace( current( ) ) && current( ) != '\\' && current( ) != '{' && current( ) != '}' ) { put( current( ) ); advance( ); } /* blank(s) */ while ( isspace( current( ) ) ) { put( current( ) ); advance( ); } freshline( ); } break; /* check for begin ({) and end (}) of block */ case '{' : case '}' : freshline( ); put( current( ) ); advance( ); freshline( ); break; default : put( current( ) ); advance( ); break; } } }