/*
* File: so_split.c
* Date: 2004-02-21
*
* This program is used to split the Open/Star Office HTML file
* into seperate files at the H1 level and H2 for a specific
* section (Split_Section_Title).
*
* Copyright 2001,2002,2003,2004 David Eastcott <david@eastcott.net>
* Released under terms of GPL
*
*/
#include "stdio.h"
#include "string.h"
#include "stdlib.h"
#include "ctype.h"
#include "so_split.h"
// Local Global Data
static char Document_Title[TITLE_NAME_SIZE+2];
static char Target_Dir[FILE_NAME_SIZE+2];
static char next_line_buffer[BUFFER_SIZE+1];
//
static void add_end_file( FILE *pout )
{
fputs( "</BODY>\n", pout );
fputs( "</HTML>\n", pout );
}
//
static void add_end_leader( FILE *pout, SECTION_INFO *section_info )
{
char **html; // raw HTML
int length;
html = end_leader;
length = strlen ( Target_Dir );
while ( *html != (char *)NULL )
{
if ( (strcmp( *html, "PREV" )) == 0 )
{ // insert previous document reference if any
if ( section_info->prvs != NULL )
{
fputs( HREF_TAG, pout );
fputs( §ion_info->prvs->filename[length], pout );
fputs( HREF_TAG_END, pout );
fputs( "Prev", pout );
fputs( ANCHOR_TAG_END, pout );
fputs( TABLE_CELL_TAG_END, pout );
}
else
{
fputs( "Prev", pout );
fputs( TABLE_CELL_TAG_END, pout );
}
}
else if ( (strcmp( *html, "NEXT" )) == 0 )
{ // insert next document reference if any
if ( section_info->pfwd != NULL )
{
fputs( HREF_TAG, pout );
fputs( §ion_info->pfwd->filename[length], pout );
fputs( HREF_TAG_END, pout );
fputs( "Next", pout );
fputs( ANCHOR_TAG_END, pout );
fputs( TABLE_CELL_TAG_END, pout );
}
// else
// fputs( "Next</td>", pout );
}
else if ( (strcmp( *html, "PREVIOUS_SECTION" )) == 0 )
{ // insert previous document title
if ( section_info->prvs != NULL )
{
if( section_info->prvs->title[0] != 0 )
fputs( section_info->prvs->title, pout );
}
}
else if ( (strcmp( *html, "NEXT_SECTION" )) == 0 )
{ // insert next document title
if ( section_info->pfwd != NULL )
{
if( section_info->pfwd->title[0] != 0 )
fputs( section_info->pfwd->title, pout );
}
}
else
fputs( *html, pout );
// next string
html++;
};
}
//
static void add_head_file( FILE *pout )
{
FILE *pin;
char line_buffer[BUFFER_SIZE+1]; // working line buffer
char filename[FILE_NAME_SIZE+1];
strcpy( filename, Target_Dir );
strcat( filename, HEAD_FILE );
pin = fopen( filename, "r" );
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
fputs( line_buffer, pout );
};
// clean up
fclose ( pin );
}
//
static void add_head_leader( FILE *pout, SECTION_INFO *section_info )
{
char **html; // raw HTML
int length;
html = head_leader;
length = strlen ( Target_Dir );
while ( *html != (char *)NULL )
{
if ( (strcmp( *html, "TITLE" )) == 0 )
{ // insert document title
if( Document_Title[0] != 0 )
fputs( Document_Title, pout );
}
else if ( (strcmp( *html, "PREV" )) == 0 )
{ // insert previous document reference if any
if ( section_info->prvs != NULL )
{
fputs( HREF_TAG, pout );
fputs( §ion_info->prvs->filename[length], pout );
fputs( HREF_TAG_END, pout );
fputs( "Prev", pout );
fputs( ANCHOR_TAG_END, pout );
fputs( TABLE_CELL_TAG_END, pout );
}
else
{
fputs( "Prev", pout );
fputs( TABLE_CELL_TAG_END, pout );
}
}
else if ( (strcmp( *html, "NEXT" )) == 0 )
{ // insert next document reference if any
if ( section_info->pfwd != NULL )
{
fputs( HREF_TAG, pout );
fputs( §ion_info->pfwd->filename[length], pout );
fputs( HREF_TAG_END, pout );
fputs( "Next", pout );
fputs( ANCHOR_TAG_END, pout );
fputs( TABLE_CELL_TAG_END, pout );
}
// else
// fputs( "Next</td>", pout );
}
else
fputs( *html, pout );
// next string
html++;
};
}
//
static void add_to_queue ( SECTION_INFO **section_info_queue, SECTION_INFO *section_info )
{
SECTION_INFO *pnext;
// if queue is empty, put item at head
if( *section_info_queue == NULL )
{
*section_info_queue = section_info;
section_info->pfwd = NULL;
section_info->prvs = NULL;
}
else
{ // move down the chain till we get to last item
for( pnext = *section_info_queue;
pnext->pfwd != NULL;
pnext = pnext->pfwd )
;
// make new item as the last in the chain
pnext->pfwd = section_info;
// make sure the new item does not point to junk
section_info->pfwd = NULL;
section_info->prvs = pnext;
}
}
//
static void add_to_symbols( SYMBOL_TABLE *symbol, SYMBOL_QUEUE *symbol_queue )
{
if ( symbol_queue->head == NULL )
{ // insert symbol at head of queue
symbol_queue->head = symbol;
symbol_queue->tail = symbol;
symbol->pfwd = NULL;
}
else
{ // add new symbol to tail, updating fwd pntr of old tail
symbol_queue->tail->pfwd = symbol;
symbol_queue->tail = symbol;
symbol->pfwd = NULL;
}
}
//
static void adjust_href( char *line_buffer, SECTION_INFO *section_info, SYMBOL_QUEUE *symbol_queue )
{
char *position, *sposition;
char *fposition;
char buffer[TITLE_NAME_SIZE+2];
int length, fragment;
SYMBOL_TABLE *symbol;
// first locate the reference point
position = strstr( line_buffer, HREF_TAG );
position += strlen ( HREF_TAG );
// check to see if this is TOC reference
if ( *position == '\"' )
{ // this means we have to find the tag based on the name text
#ifdef DEBUG
printf ( "\nSearching for TOC Reference in >>%s<<", line_buffer );
#endif
// move to Text area
sposition = position + 2;
// get copy of the text string
length = strcspn( sposition, "<" );
strncpy( buffer, sposition, length );
buffer[length] = 0;
// locate the string in the available symbols
for( symbol = symbol_queue->head; symbol != NULL; symbol = symbol->pfwd )
{
if ( (strcmp ( symbol->name_text, buffer )) == 0 )
break; // we have a match
}
// on error, report and exit
if( symbol == NULL )
{
printf ( "Error: unable to locate symbol for %s\n", buffer );
return;
}
// make room for tag, and isert it with leading #
length = strlen ( position );
my_memmove( position+(strlen(symbol->name) + 1), position, length );
*position = '#';
strncpy ( position+1, symbol->name, strlen( symbol->name ) );
}
// exit now if href is not a relative type
if ( *position != '#' )
return;
//
// SO 6.0Beta Bug Fix
// if ( (strncmp ( position+1, "http:", 5)) == 0 ||
// (strncmp ( position+1, "ftp:", 4)) == 0 )
// return;
#ifdef DEBUG
printf ( "\nSearching for Relative Reference in >>%s<<", line_buffer );
#endif
// get copy of the tag string
length = strcspn( position, "\"" );
strncpy( buffer, position+1, length-1 );
buffer[length-1] = 0;
// locate the string in the available symbols
for( symbol = symbol_queue->head; symbol != NULL; symbol = symbol->pfwd )
{
if ( (strcmp ( symbol->name, buffer )) == 0 )
break; // we have a match
}
// on error, report and exit
if( symbol == NULL )
{
printf ( "Error: unable to locate symbol for %s\n", buffer );
return;
}
// if the symbols section is the same as the one we are in now, just leave it relative
if ( symbol->section_info == section_info )
return;
// make room in buffer for filename and insert filename ahead of tag
fposition = &symbol->section_info->filename[0] + strlen( Target_Dir );
fragment = strlen( position );
my_memmove( (position+(strlen ( fposition ))), position, fragment );
strncpy( position, fposition, (strlen(fposition)) );
}
//
static void build_symbol_table( char *line_buffer, SECTION_INFO **section_info_queue, SYMBOL_QUEUE *symbol_queue )
{
SECTION_INFO *section_info;
FILE *pin;
char *position;
int length;
SYMBOL_TABLE *symbol;
// go thru each section and extract Anchor symbols
for( section_info = *section_info_queue;
section_info != NULL;
section_info = section_info->pfwd )
{
// open file for reading
pin = fopen ( section_info->section_id_name, "r" );
// read lines from file
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
// keep reading till we find the start of the Anchnor
if ( (position = strstr( line_buffer, ANCHOR_TAG_NAME )) == NULL )
continue;
// tag found, create structure to hold data
symbol = calloc( 1, sizeof(SYMBOL_TABLE) );
// load structure
symbol->section_info = section_info;
// transfer anchor attribute
position += strlen( ANCHOR_TAG_NAME );
get_html_field( pin, NULL, line_buffer, position, symbol->name, ANCHOR_TAG_NAME_END );
// now the text
position = strstr( position, ANCHOR_TAG_END );
position += strlen ( ANCHOR_TAG_END );
get_html_field( pin, NULL, line_buffer, position, symbol->name_text, HEADER_TAG_END_OPEN );
// add it to the symbol table
add_to_symbols ( symbol, symbol_queue );
#ifdef DEBUG
printf( "Anch: %s, Text: %s, File: %s\n", symbol->name, symbol->name_text, symbol->section_info->filename );
#endif
}
// clean up
fclose ( pin );
}
}
//
static void get_head( FILE *pin, char *line_buffer )
{
FILE *pout;
char *position, **html;
int title_size;
char filename[FILE_NAME_SIZE+1];
strcpy( filename, Target_Dir );
strcat( filename, HEAD_FILE );
pout = fopen( filename, "w" );
// read lines from source until </HEAD> encountered,
// transfering them to the HEAD_FILE as we go
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
// extract document title
if( (position = strstr( line_buffer, TITLE_TAG )) )
{ // end of where we want
position += strlen( TITLE_TAG );
title_size = strcspn( position, "<" );
strncpy( Document_Title, position, (title_size < TITLE_NAME_SIZE) ? title_size : TITLE_NAME_SIZE );
}
if( (position = strstr( line_buffer, HEAD_TAG_END )) )
{ // end of where we want
// write last line and exit
fputs( line_buffer, pout );
break;
}
// write line
fputs( line_buffer, pout );
};
// insert basic head stuff
for( html = head_lines; *html != NULL; html++ )
fputs( *html, pout );
// clean up
fclose( pout );
}
//
static void get_html_field( FILE *pin, FILE *pout, char *line_buffer, char *position, char *token_buffer, char *end_tag )
{
char *tposition;
char end[2];
int length;
// preset for null token
*token_buffer = 0;
end[0] = *end_tag;
end[1] = 0;
// see if the end tag is on this line
if ( (tposition = strstr (position, end_tag )) != NULL )
{ // YES, transfer to buffer and mark end with null
position = move_past_tokens( position );
length = strcspn( position, end ); // find length of text
strncpy( token_buffer, position, length );
token_buffer[length] = 0;
}
else
{ // Nope, hard times now
position = move_past_tokens( position );
strcpy( token_buffer, position ); // get first part
strip_tail_white_space( token_buffer );
// add a SPACE because the text crossed line boundaries
if ( (strlen ( token_buffer )) > 0 )
strcat ( token_buffer, " " );
// write out current line
if ( pout != NULL )
fputs( line_buffer, pout );
// get next line
fgets( line_buffer, BUFFER_SIZE, pin );
// append text up to end marker
position = move_past_tokens( line_buffer );
length = strcspn( position, end );
strncat ( token_buffer, position, length );
strip_tail_white_space( token_buffer );
}
}
//
static void get_section_title( char *line_buffer, SECTION_INFO *section_info, char *end_tag, FILE *pin, FILE *pout )
{
char *position, *tposition;
int length;
// move to the begining of the name
// <H1 STYLE="page-break-before: always"><A NAME="Introduction"></A>Introduction</H1> <-- if we're lucky
position = strstr( line_buffer, ANCHOR_TAG_END );
position += strlen ( ANCHOR_TAG_END );
get_html_field ( pin, pout, line_buffer, position, section_info->title, end_tag );
}
//
static void get_title_section( FILE *pin, char *line_buffer, SECTION_INFO **section_info_queue )
{
FILE *pout;
char *position;
SECTION_INFO *section_info;
// create section for the title page
section_info = calloc( 1, sizeof(SECTION_INFO) );
section_info->section_id = TITLE_SECTION_ID;
sprintf( section_info->section_id_name, "%s%d.tmp.html", Target_Dir, section_info->section_id );
strcpy ( section_info->filename, Target_Dir );
strcat ( section_info->filename, TITLE_FILE );
strcpy ( section_info->title, "Title" );
// add the new section to chain
add_to_queue( section_info_queue, section_info );
// open file for output
pout = fopen( section_info->section_id_name, "w" );
// read lines from source starting at current position until <DIV> encountered,
// transfering them to the temp Title file as we go
// seek to next line after <BODY>
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
if( (position = strstr( line_buffer, BODY_TAG_OPEN )) )
break;
};
// now transfer the stuff we want
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
// Stop when we get to the TOC stuff
if( (position = strstr( line_buffer, TOC_DIV_TAG )) )
break;
// write line
fputs( line_buffer, pout );
};
// clean up
fclose( pout );
}
//
static void get_toc_section( FILE *pin, char *line_buffer, SECTION_INFO **section_info_queue )
{
FILE *pout;
int iteration;
char *position, *tposition;
SECTION_INFO *section_info;
int offset, toc_error;
toc_error = FALSE;
// create section for the Table of Contents page
section_info = calloc( 1, sizeof(SECTION_INFO) );
section_info->section_id = TOC_SECTION_ID;
sprintf( section_info->section_id_name, "%s%d.tmp.html", Target_Dir, section_info->section_id );
strcpy ( section_info->filename, Target_Dir );
strcat ( section_info->filename, TOC_FILE );
strcpy ( section_info->title, "Table of Contents" );
// add the new section to chain
add_to_queue( section_info_queue, section_info );
// open file for output
pout = fopen( section_info->section_id_name, "w" );
// then the opening <DIV line that was picked up by get_title_section for us
fputs( line_buffer, pout );
// now read lines from source starting at current position until <DIV> encountered,
// transfering them to the temp TOC file as we go
iteration = 1;
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
// Watch out for the 2nd one because a title for TOC was entered
if( (strstr( line_buffer, TOC_DIV_TAG )) )
iteration += 1;
// stop reading when 2nd occurance of </DIV seen
if( (strstr( line_buffer, DIV_TAG_END_OPEN )) )
{
iteration -= 1;
// second time we're done
if ( iteration == 0 )
{
fputs( line_buffer, pout );
break;
}
}
// locate the lines which are the actual table of contents
// and strip page number, and add empty HREF for later
//
// this is VERY SO 6 beta specific
if ( iteration == 1 )
{
// locate line with desired text on it
if ( (strstr( line_buffer, PARA_TAG_OPEN )) != NULL )
{
// now see if we have all we need
if ( (position = strstr( line_buffer, PARA_TAG_END )) == NULL )
{ // nope - magic time
do
{
// get next line
fgets( next_line_buffer, BUFFER_SIZE, pin );
if ( (strlen ( line_buffer ) + strlen ( next_line_buffer )) >= BUFFER_SIZE )
{
printf ( "Error: Paragraph size too large in TOC\n" );
toc_error = TRUE;
break;
}
// first strip whitespace and EOL from previous line
position = &line_buffer[(strlen(line_buffer))-1];
*position-- = '\0';
while (position != line_buffer && isspace(*position) )
*position-- = '\0';
// now move past all whitespace at front of current line
tposition = next_line_buffer;
while ( isspace( *tposition ) ) tposition++;
if ( !(*tposition == '<' || *position == '>') )
{
*++position = ' ';
*++position = '\0';
}
strcat( position, tposition ); // merge the lines into one till we find the end of tag
#ifdef DEBUG1
printf ( "\nTOC BEG:\n%s\n:END", line_buffer );
#endif
} while ( (strstr ( next_line_buffer, PARA_TAG_END )) == NULL );
if ( toc_error )
{ // stop processing on a line error
toc_error = FALSE;
continue;
}
// now reset as though it was all one line
position = strstr( line_buffer, PARA_TAG_END );
}
// remove page number from line
tposition = position = move_past_token ( BACKWARD, position ); // get to page number
position++;
while ( isdigit( *tposition ) ) tposition--; // move past number
while ( isspace( *tposition ) ) tposition--; // move to end of actual entry
tposition++;
strcpy ( tposition, position );
// now add tail stuff before paragraph close
my_memmove ( tposition+(strlen(ANCHOR_TAG_END)), tposition, strlen( tposition ));
strncpy( tposition, ANCHOR_TAG_END, strlen( ANCHOR_TAG_END ) );
// now insert dummy HREF for later
tposition = move_past_token ( FORWARD, line_buffer ); // move to real text
my_memmove( tposition+(strlen(HREF_TAG)+(strlen(HREF_TAG_END))), tposition, (strlen (tposition)) );
strncpy ( tposition, HREF_TAG, strlen (HREF_TAG) ); // add lead stuff
strncpy ( tposition+strlen(HREF_TAG), HREF_TAG_END, strlen( HREF_TAG_END) );
}
}
// write line
fputs( line_buffer, pout );
};
// clean up
fclose( pout );
}
//
static void make_adjusted_html( char *line_buffer, SECTION_INFO **section_info_queue, SYMBOL_QUEUE *symbol_queue )
{
FILE *pin, *pout;
SECTION_INFO *section_info;
int toc_found, toc_count, toc_error;
char *position, *tposition, **html;
toc_found = FALSE;
toc_count = 0;
toc_error = FALSE;
// now convert all temp sections to final html files
for( section_info = *section_info_queue; section_info != NULL; section_info = section_info->pfwd )
{
pout = fopen( section_info->filename, "w" );
pin = fopen( section_info->section_id_name, "r" );
// transfer all body stuff now, adjust hrefs as we go
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
if( (strstr( line_buffer, TOC_DIV_TAG )) != NULL )
{
toc_found = TRUE;
toc_count += 1;
}
if ( toc_found == TRUE )
{
if( (strstr( line_buffer, DIV_TAG_END )) != NULL )
toc_count -= 1;
}
#ifdef DEBUG1
printf ( "\nF=%d, C=%d, >>%s<<\n", toc_found, toc_count, line_buffer );
#endif
if ( toc_found && toc_count == 1 )
{ // adjust TOC text line with a dummy href
// locate line with desired text on it
if ( (strstr( line_buffer, PARA_TAG_OPEN )) != NULL )
{
// now see if we have all we need
if ( (position = strstr( line_buffer, PARA_TAG_END )) == NULL )
{ // nope - magic time
do
{
// get next line
fgets( next_line_buffer, BUFFER_SIZE, pin );
if ( (strlen ( line_buffer ) + strlen ( next_line_buffer )) >= BUFFER_SIZE )
{
printf ( "Error: Paragraph size too large in TOC\n" );
toc_error = TRUE;
break;
}
// first strip whitespace and EOL from previous line
position = &line_buffer[(strlen(line_buffer))-1];
*position-- = '\0';
while (position != line_buffer && isspace(*position) )
*position-- = '\0';
// now move past all whitespace at front of current line
tposition = next_line_buffer;
while ( isspace( *tposition ) ) tposition++;
if ( !(*tposition == '<' || *position == '>') )
{
*++position = ' ';
*++position = '\0';
}
strcat( position, tposition ); // merge the lines into one till we find the end of tag
#ifdef DEBUG1
printf ( "\nHTML BEG:\n%s\n:END", line_buffer );
printf ( "\ntoc_found %d, toc_count %d\n", toc_found, toc_count );
#endif
} while ( (strstr( next_line_buffer, PARA_TAG_END )) == NULL );
if ( toc_error )
{ // stop processing line on error
toc_error = FALSE;
continue;
}
// now reset as though it was all one line
position = strstr( line_buffer, PARA_TAG_END );
}
// remove page number from line
tposition = position = move_past_token ( BACKWARD, position ); // get to page number
position++;
while ( isdigit( *tposition ) ) tposition--; // move past number
while ( isspace( *tposition ) ) tposition--; // move to end of actual entry
tposition++;
strcpy ( tposition, position );
// now add tail stuff before paragraph close
my_memmove ( tposition+(strlen(ANCHOR_TAG_END)), tposition, strlen( tposition ));
strncpy( tposition, ANCHOR_TAG_END, strlen( ANCHOR_TAG_END ) );
// now insert dummy HREF for later
tposition = move_past_token ( FORWARD, line_buffer ); // move to real text
my_memmove( tposition+(strlen(HREF_TAG)+(strlen(HREF_TAG_END))), tposition, (strlen (tposition)) );
strncpy ( tposition, HREF_TAG, strlen (HREF_TAG) ); // add lead stuff
strncpy ( tposition+strlen(HREF_TAG), HREF_TAG_END, strlen( HREF_TAG_END) );
}
}
// replace the <BODY tag with our own, so that we get the attributes too
if ( (strstr( line_buffer, BODY_TAG_OPEN )) != NULL )
{
for( html = head_lines; *html != NULL; html++ )
fputs( *html, pout );
continue;
}
// adjust all href so that they point to the correct files
if ( (strstr( line_buffer, HREF_TAG )) != NULL )
adjust_href( line_buffer, section_info, symbol_queue );
// write line
fputs( line_buffer, pout );
};
// close this one and remove temp
fclose( pin );
fclose( pout );
};
}
//
static void make_html( char *line_buffer, SECTION_INFO **section_info_queue, SYMBOL_QUEUE *symbol_queue )
{
FILE *pin, *pout;
SECTION_INFO *section_info;
// now convert all temp sections to final html files
for( section_info = *section_info_queue; section_info != NULL; section_info = section_info->pfwd )
{
pout = fopen( section_info->filename, "w" );
pin = fopen( section_info->section_id_name, "r" );
if( section_info->section_id == TITLE_SECTION_ID )
{ // process title page differnent than others
add_head_file( pout );
}
else
{
add_head_file( pout );
add_head_leader( pout, section_info );
}
// transfer all boady stuff now, adjust hrefs as we go
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
// adjust all href so that they point to the correct files
if ( (strstr( line_buffer, HREF_TAG )) != NULL )
adjust_href( line_buffer, section_info, symbol_queue );
// write line
fputs( line_buffer, pout );
};
// closing remarks
if ( section_info->section_id != TITLE_SECTION_ID )
add_end_leader( pout, section_info );
add_end_file( pout );
// close this one and remove temp
fclose( pin );
fclose( pout );
unlink( section_info->section_id_name );
}
}
//
static char *move_past_tokens( char *position )
{
// if the first character is not a token start, we're done
if ( *position != '<' )
return ( position );
else
{
// as long as there are characters to check, look for close of token
while ( *position != 0 )
{
// exit if end of token, but watch for second in a row
if ( *position == '>' && *(position+1) != '<' )
return ( ++position );
position++;
}
return ( position );
}
}
//
static char *move_past_token( int direction, char *buffer )
{
if ( direction == FORWARD )
{
while ( TRUE )
{
while ( isspace( *buffer ) ) buffer++; // skip any white space
if ( *buffer != '<' )
break; // no token start, so must be done
while ( *buffer != '>' ) buffer++;
buffer++; // move to next charc after token stop
};
}
else // must be BACKWARDS
{
while ( TRUE )
{
if ( *buffer == '<' )
buffer--; // back up if token start
while ( isspace( *buffer ) ) buffer--; // skip any white space
if ( *buffer != '>' )
break; // done, if this is not the end on another token
while ( *buffer != '<' ) buffer--; // move to start of token
};
}
}
//
static void my_memmove( char *dest, char *src, int count )
{
// adjust src and dest to end of strings
dest += count;
src += count;
count += 1; // to account for the null
while( count-- )
*dest-- = *src--;
}
//
static void split_source( FILE *pin, char *line_buffer, SECTION_INFO **section_info_queue )
{
int section_id, sub_section_id, name_len;
int enable_H2;
char *position, *last_position;
FILE *pout;
SECTION_INFO *section_info;
// make sure we are at the beginning of the source file
rewind( pin );
pout = NULL;
section_id = BODY_SECTION_ID;
sub_section_id = 0;
enable_H2 = FALSE;
// run through the source file and create temp section files as we go
// that contain the "body" of each <H1 tag
// target file names based on the <H1 tag's title
while( (fgets( line_buffer, BUFFER_SIZE, pin )) != NULL )
{
// stop scanning lines if end of BODY seen
if( (position = strstr( line_buffer, BODY_TAG_END )) != NULL )
{
// if there is an open file, do end processing
if ( pout != NULL )
{
fclose ( pout );
pout = NULL;
// set for next section id
section_id += 1;
}
// go get next line
continue;
}
// we have a line, lets see if its the start of a desired definition
if( (position = strstr( line_buffer, HEADER_1_TAG )) == line_buffer ||
(enable_H2 && ((position = strstr( line_buffer, HEADER_2_TAG )) == line_buffer)) )
{ // extract section title, and make it the section filename
// if there is an open file, do end processing
if ( pout != NULL )
{
fclose ( pout );
pout = NULL;
// set for next section id
if ( (position = strstr( line_buffer, HEADER_1_TAG )) == line_buffer )
{ // move to next section id and stop splitting at H2
section_id += 1;
enable_H2 = FALSE;
}
else
sub_section_id +=1;
}
// get and initialize a section block with the name
section_info = calloc( 1, sizeof(SECTION_INFO) );
section_info->section_id = section_id;
if ( enable_H2 )
{ // for H2, extend names to include sub-section ID
section_info->sub_section_id = sub_section_id;
sprintf( section_info->section_id_name, "%s%d-%d.tmp.html", Target_Dir,
section_info->section_id, section_info->sub_section_id );
sprintf( section_info->filename, "%ssection%d-%d.html", Target_Dir,
section_info->section_id-BODY_SECTION_ID+1, section_info->sub_section_id );
}
else
{ // H1, names only include section ID
sprintf( section_info->section_id_name, "%s%d.tmp.html", Target_Dir,
section_info->section_id );
sprintf( section_info->filename, "%ssection%d.html", Target_Dir,
section_info->section_id-BODY_SECTION_ID+1 );
}
// remove old file, if it exists
unlink ( section_info->filename );
// create and initialize the new file
pout = fopen( section_info->section_id_name, "w" );
// get section title, and put in structure
get_section_title( line_buffer, section_info,
(enable_H2 ? HEADER_2_TAG_END : HEADER_1_TAG_END), pin, pout );
// add the new section to chain
add_to_queue( section_info_queue, section_info );
// then check to see if this section is the one that we are to split at the H2 level
if( (strcmp( section_info->title, Split_Section_Title )) == 0 )
enable_H2 = TRUE;
#ifdef DEBUG
printf ( "Section %d = %s\n", section_info->section_id, section_info->filename );
printf( "Title: <%s> Look:<%s>\n\n", section_info->title, Split_Section_Title );
#endif
}
// output lines if we have skipped all the intro stuff
if ( pout != NULL )
fputs( line_buffer, pout );
};
// if there is an open files, do end processing
if ( pout != NULL )
fclose ( pout );
}
//
static void strip_tail_white_space( char *token_buffer )
{
char *end;
// get end of buffer
end = &token_buffer[(strlen(token_buffer))-1];
// moving backward, strip white space from end of line
while ( isspace( *end ) )
{
*end = 0;
if ( end == token_buffer )
break;
else
end--;
}
}
//
// Local Private Data
static SECTION_INFO *section_info_queue = NULL; // section structure chain origin
static SYMBOL_QUEUE symbol_queue = { NULL, NULL }; // symbol table chain
static char line_buffer[BUFFER_SIZE+1]; // working line buffer
static char filename[FILE_NAME_SIZE+1]; // temp filename buffer
static int split_main( int argc, char *argv[] )
{
FILE *pin;
// set up directory leader
if ( argc > 3 )
{
strncpy ( Target_Dir, argv[3], FILE_NAME_SIZE );
// make sure path has terminating '/'
if ( strlen ( Target_Dir ) > 1 )
{
if ( Target_Dir[(strlen(Target_Dir)-1)] != '/' )
strcat( Target_Dir, "/" );
}
}
else // use current directory for files
Target_Dir[0] = 0;
// open source file for reading
if ( !(pin = fopen ( argv[2], "r" )) )
{
printf ( "Error: Unable to open source file: %s\n", argv[2] );
return ( 1 );
}
// clear document title
memset ( Document_Title, 0, sizeof(Document_Title) );
// First, extract up to </HEAD> into a temp file
// picking up document title as we go
get_head ( pin, line_buffer );
// Then get the TITLE data into a seperate file
get_title_section ( pin, line_buffer, §ion_info_queue );
// Then extract the <DIV ... /DIV> since it contains the Table of contents
get_toc_section ( pin, line_buffer, §ion_info_queue );
// Next run thru the source file and create independant files
// which are delineated by the <H1 tag
split_source ( pin, line_buffer, §ion_info_queue );
// Next, build Anchor/Href symbol table
build_symbol_table( line_buffer, §ion_info_queue, &symbol_queue );
// Finally create file HTML files and add some beautifying code
// that simplifies navigating about
make_html ( line_buffer, §ion_info_queue, &symbol_queue );
// all done, close up shop
fclose ( pin );
strcpy ( filename, Target_Dir );
strcat ( filename, HEAD_FILE );
unlink ( filename );
return ( 0 );
}
static int toc_main( int argc, char *argv[] )
{
FILE *pin;
SECTION_INFO *section_info;
// set up directory leader
if ( argc > 3 )
{
strncpy ( Target_Dir, argv[3], FILE_NAME_SIZE );
// make sure path has terminating '/'
if ( strlen ( Target_Dir ) > 1 )
{
if ( Target_Dir[(strlen(Target_Dir)-1)] != '/' )
strcat( Target_Dir, "/" );
}
}
else // use current directory for files
Target_Dir[0] = 0;
// open source file for reading
if ( !(pin = fopen ( argv[2], "r" )) )
{
printf ( "Error: Unable to open source file: %s\n", argv[2] );
return ( 1 );
}
// close it for now
fclose ( pin );
// First, build Anchor/Href symbol table
// create section for the Table of Contents page
section_info = calloc( 1, sizeof(SECTION_INFO) );
section_info->section_id = TOC_SECTION_ID;
sprintf( section_info->section_id_name, "%s", argv[2] );
strcpy ( section_info->filename, Target_Dir );
strcat ( section_info->filename, "new." );
strcat ( section_info->filename, argv[2] );
strcpy ( section_info->title, "Table of Contents" );
// add the new section to chain
add_to_queue( §ion_info_queue, section_info );
// now build up symbol table
build_symbol_table( line_buffer, §ion_info_queue, &symbol_queue );
// Finally create file HTML files and add some beautifying code
// that simplifies navigating about
make_adjusted_html ( line_buffer, §ion_info_queue, &symbol_queue );
// all done, close up shop
return ( 0 );
}
int main( int argc, char *argv[] )
{
// Check that we have some parameters
if ( argc < 3 )
{
printf ( usage );
return ( 1 );
}
// see which mode wanted
if ( (strncmp( argv[1], "-t", 2 )) == 0 )
{ //wants to adjust TOC only
return ( toc_main( argc, argv ) );
}
else if ( (strncmp( argv[1], "-s", 2 )) == 0 )
{ // wants to split the file up
return ( split_main( argc, argv ) );
}
else
{ // report error and exit
printf ( usage );
return ( 1 );
}
}
