#!/usr/bin/perl -w

#
# Test code to read event database for Ted Wang
#
# S. Diggs: 2003.11.05
#

require "fix_date.pl";
require "html_creation.pl";

##print STDOUT "Enter database filename: ";
##chomp($input_filename = <STDIN>);
##print STDOUT "Input filename is: $input_filename\n\n";
#$input_filename = "2004.02.27_event.hist";

my $input_filename = "Current_Event_Hist.txt";
if (! ( -e $input_filename)){

  print "$input_filename cannot be found in the current directory!\n";
  exit(1);
}

print "Reading database file...\n";

open (INPUT, $input_filename) or 
	die "Cannot open $input_filename for reading";

#
#--> leave in line feeds (the 'NOTE' section of the database is multi-line
#--> only separated by the '$' character)
#	
@file_buffer = <INPUT>;
print STDOUT "Got ", ($#file_buffer + 1), " lines\n";
close(INPUT);	
	
#
#--> In the database file, the first line is field names, the last
#--> line is a string of $'s to signal the end.
#--> read in fields from 1st line
#--> Database is held in an array with each line as a separate element
#	
chomp($file_buffer[0]);
my @event_field = split (/\$/, $file_buffer[0]);
print STDOUT "Fields are: \n---\n";
$num_of_fields = 0;
foreach my $thing (@event_field)	{
	$num_of_fields++;
	print STDOUT $thing . "\n";
}
print STDOUT "\nThere are $num_of_fields fields in the events table\n";

#
#--> In the database file, the first line is field names, the last 
#--> line is a string of $'s to signal the end.
#--> Keep the first line for completeness. Remove the last line.
#
pop(@file_buffer);

#
#--> Database is in the format data$data$data\ndata$data$data. 
#--> If it is parsed by $, the third element will be data\ndata.
#--> Cannot parse by \n, since they are allowed in the NOTES 
#--> section. 
#
#--> My workaround (not necessarily the best) is to change the format to
#--> data$data$data\n$data$data$data while it is still in the file_buffer 
#--> array.  First see if a line is a new entry.  Lines that are new 
#--> entries have dollar signs.  Lines  within notes do not. If it is 
#--> a new entry, insert a dollar sign at the beginning.  Do this for all 
#--> lines except for the first line in the file. 
#
#--> Line by line, the original database looks like this:
#--> line 1 (new entry):    data$data$data$notes\n 
#--> line 2 (new entry):    data$data$data$notes\n 
#--> line 3 (continuation): notes\n
#
#--> Line by line, the adjusted database looks like this:
#--> line 1 (new entry):    data$data$data$notes\n 
#--> line 2 (new entry):    $data$data$data$notes\n 
#--> line 3 (continuation): notes\n
#
#--> NOTE: this will work as long as newline chars 
#--> are allowed only in the last field.
#
$num_of_entries = 0;

##open(OUTPUT_FILE, ">fixtest");
for ($i=1; $i <= $#file_buffer; $i++) { 
	if ($file_buffer[$i] =~ /\$/) { 
		$num_of_entries++;
		$file_buffer[$i] = "\$$file_buffer[$i]";
##		print OUTPUT_FILE "$file_buffer[$i]";
	}
}
##close(OUTPUT_FILE);
#
#--> There are $num_of_entries entries plus 1 line with the fields
#
print "There are $num_of_entries entries\n";

#
#--> Join the file input buffer back together so that you
#--> can parse the multi-line 'NOTES' section at the end 
#--> Database is now held in one long string 
#
$input_buffer_strung_together = join('', @file_buffer);

#
#--> Re-parse the scalar "$input_buffer_strung_together" by "$"
#--> Database is now held in an array, each element is text 
#--> between the dollar signs in the original file
#
@reparse_input_buffer = split(/\$/, $input_buffer_strung_together);

#
#--> Now, print out the array elements, one at a time until you're 
#--> satisfied that it worked
#
#my $dummy = <STDIN>;
#for ($i=0 ; $i <= $#reparse_input_buffer; $i++)	{
##for ($i=0 ; $i <= 1000; $i++)	{
#for ($i=60021 ; $i <= 60121; $i++)	{
#for ($i=25000 ; $i <= 25100; $i++)	{

##	if ($i % 13 == 0) {
##	printf STDOUT ("%5d\t --> %s\n",$i, $reparse_input_buffer[$i]);
##	}
##}	

#
#--> Make a "virtual" database with a hash of arrays.
#-->            |0        |1        |2        |3|4|...|num_of_entries|
#--> |line | -> |line     |a02      |a03      |3|4|...|num_of_entries|
#--> |date | -> |date     |10/27/03 |02/28/94 |3|4|...|num_of_entries|
#-->    .
#-->    .
#-->    .
#--> |notes| -> |notes    |fixed    |deleted  |3|4|...|num_of_entries|
#
for ($i=0 ; $i <= $num_of_entries ; $i++)	{
	my @temp_array = splice(@reparse_input_buffer, 0, $num_of_fields);
	$field=0;
	foreach my $array_element ( @event_field )	{
	#
	#--> set hash_element[i] = the line it's found on (i)
	#--> and the name of the proper field column
	#
	#	print "$i||$array_element||$temp_array[$field]\n";
		$data_hash{$array_element}[$i] = $temp_array[$field]; 
		$field++;
	}
}

print "\nData hash created (PRESS ENTER TO CONTINUE): ";
my $dummy = <STDIN>;

#
#--> Check to see if the database was parsed correctly, if not, there may be 
#--> misplaced entries.  One sign of misplaced entries is newlines in 
#--> fields other than notes.  This will result in a num_of_entries count to 
#--> not equal the actual number of entries as given by th 4D database.
#

#
#--> Change the date formats from (M or MM)/(D or DD)/YY to YYYY/MM/DD using
#--> fix_date.pl
#
#--> Change newlines in Notes section to <br> for html pages
#
for ($i=1 ; $i <= $num_of_entries ; $i++) {
    $data_hash{'Cruise Date'}[$i] = 
        &fix_date($data_hash{'Cruise Date'}[$i]);
    $data_hash{'Date Entered'}[$i] = 
        &fix_date($data_hash{'Date Entered'}[$i]);

    $data_hash{'Note'}[$i]=~ s/\n/<br>\n/g;

}

#
#--> Allow user to enter cmd line arguments to check data_hash. 
#--> Usage:
#--> -e 06MT22_5  (-e followed by expocode for all entries with that code)
#--> -i 456 (-i followed by index number for entry with that index number)
#
if ( $#ARGV > 0) {
	if ( grep(/-e/, $ARGV[0])) {
	for ($i=0; $i<=$num_of_entries; $i++) {
		if (grep(/$ARGV[1]/, $data_hash{'ExpoCode'}[$i])){
		foreach $array_element (@event_field) {
                print "$i: $array_element: $data_hash{$array_element}[$i]\n";
                }
 
		}	
	}	
	}
	if (grep(/-i/, $ARGV[0]) && $ARGV[1] =~ /[0-9]+/ 
	&& $ARGV[1] <= $num_of_entries && $ARGV[1] >= 0) {
		print "Entry $ARGV[1]:\n---------------\n"; 
		foreach $array_element (@event_field) {
		print "$array_element: $data_hash{$array_element}[$ARGV[1]]\n";
		}	
	}
}

#
#--> Open the data file with the list of all the expos and directories.
#--> Use this info to search for entries and create their HTML pages.
#--> This data file is $ delimited with 1 entry per line.
#

#$list_of_expos="/home/whpo/sdiggs/HTML/cgi/PERL/MAKE_TABLES/list_of_expos";
$list_of_expos="list_of_expos";
open(LIST, $list_of_expos) or
        print STDOUT "Cannot open datafile ---> $list_of_expos\n";
@expo_buffer = <LIST>;
chomp(@expo_buffer);
close(LIST);
print "Reading list of expos file...\n";

#
#--> Make a hash of arrays for this file we just opened.
#-->            |0    |1    |2    | ... |n    |
#--> |line| ->  |a01  |a02  |a02  | ... |s04  |
#--> |dir | ->  |dir1 |dir2 |dir2 | ... |dir9 |
#--> |expo| ->  |121  |122  |123  | ... |9999 |
#
@expo_header=( "Line", "Directory", "Expocode" );
for ($c=0; $c<=$#expo_buffer ; $c++) {
        @expo_tmp = split(/\$/,$expo_buffer[$c]);
        $expo_field=0;
        foreach $name (@expo_header) {
	#	print "$c||$name||$expo_tmp[$expo_field]\n";
                $expo_hash{$name}[$c] = $expo_tmp[$expo_field];
                $expo_field++;
        }
}

#
#--> Some adjacent entries have the same directory.  Collapse these entries 
#--> into one entry.  Separate the expocodes with "and".  This will be useful 
#--> in creating the HTML pages.  Do not remove the entries with the same 
#--> directory.  Instead, add a field to the hash to indicate which 
#--> entries to use when creating the HTML pages.
#-->            |0    |1     |2            | ... |n    |
#--> |line| ->  |a01  |a02   |a02          | ... |s04  |
#--> |dir | ->  |dir1 |dir2  |dir2         | ... |dir9 |
#--> |expo| ->  |121  |122   |122 and 123  | ... |9999 |
#--> |use | ->  |true |false |true         | ... |true |

$expo_hash{'Use'}[0] = "true";
for ($i=1 ; $i <= $#expo_buffer ; $i++ ) {
	$expo_hash{'Use'}[$i] = "true";
	if(grep(/$expo_hash{'Directory'}[$i]/, $expo_hash{'Directory'}[$i-1])){
		$expo_hash{'Expocode'}[$i] = 	
		"$expo_hash{'Expocode'}[$i-1] and $expo_hash{'Expocode'}[$i]";
		$expo_hash{'Use'}[$i-1] = "false";
	}
}
# count how many directories are used
#$numfiles = 0;
#for ($i=0; $i<= $#expo_buffer; $i++) {
#    if (grep(/$expo_hash{'Use'}[$i]/, "true"))
#    {$numfiles++;}
#}
#print $numfiles;

print "Expo hash created (PRESS ENTER TO CONTINUE): ";
my $dummy2 = <STDIN>;

#
#--> For each directory, use its expocodes to search for the corresponding 
#--> expocodes in the data hash.  Save the index numbers where the expocodes 
#--> match in an array.  Then pass directory, expocode(s), and matching array
#--> to HTML creation script.
#

#in list of expos.. some codes are in the format "blah_1" while in the database
#they are just "blah" causing entries not to show up.

#for($i=36; $i<=36 ; $i++){
$numfiles = 0;
for($i=0; $i<=$#expo_buffer ; $i++){
if(grep(/$expo_hash{'Use'}[$i]/, "true")) { 
        $numfiles++;########
#	print "$i||$expo_hash{'Line'}[$i]||$expo_hash{'Directory'}[$i]||$expo_hash{'Expocode'}[$i]||$expo_hash{'Use'}[$i]\n"; 
	@expocodes = split(/ and /, $expo_hash{'Expocode'}[$i]);
        @matches = ();
        foreach $thing (@expocodes) {
	for($j=0; $j<=$num_of_entries ; $j++){
		if(grep(/^$thing$/, $data_hash{'ExpoCode'}[$j])){
#			print"$j\n";
			# add $j to array of results
			push(@matches, $j);	
		}
	}
	}
#       print "@matches\n";
        *arr=\@matches;
	#call HTML creation script
        print $numfiles;#########
        &html_creation($expo_hash{'Line'}[$i], $expo_hash{'Directory'}[$i], $expo_hash{'Expocode'}[$i], *arr);
}
}