#!/usr/bin/perl
use Pg;
######################
#build indices
#####################
#1.build array with all the fields to get values in
#2. extract unique values from each field
#3. return a unique array for each field
#4. build a linked index page, call a 'buildtrajanpage?key' for each value
##NB decide how to send several params after the ?
@ind=qw(subjectkeywords toolkeywords sideofcolumn scenenumber drumnumber);
foreach $i(@ind) {
print "processing $i\n";
$fld=$i;
&GETUNIQUE;
}
#get header and footer
$header=`cat stndheader.txt`;
$footer=`cat stndfooter.txt`;
#print total num of matches
$S_sk=@sk;
print "$S_sk unique matches for subjectkeywords \n";
$S_tk=@tk;
print "$S_tk unique matches for toolkeywords \n";
$S_sc=@sc;
print "$S_sc unique matches for sideofcolumn \n";
$S_sn=@sn;
print "$S_sn unique matches for scenenumber \n";
$S_dn=@dn;
print "$S_dn unique matches for drumnumber \n";
$S_s=@s;
print "$S_s unique matches for spiral \n";
#############################################
##BUILD INDICES
#1. subjectkeywords
print "building subject keyword index...\n";
#$skout="Index organized by Subject Keyword\n";
$c=1;
#@sortedsk= sort @sk;
#foreach $elem(@sortedsk){
foreach $elem(@sk){
if (!($c==8)){$p="";}else{$p="";$c=1;}
$skout.= " $elem$p\n";
$c++;
}
$c=1;
#2. toolkeywords
print "building tool keyword index...\n";
#$tkout="Index organized by Tool Keyword\n";
#@sortedtk= sort @tk;
#foreach $elem(@sortedtk){
foreach $elem(@tk){
if (!($c==6)){$p="";}else{$p="
";$c=1;}
$tkout.= " $elem$p\n";
$c++
}
$c=1;
#3. sideofcolumn
print "building side of column index...\n";
#$scout="Index organized by Side Of Column\n";
#@sortedsc= sort {$a <=> $b } @sc;
#foreach $elem(@sortedsc){
foreach $elem(@sc){
if (!($c==12)){$p="";}else{$p="
";$c=1;}
$scout.= " $elem$p\n";
$c++;
}
$c=1;
#4. scenenumber
print "building scene number index...\n";
#$snout="Index organized by Scene Number\n";
#@sortedsn= sort {$a <=> $b } @sn;
#foreach $elem(@sortedsn){
foreach $elem(@sn){
if (!($c==12)){$p="";}else{$p="
";$c=1;}
$snout.= " $elem$p\n";
$c++;
}
$c=1;
#5. drumnumber
print "building drum number index...\n";
#$dnout="Index organized by Drum Number\n";
#@sorteddn= sort {$a <=> $b } @dn;
#foreach $elem(@sorteddn){
foreach $elem(@dn){
if (!($c==10)){$p="";}else{$p="
";$c=1;}
$dnout.= " $elem$p\n";
$c++;
}
$c=1;
#6. spiral
print "building spiral index...\n";
#$sout="Index organized by Spiral Number\n";
#@sorteds= sort {$a <=> $b } @s;
#foreach $elem(@sorteds){
foreach $elem(@s){
if (!($c==10)){$p="";}else{$p="
";$c=1;}
$sout.= " $elem$p\n";
$c++;
}
#foreach $i(0..@s) {
# print "Spiral=$s[$i] Drum=$dn[$i]\n";
#}
#open a file and write the data to it
open (F,">indices.html");
print F <\n
\n
Trajan's Column - Index\n
\n
\n
\n
$header
\n
\n
\n
| Index organized by Subject Keyword |
\n
| $skout |
\n
\n
| Index organized by Tool Keyword\n
|
| $tkout |
\n
\n
| Index organized by Side Of Column\n
|
| $scout |
\n
\n
| Index organized by Scene Number\n
|
| $snout |
\n
\n
| Index organized by Drum Number\n
|
| $dnout |
\n
\n
| Index organized by Spiral Number\n
|
| $sout |
\n
\n
\n
\n
$footer
\n
eof5
close (F);
#########################################
sub GETUNIQUE {
#build a hash, check the hash and then launch the appropriate sub handler
#in that handler call PG
%FIELDS = (
'subjectkeywords' => \&FILTER_SK,
'toolkeywords' => \&FILTER_TK,
'scenenumber' => \&FILTER_SN,
'sideofcolumn' => \&FILTER_SC,
'drumnumber' => \&FILTER_DN
);
#check the hash, the block checks for the key letter and fires off the sub assciated with it.
if ($FIELDS{$fld}) {$FIELDS{$fld}->();}
}
#SUBJECT KEYWORDS
sub FILTER_SK {
&CALLPG;
#get the data from the array
for $i ( 0 .. $#ary ) {
for $j ( 0 .. $#{$ary[$i]} ) {
$test= "$ary[$i][$j]";
#sometimes there is more than one word per field with commas
$test=~s/ //g;
#$test=~s/_//g;
if ($test && $test=~s/,/,/){
@testary=split ",",$test;
foreach $elem(@testary){
if ($test && !($uniqueSK=~s/$elem/$elem/)){ $uniqueSK.="$elem,";}
}
} elsif ($test && !($uniqueSK=~s/$elem/$elem/)){ $uniqueSK.="$elem,"; }
}
}
#now set the value we found to an array to check later
#trim the trailing comma
$uniqueSK=~s/,$//;
@sk = split ",", $uniqueSK;
}
#TOOL KEYWORDS
sub FILTER_TK {
&CALLPG;
#get the data from the array
for $i ( 0 .. $#ary ) {
for $j ( 0 .. $#{$ary[$i]} ) {
$test= "$ary[$i][$j]";
#sometimes there is more than one word per field with commas
$test=~s/ //g;
#$test=~s/_//g;
if ($test && $test=~s/,/,/){
@testary=split ",",$test;
foreach $elem(@testary){
if ($test && !($uniqueTK=~s/$elem/$elem/)){
$uniqueTK.="$elem,";
print "$elem\n";
}
}
} elsif ($test && !($uniqueTK=~s/$elem/$elem/)){ $uniqueTK.="$elem,"; }
}
}
#now set the value we found to an array to check later
#trim the trailing comma
$uniqueTK=~s/,$//;
@tk = split ",", $uniqueTK;
}
#SCENE NUMBER, This may need some tlc to make sure to get all the numbers out
sub FILTER_SN {
&CALLPG;
#get the data from the array
for $i ( 0 .. $#ary ) {
for $j ( 0 .. $#{$ary[$i]} ) {
$test= "$ary[$i][$j]";
#sometimes there is more than one word per field with commas
$test=~s/ //g;
$test=~s/_//g;
if ($test && ($test=~s/,/,/)){
@testcom=split ",", $test;
foreach $elem(@testcom) {
if (length $elem ==1 && !($single=~s/$elem/$elem/)){
$single.="$elem,";
}elsif (length $elem == 2 && !($double=~s/$elem/$elem/)){
$double.="$elem,";
}elsif (length $elem == 3 && !($tripple=~s/$elem/$elem/)){
$tripple.="$elem,";
}elsif (length $elem > 3 && !($other=~s/$elem/$elem/)){
$other.="$elem,";
}
}
}else {
if ($test && (length $test ==1) && !($single=~s/$test/$test/)){
$single.="$test,";
}elsif ( $test && (length $test == 2) && !($double=~s/$test/$test/)){
$double.="$test,";
}elsif ( $test && (length $test == 3) && !($tripple=~s/$test/$test/)){
$tripple.="$test,";
}elsif ($test && (length $test > 3) && !($other=~s/$test/$test/)){
$other.="$test,";
}
}
}
}
$uniqueSN.="$single$double$tripple$other";
#now set the value we found to an array to check later
#trim the trailing comma
$uniqueSN=~s/,$//;
@sn = split ",", $uniqueSN;
}
#SIDE OF COLUMN
sub FILTER_SC {
#no commas here
&CALLPG;
#get the data from the array
for $i ( 0 .. $#ary ) {
for $j ( 0 .. $#{$ary[$i]} ) {
$test= "$ary[$i][$j]";
#sometimes there is more than one word per field with commas
$test=~s/ //g;
$test=~s/_//g;
if($test && !($uniqueSC=~s/$test/$test/)){ $uniqueSC.="$test,";}
}
}
#now set the value we found to an array to check later
#trim the trailing comma
$uniqueSC=~s/,$//;
@sc = split ",", $uniqueSC;
}
#drum number
#DRUM NUMBER, This may need some tlc to make sure to get all the numbers out
sub FILTER_DN {
&CALLPG;
#get the data from the array
for $i ( 0 .. $#ary ) {
for $j ( 0 .. $#{$ary[$i]} ) {
$test= "$ary[$i][$j]";
#sometimes there is more than one word per field with commas
$test=~s/ //g;
$test=~s/_//g;
#separate spirals from drumnum
@both=split ",", $test;
$spirals.="$both[0],";
$drums.="$both[1],";
#print "SN=$both[0] : DN=$both[1]\n\n";
}
}
#now we have separated spirals and drums
$spirals=~s/,$//;
@spiralary= split ",", $spirals;
$drums=~s/,$//;
@drumary=split "," ,$drums;
#show count
$ts=@spiralary;
print "\n$ts spirals\n";
$ts=@drumary;
print "\n$ts drums\n";
#check for hyphenated entries
foreach $test (@spiralary) {
if ($test){
if (length $test ==1 && !($singleSN=~s/$test/$test/)){
$singleSN.="$test,";
}elsif (length $test == 2 && !($doubleSN=~s/$test/$test/)){
$doubleSN.="$test,";
}elsif (length $test == 3 && !($trippleSN=~s/$test/$test/)){
$trippleSN.="$test,";
}elsif ((length $test > 3) && !($otherSN=~s/$test/$test/)){
$otherSN.="$test,";
}
}
}
$uniqueS="";
$uniqueS.="$singleSN$doubleSN$trippleSN$otherSN";
$uniqueS=~s/,$//;
@s = split ",", $uniqueS;
print "sprial unique array=$uniqueS\n";
#Drum numbers
#check for hyphenated entries
foreach $test (@drumary) {
if ($test){
if (length $test ==1 && !($singleDN=~s/$test/$test/)){
$singleDN.="$test,";
}elsif (length $test == 2 && !($doubleDN=~s/$test/$test/)){
$doubleDN.="$test,";
}elsif (length $test == 3 && !($trippleDN=~s/$test/$test/)){
$trippleDN.="$test,";
}elsif ((length $test > 3) && !($otherDN=~s/$test/$test/)){
$otherDN.="$test,";
}
}
}
$uniqueDN="";
$uniqueDN.="$singleDN$doubleDN$trippleDN$otherDN";
#now set the value we found to an array to check later
#trim the trailing comma
$uniqueDN=~s/,$//;
@dn = split ",", $uniqueDN;
#print "unique dn array=$otherDN\n";
}
##############
sub old_FILTER_DN {
#NB there are two pieces here, 1 spiral, 2 drum, just get drum here
&CALLPG;
#get the data from the array
for $i ( 0 .. $#ary ) {
for $j ( 0 .. $#{$ary[$i]} ) {
$test= "$ary[$i][$j]";
#sometimes there is more than one word per field with commas
$test=~s/ //g;
$test=~s/_//g;
@testary=split ",",$test;
if ($test && !($uniqueDN=~s/$testary[1]/$testary[1]/)){ $uniqueDN.="$testary[1],";}
#while we are here get spiral info too
#check to see if it is <9 first
if ($test && ($testary[0] < 9) && !($spiralA=~s/$testary[0]/$testary[0]/)){
$spiralA.="$testary[0],";
}elsif ($test && !($spiral=~s/$testary[0]/$testary[0]/)){
$spiral.="$testary[0],";
}
}
}
#print "$uniqueDN\n";
#now set the value we found to an array to check later
#trim the trailing comma
#spirals
$uniqueS="$spiralA$spiral";
$uniqueS=~s/,$//;
@s=split ",", $uniqueS;
#drumnumbers
$uniqueDN=~s/,$//;
@dn = split ",", $uniqueDN;
}
sub FILTER_KEYWORDS {
#sometimes there is more than one word per field with commas
$test=~s/ //g;
$test=~s/_//g;
if ($test && $test=~s/,/,/){
@testary=split ",",$test;
foreach $elem(@testary){
#print $elem,"\n";
if (!($unique=~s/$elem/$elem/)){
$unique.="$elem,";
# print "mult unique -$elem\n";
}
}
}elsif ($test && !($unique=~s/$elem/$elem/)){
$unique.="$elem,";
#print "unique $elem\n";
}
}
sub CALLPG {
#CALL PG AND GET LIST OF A FIELD
$sql ="SELECT $fld from traj";
$conn = Pg::connectdb("dbname=trajan");
#execute the query
Pg::doQuery($conn, $sql, \@ary);
}
exit;