#!/usr/bin/perl

#(9/27/2023): Accepts an SRT captioning file and runs it through Bison-32K

use JSON::XS;

#./ INFILE OUTFILE FORMAT(SRT/TXT) PROMPTNUM
if ($ARGV[0] eq '' || $ARGV[1] eq '' || $ARGV[2] eq '' || $ARGV[3] eq '' || (-s $ARGV[0]) < 50 ) { print "FATAL: Invalid Parameters...\n"; exit; }

##################################################################
#LOAD TRANSCRIPT...

if ($ARGV[2] eq 'SRT') {

    #load from SRT...
    open(FILE, $ARGV[0]); binmode(FILE, ":utf8");
    while(<FILE>) {
        $_=~s/[\x{FEFF}\x{FFFF}]/ /g; #first transcode a few characters...
        $_=~s/\s+$//; $_=~s/^\s+//; #remove leading/trailing spaces...
    
        #skip formatting lines...
        if ($_=~/^\d+$/ || $_ eq '') { next; };
        if ($_=~/^[\d:,\-\> ]+$/) { next; };
        #remove speaker notations...
        $_=~s/^>>\s*//;    
        #$mid = ord(substr($_,0,1)); #codify oddities...
    
        $TRANSCRIPT .= $_ . ' ';
    
    }
    close(FILE);

    #clean up extra spaces...
    $TRANSCRIPT=~s/\s+/ /gs;

} else {

    #load from TXT...
    open(FILE, $ARGV[0]); binmode(FILE, ":utf8"); read(FILE, $TRANSCRIPT, (-s FILE)); close(FILE);

}
##################################################################

##################################################################
#CONSTRUC, SUBMIT AND RECEIVE OUR RESULTS...

#$TRANSCRIPT = 'Welcome to tonights broadcast. Our first story is that Donald Trump announced he was running for president again. Joe Biden is traveling to China next week to discuss Ukraine.';

#add our prompt...
if ($ARGV[3] eq 'SUMMARIZE') {
    $TRANSCRIPT = "Divide the television news transcript below into a set of discrete stories, make them into a bulleted list with a title and short description and summary of each in English. Each story should include a Title, Description and Summary and be written in English. Do not use any information beyond what is in the transcript below. \n\nTELEVISION NEWS TRANSCRIPT:\n" . $TRANSCRIPT;
} elsif ($ARGV[3] eq 'SUMMARIZESUMMARIES') {
    $TRANSCRIPT = "Below is a collection of television news summaries describing a day's news on a television news channel. Take all of the summaries below, combine the ones that describe the same story and divide the final summary into a set of discrete stories, make them into a bulleted list with a title and short description and summary of each in English. Each story should include a Title, Description and Summary and be written in English. Do not use any information beyond what is in the summaries below. \n\nTELEVISION NEWS SUMMARIES:\n" . $TRANSCRIPT;
}

#construct the submission...
$JSON->{'instances'}[0]->{'prompt'} = $TRANSCRIPT;
$JSON->{'parameters'}->{'temperature'} = 0.2;
$JSON->{'parameters'}->{'maxOutputTokens'} = 8192;
#$JSON->{'parameters'}->{'topK'} = 40;
#$JSON->{'parameters'}->{'topP'} = 0.95;
open(OUT, ">$ARGV[1].toapi"); print OUT JSON::XS->new->allow_nonref(1)->utf8->encode( $JSON ); close(OUT);

#submit it...
unlink("$ARGV[1].fromapi");
open(LOG, ">>SUBMIT.LOG"); print LOG localtime(time) . ": curl -s -X POST -H \"Authorization: Bearer \$(gcloud auth print-access-token)\" -H \"Content-Type: application/json\" https://us-central1-aiplatform.googleapis.com/v1/projects/[YOURPROJECTID]/locations/us-central1/publishers/google/models/text-bison-32k:predict -d \@$ARGV[1].toapi > $ARGV[1].fromapi\n"; close(LOG);
system("curl -s -X POST -H \"Authorization: Bearer \$(gcloud auth print-access-token)\" -H \"Content-Type: application/json\" https://us-central1-aiplatform.googleapis.com/v1/projects/[YOURPROJECTID]/locations/us-central1/publishers/google/models/text-bison-32k:predict -d \@$ARGV[1].toapi > $ARGV[1].fromapi");

#and get the results...
system("cat $ARGV[1].fromapi | jq -r .predictions[0].content > $ARGV[1]");


##################################################################