rename_ambetter_eob

#!/usr/bin/env perl

# Member ID to initial$ mapping
my %MEMBER_TO_INITIAL = (
   'U9728842101' => 'M',
   'U9728842102' => 'F',
   # Add more mappings as needed
);

use strict;
use warnings;
use autodie;
use File::Basename;
use Data::Dumper;
use Getopt::Long;

# Parse command line options
my $debug = 0;  # Default to debug off
GetOptions("debug|d" => \$debug) or die "Usage: $0 [-d] pdf-file ...\n";

# Define the base pattern
my $DATE_PATTERN_STR = '\d{1,2}/\d{1,2}/\d{4}';
my $DATE_PATTERN = qr/$DATE_PATTERN_STR/;
my $PDFGREP_PATTERN = "(Provider Name:|$DATE_PATTERN_STR)";

sub normalize_provider {
   my ($provider) = @_;
   warn "DEBUG: normalizing provider: '$provider'\n" if $debug;
   # Strip after comma and normalize whitespace
   $provider =~ s/,.*$//;
   $provider =~ s/\s+/ /g;
   $provider =~ s/^\s+|\s+$//g;

   my @words = split /\s+/, $provider;

   # If second word is AND/OF, take three words
   if ($#words >= 2 && $words[1] =~ /^(AND|OF)$/) {
       $provider = join '-', @words[0..2];
   } else {
       # Otherwise take up to two words
       $provider = join '-', @words[0..($#words > 0 ? 1 : 0)];
   }

   warn "DEBUG: normalized to: '$provider'\n" if $debug;
   return $provider;
}

sub process_file {
   my ($pdf_file) = @_;
   my %claims;          # Store claim info
   my $member_id;       # Track member ID
   my $eob_date;        # File-level EOB date
   my @errors;          # Collect validation errors

   warn "DEBUG: Processing file: $pdf_file\n" if $debug;
   warn "DEBUG: PDFGREP pattern: $PDFGREP_PATTERN\n" if $debug;

   # Run pdfgrep and capture all output first
   warn "DEBUG: Raw pdfgrep output:\n" if $debug;
   open(my $grep, "-|", "pdfgrep", "-P", $PDFGREP_PATTERN, $pdf_file)
       or die "Failed to run pdfgrep: $!";
   my @lines = <$grep>;
   warn join("", @lines) if $debug;
   close($grep);

   # Now process the captured lines
   for my $line (@lines) {
       chomp $line;
       warn "DEBUG: Processing line: $line\n" if $debug;

       if ($line =~ /Member ID: (\w+).*EOB Date: ($DATE_PATTERN)/) {
           my $id = $1;
           my ($month, $day, $year) = split '/', $2;
           my $this_eob_date = sprintf("%04d%02d%02d", $year, $month, $day);

           # Check for EOB date conflicts
           if (defined $eob_date && $this_eob_date ne $eob_date) {
               push @errors, "Conflicting EOB dates found: $eob_date vs $this_eob_date";
           }
           $eob_date = $this_eob_date;

           warn "DEBUG: Found Member ID: $id, EOB Date: $eob_date\n" if $debug;

           if (defined $member_id && $id ne $member_id) {
               push @errors, "Multiple Member IDs found: $member_id and $id";
           }
           $member_id = $id;
       }
       elsif ($line =~ /Provider Name: (.*?)\s+Network Status:.*Claim Number: (\w+)/) {
           my $provider = normalize_provider($1);
           my $claim_num = $2;
           warn "DEBUG: Found Provider: $provider, Claim Number: $claim_num\n" if $debug;

           $claims{$claim_num} = {
               provider => $provider,
               claim_num => $claim_num,
               dates => {},
               eob_date => $eob_date,
           };
       }
       elsif ($line =~ /^\d{4}\s+($DATE_PATTERN)\s+$DATE_PATTERN/) {
           my ($month, $day, $year) = split '/', $1;
           my $date = sprintf("%04d%02d%02d", $year, $month, $day);
           warn "DEBUG: Found service date: $date\n" if $debug;
           my ($claim) = grep { !defined $_->{service_date} } values %claims;
           if ($claim) {
               $claim->{dates}{$date}++;
               # Check for multiple dates in claim
               if (keys %{$claim->{dates}} > 1) {
                   push @errors, "Multiple service dates found for a claim";
               }
               $claim->{service_date} = $date;
           }
       }
   }

   warn "DEBUG: Final claims structure:\n" . Dumper(\%claims) if $debug;

   # Validate we found necessary data
   push @errors, "No Member ID found" unless defined $member_id;
   push @errors, "No EOB date found" unless defined $eob_date;
   push @errors, "No claims found" unless %claims;

   # Validate member ID exists in mapping
   if (!exists $MEMBER_TO_INITIAL{$member_id}) {
       push @errors, "Member ID $member_id not found in mapping table";
   }

   # Validate service dates
   for my $claim (values %claims) {
       push @errors, "No service date found for claim $claim->{claim_num}"
           unless defined $claim->{service_date};
   }

   # If we have errors, report them and return
   if (@errors) {
       warn "$pdf_file: " . join("\n\t", @errors) . "\n";
       return;
   }

   # Sort claims by service date and then by claim number for stability
   my @sorted_claims = sort {
       $a->{service_date} cmp $b->{service_date}
       ||
       $a->{claim_num} cmp $b->{claim_num}
   } values %claims;

   warn "DEBUG: Sorted claims:\n" . Dumper(\@sorted_claims) if $debug;

   # Generate new filename - now handling N claims
   my $new_name = sprintf("%s.%s_%s_%s_%s.pdf",
       $MEMBER_TO_INITIAL{$member_id},
       $sorted_claims[0]->{service_date},
       $sorted_claims[-1]->{service_date},
       $eob_date,
       join('_', map { "$_->{claim_num}.$_->{provider}" } @sorted_claims)
   );

   warn "DEBUG: Generated new name: $new_name\n" if $debug;

   my $current_base = basename($pdf_file);
   if ($new_name eq $current_base) {
       warn "$pdf_file: Already has correct name\n";
       return;
   }

   my $dir = dirname($pdf_file);
   my $new_path = "$dir/$new_name";

   if (-e $new_path) {
       warn "$pdf_file: Target file $new_path already exists, skipping\n";
       return;
   }

   rename($pdf_file, $new_path);
   warn "$pdf_file -> $new_path\n";
}

die "Usage: $0 [-d] pdf-file ...\n" unless @ARGV;
process_file($_) for @ARGV;