package PubMed::Files;

use strict;
use warnings;

use Carp qw(confess);
use Data::Dumper;
use Date::Format;
use File::Basename;
use File::Path;
use File::Touch;
use File::Copy;

use PubMed::Common;
use PubMed::Paths;

sub prepare {
  my $file=shift;
  my $dirname=dirname($file);
  if(not -d $dirname) {
    mkpath($dirname);
  }
}

sub find_pryr {
  my $dir=shift;
  my $pryr=0;
  my $verbose=&PubMed::Common::isatty();
  foreach my $file (glob("$dir/*")) {
    if($verbose) {
      print "file $file\n";
    }
    if(not $file=~m|pubmed(\d\d)n|) {
      if(not basename($file) eq 'README.txt') {
	warn "I can't deal with $file.";
      }
      next;
    }
    if($1 > $pryr) {
      $pryr=$1;
    }
  }
  return $pryr;
}

sub check_for_pryr {
  my $file=shift;
  my $pryr=shift;
  my $regex="pubmed$pryr".'n\d+';
  if($file=~m|$regex|) {
    return 1;
  }
  return 0;
}

#sub pmid_to_fapi {
#  my $pmid=shift // confess "I need a pmid here";
#  my $pemi=&PubMed::Common::pmid_to_pemi($pmid);
#  my $file=$PubMed::Paths::fapi_dir."/$pemi/$pmid.xml";
#  ## we can save this overhead now
#  # &PubMed::Files::prepare($file);
#  return $file;
#}

#sub fapi_to_pmid {
#  my $fapi=shift // confess "I need a fapi here";
#  my $bana=basename($fapi);
#  my $pmid=substr($bana,0,length($bana)-4);
#  return $pmid;
#}

sub find_newest {
  my $dir=shift;
  my $regx=shift // '';
  my $verbose=shift // '';
  if(not -d $dir) {
    confess "I need a directory here, you gave me '$dir'";
  }
  my $o={};
  my $min_age=time;
  my $min_file='';
  foreach my $file (glob("$dir/*")) {
    if($regx) {
      if(not $file=~m|$regx|) {
	if($verbose) {
	  print "find_newest: $file does not match '$regx'\n";
	}
	next;
      }
    }
    my $age=-M $file;
    if($age < $min_age) {
      $min_age=$age;
      $min_file=$file;
    }
  }
  $o->{'file'}=$min_file;
  $o->{'age'}=$min_age;
  return $o;
}

sub find_oldest {
  my $dir=shift;
  my $regx=shift // '';
  my $verbose=shift // '';
  if(not -d $dir) {
    confess "I need a directory here, you gave me '$dir'";
  }
  my $o={};
  my $max_age=0;
  my $max_file='';
  foreach my $file (glob("$dir/*")) {
    if($regx) {
      if(not $file=~m|$regx|) {
	if($verbose) {
	  print "find_oldest: $file does not match '$regx'\n";
	}
	next;
      }
    }
    my $age=-M $file;
    if($age > $max_age) {
      $max_age=$age;
      $max_file=$file;
    }
  }
  $o->{'file'}=$max_file;
  $o->{'age'}=$max_age;
  return $o;
}

sub file_to_date {
  my $file=shift // confess "I need a file here.";
  my $mtime=&mtime($file);
  my $date=time2str('%Y-%m-%d',$mtime);
  return $date;
}

sub sou_in_file {
  my $file=shift // confess "I need a file here.";
  if(not -f $file) {
    confess "I can't open $file";
  }
  my $sou_file="$file.sou";
  my $s="export LC_ALL=C; $PubMed::Paths::sort $file | $PubMed::Paths::uniq > $sou_file";
  system($s);
  my $toucher = File::Touch->new(reference => $file, no_create => 1);
  $toucher->touch($sou_file);
  move($sou_file,$file);
}


sub mtime {
  my $file=shift // confess "I need a file here.";
  if(not -f $file) {
    confess "I can't see the file $file";
  }
  my ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
      $atime,$mtime,$ctime,$blksize,$blocks)
    = stat($file);
  return $mtime;
}

## a function reference
## sorting for raw files, probably not required every time.
## Fixme:: there is different version of this in PubMed::File
our $sort_uniq = sub {
  my $fufi=shift // confess "I need a file here.";
  my $tmp = File::Temp->new(UNLINK => 0);
  my $s="export LC_ALL=C; $PubMed::Paths::sort $fufi | $PubMed::Paths::uniq > $tmp";
  system($s);
  move($tmp,$fufi);
};

sub does_file_need_renewal {
  my $file=shift;
  my @others=@_;
  my $verbose=0;
  if(not -f $file) {
    if($verbose) {
      print "file $file is not there, it needs renewing\n.";
    }
    return 1;
  }
  if($verbose) {
    print "file $file is there ... ";
  }
  if(-z $file) {
    if($verbose) {
      print "file $file is empty, it needs renewing\n.";
    }
    return 1;
  }
  ## -M  Script start time minus file modification time, in days.
  my $target_time=-M $file;
  if($verbose) {
    print "target_time is $target_time\n";
  }
  if($verbose) {
    print Dumper @others;
  }
  foreach my $file (@others) {
    if($verbose) {
      print "considering $file as renewal target\n";
    }
    if(not $file) {
      confess "I don't have the file $file";
    }
    if(-d $file) {
      ## directory case, not treated recursively
      my $dirname=$file;
      opendir( my $dir, $dirname ) or die "Error: can't open dir $dirname";
      my $file;
      while ($file = readdir $dir ) {
        ## skipping "." and ".."
        if ( ($file eq "." ) or ( $file eq ".." )) {
          next;
        }
        $file="$dirname/$file";
        if(-M $file < $target_time) {
          if($verbose) {
            print " $file is newer, renewal required\n";
          }
          return 1;
        }
      }
      ## finished with this
      next;
    }
    if(not -f $file) {
      die "I can not find the file '$file'";
    }
    if($verbose) {
      print "time on file $file is ", -M $file, "\n";
    }
    if(-M $file < $target_time) {
      if($verbose) {
        print " but $file is newer, renewal required\n";
      }
      return 1;
    }
  }
  if($verbose) {
    print " no reed to renew\n";
  }
  return 0;
}




1;