zhangbo-NLP-plugin_eng/lib/zhangbo/NLP/gen_babelnet.pl
#!/usr/bin/perl
#use JSON::XS;
use Clone qw(clone);
use MongoDB;
use Smart::Comments;
use lib "/home/wyb/shell/";
use Conn_mongo_jc;
use Data::Dumper;
use IO::File;
#use Add_info;
# http://poe.perl.org/?POE_Cookbook/TCP_Servers
# Include POE and POE::Component::Server::TCP.
our $hash=();
$hash_mn=();
use URI::Escape;
my $hash_sc=();
our $hash_babelnet;
open(FD, "/root/babelnet.txt");
open(FD2, ">/root/babelnet_dic.txt");
open(FD3, ">/root/babelnet_dic_r.txt");
my $tmp=();
my $netn=0;
while(<FD>)
{
my $line= $_;
if($netn<71)
{$netn++;next;}
$netn++;
my @a=split("\t",$line);
#bn:03816372n WIKI CONCEPT 20 WIKIRED:EN:Oxalic_acid_diamide WIKIRED:EN:Oxamimidic_acid WIKI:EN:Oxamide WIKIRED:FA:ﺍکﺱﺎﻤﻳﺩ WIKIRED:FA:ﺎﻜﺳﺎﻤﻳﺩ WIKIRED:PT:Oxalamida WIKI:DE:Oxamid WIKI:JA:オキサミド WIKI:PT:Oxamida WIKI:SV:Oxamid WIKI:ZH:乙二酰二胺 WIKI:PL:Oksamid WIKI:FA:ﺍکﺱﺎﻣیﺩ WIKIRED:FA:ﺎﻜﺳﺎﻣیﺩ WIKIRED:ZH:草酰胺 WIKIRED:EN:C2H4N2O2 WIKIRED:EN:Oxalamide WIKIRED:EN:Ethanediamide WIKIRED:EN:1-carbamoyl-formimidic_acid WIKIRED:EN:Diaminoglyoxal 7 2_6,7,8,9,10,11,12 1 DE:File:Oxamide.png 6 EN:Amides DE:Carbonsäureamid FA:ﺂﻣیﺪﻫﺍ PL:Amidy ZH:酰胺 ZH:二碳有机物 48 r bn:03892978n r bn:00072720n r bn:00027019n r bn:00045383n r bn:00015849n r bn:03130094n r bn:00068085n r bn:00059475n r bn:00042379n r bn:01399263n r bn:00003518n r bn:00024691n r bn:00031692n r bn:00031694n r bn:00993461n r bn:00035949n|FROM_PL r bn:00035930n|FROM_PL r bn:00055604n|FROM_SV r bn:00003549n|FROM_SV r bn:00018110n|FROM_SV r bn:00002355n|FROM_SV r bn:00068091n|FROM_SV r bn:00045397n|FROM_SV r bn:03374210n|FROM_SV r bn:00041296n|FROM_FA r bn:00018110n|FROM_FA r bn:00018101n|FROM_FA r bn:00073790n|FROM_ZH r bn:00015870n|FROM_ZH r bn:00024227n|FROM_ZH r bn:00006904n|FROM_ZH r bn:00017060n|FROM_ZH r bn:00037689n|FROM_ZH r bn:00004212n|FROM_ZH r bn:00003549n|FROM_ZH r bn:00021370n|FROM_ZH r bn:00505400n|FROM_ZH r bn:00045397n|FROM_ZH r bn:00006901n|FROM_ZH r bn:00017060n|FROM_JA r bn:00018110n|FROM_JA r bn:00034136n|FROM_DE r bn:00006912n|FROM_DE r bn:00045380n|FROM_DE r bn:00059965n|FROM_DE r bn:00072758n|FROM_DE r bn:03359325n|FROM_DE r bn:00045397n|FROM_DE
#bn:00000002n WIKIWN 08950407n NAMED_ENTITY 377 WIKIRED:EN:Hague WIKIRED:EN:S`Gravenshage WIKIRED:EN:Den_hague WIKIRED:EN:'s_GravenhageWIKIRED:EN:’s-Gravenhage WIKIRED:EN:The_hague WIKIRED:EN:The_haig WIKIRED:EN:Den_haag WIKIRED:EN:Den_Haag WIKI:EN:The_Hague WIKI:PT:Haia WIKIRED:SL:Den_Haag WIKIRED:IS:Den_Haag WIKIRED:SV:Gravenhage WIKI:ZH:海牙 WIKI:MS:The_Hague WIKIRED:CA:L'Haia WIKI:GA:An_Háig WIKIRED:LA:Haga_Comitum WIKIRED:TR:The_Hague WIKI:NL:Den_Haag WIKIRED:PT:A_Haia WIKIRED:MS:Hague WIKIRED:CY:Yr_Hâg WIKI:VI:Den_Haag WIKIRED:PL:La_Haye WIKIRED:RO:Den_Haag WIKIRED:CY:’s-Gravenhage WIKI:BG:Хага WIKIRED:VI:La_Haye WIKI:ET:Haag WIKI:HE:האג WIKIRED:UK:The_Hague WIKI:DA:Haag WIKIRED:NL:De_weduwe_van_Indie WIKI:ID:Den_Haag WIKIRED:ET:'s-Gravenhage WIKI:FR:La_Haye WIKIRED:SK:Den_Haag WIKI:SIMPLE:The_Hague WIKIRED:PT:The_Hague WIKIRED:DE:’s_G
my $bnid=$a[0];
my $bntype,$bnof1,$bnof2;
#print $line;
if($a[2] eq "CONCEPT")
{
$bnof1=$a[2];
# $hash_bnid->{$bnid}->{"t"}=$bnof1;
}
elsif($a[3] eq "NAMED_ENTITY")
{
$bnof1=$a[3];
# $hash_bnid->{$bnid}->{"t"}=$bnof1;
}
elsif($a[3] eq "CONCEPT")
{$bnof1=$a[3];
# $hash_bnid->{$bnid}->{"t"}=$bnof1;
}
elsif($a[2] eq "NAMED_ENTITY")
{$bnof1=$a[2];
# $hash_bnid->{$bnid}->{"t"}=$bnof1;
}
else{
print "$a[0]-$a[1]-$a[2]-$a[3]-$a[4]\n";
}
print FD3 "\n$bnid";
if(length($bnid)<=1)
{
print $line;
exit;
}
for(4 .. scalar(@a))
{
my $n=$_;
my $n_n,$w;
if($a[$n]=~/Image/)
{next;}
if($a[$n]=~/File/)
{next;}
$a[$n]=~s/_//g;
if($a[$n]=~/ZH\:(.*)\:/)
{
$w=$1;
print FD2 "$w\t$bnid\t$bnof1\n";
}
elsif($a[$n]=~/ZH\:(.*)/)
{
$w=$1;
print FD2 "$w\t$bnid\t$bnof1\n";
}
elsif($a[$n]=~/^r (bn\:.*?n)$/)
{
$n_n=$1;
print FD3 "\t$n_n";
}
elsif($a[$n]=~/^r (bn\:.*?n).*ZH$/)
{
$n_n=$1;
print FD3 "\t$n_n";
}
}
#print Data::Dumper->Dump([%{$hash}]);
#print Data::Dumper->Dump([%{$hash_bnid}]);
}
print "server start\n";
exit;
use POE qw(Component::Server::TCP);
my $n=0;
#our $f=0;
POE::Component::Server::TCP->new(
Alias => "echo_server",
Port => 11219,
ClientInput => sub {
my ($session, $heap, $input) = @_[SESSION, HEAP, ARG0];
#print "Session ", $session->ID(), " got input: ".$input."\n";
#print "Session ", $session->ID(), " got input: ".length($input)."\n";
my $d;
open(FD_log,">/tmp/engine.log");
print FD_log "$input\n";
#system("cp $input /tmp/");
print "input file $input",length($input),"\n";
my @res;
if ($input=~/^cluster/)
{
@res=search_article2($input);
}
else
{
@res=search_article($input);
}
print FD_log @res;
close(FD_log);
#print scalar(@res)," num\n";
$heap->{client}->put(@res);
#$session->yield("shutdown");
$_[KERNEL]->yield("shutdown");
return;
},
ClientDisconnected => sub {
#print "Client disconnected\n"; # log it
}
#, ClientFlushed => sub {
# my $data_source = $_[HEAP]{file_handle};
# my $read_count = sysread($data_source, my $buffer = "", 65536);
# if ($read_count) {
# $_[HEAP]{client}->put($buffer);
# }
# else {
# print FD_log "------------------error\n";
# $_[KERNEL]->yield("shutdown");
# }
#}
);
POE::Kernel->run;
#$input="坦克,我的希望非常诱人电视里有主持人，那里有大熊猫";
# $input="网络和网民的意见，现在正行驶在通州回北京的高速公路上，十评论员单仁平的文章，标题是“做大众政治焦点，茅于轼的选择”。这篇文章的核心意见是，茅于轼应该做中国社会团结的促进者，不应该";
#$input="致力于宣传 市场万能 剥削有理 汉奸人性 保钓无用 保粮错误替富人说话的茅于轼，今天下午两点在北京海淀翠宫饭店演讲顽强继续。未知海淀区委书记隋振江，宣传部长陈名杰是否到场。外媒问我是否到场，告曰:先参加央视《苦难辉煌》座谈会，或会晚到一会儿。今天要长见识了";
# $input2="茅于轼刚刚吃完胡辣汤和烧饼夹猪头肉。现在正行驶在通州回北京的高速公路上，十五分钟以后要开始腾讯微访谈直播，我和方舟子拟就昨天的热点问题，回答网友的提问。敬请各位网友提示一下，昨天有什么热点问题需要谈一谈？";
#$file=$ARGV[0];
sub search_article
{
$file=$_[0];
chomp($file);
#print "file=$file\n";
$expword=$ARGV[1];
$exp_hash=();
@exp_w=split(",",$expword);
foreach $exp (@exp_w)
{
$exp_hash->{$exp}=1;
}
my $f=0;
my $n=0;
my $kk;
my $thash=();
my @input_ar=`cat $file`;
my %my_hash;
my %my_hash2;
my %thash;
my $j;
#my $rand2=rand();
#open(FD_hash,">$file.txt");
#print "openfile $file\n";
open(FD_arr,"<$file");
my $first=0;
my $rand=rand();
open(FD_aro,">/tmp/$rand.sctxt");
my $j=1;
while(<FD_arr>)
{
#print FD_log $j,"----------\n";
#print $j,"----------\n";
my $in=$_;
if($first==0)
{
$first++;
%my_hash=txt2arr3($in);
# while (($k,$v)=each(%my_hash))
# {
#print "txt2arr_one $k,$v\n";
# }
# print FD_hash "\n";
$d= clone(\%my_hash);
print FD_aro $in;
}
else
{
if(length($in)>1000)
{ $in=substr $in,0,500;}
# $in=~s/\@.*?\:/ /g;
# $in=~s/\@.*?\s/ /g;
# $in=~s/机场|航班|深圳/ /g;
print FD_aro "$j $in";
$j++;
#print FD_log $j,"-p----",length($in),"\n";
#print $j,"-p----",length($in),"\n";
}
}
close FD_arr;
close FD_aro;
system("/home/wyb/xs/bin/scws -I -E -N -d /home/lzj/shell2/zhonto.xdb -c utf-8 /tmp/$rand.sctxt -o /tmp/$rand.sctxt2");
#system("/home/wyb/xs/bin/scws -I -E -N -d /home/lzj/shell/bigdic.xdb:/home/wyb/stanford/dict_user.txt -c utf-8 /tmp/$rand.sctxt -o /tmp/$rand.sctxt2");
system("cp /tmp/$rand.sctxt ./last_scws.txt");
unlink("/tmp/$rand.sctxt");
open(FD_scws,"/tmp/$rand.sctxt2");
while(<FD_scws>)
#foreach my $in (@output)
{
my $in=$_;
print FD_log "line ttt $kk $in \n";
chomp($in);
#$input_ar[$f]=$in;
if($f==0)
{
$f++;
$kk++;
#if(length($in)>12)
#{
($d,$c)=txt2arr3($in,$exp_hash);
# $d=clone($my_hash);
#}
next;
#print Data::Dumper->Dump([%{$d}]);
# %my_hash2=txt2arr($input2);
# $n=comp_hash(\%my_hash,\%my_hash2);
}
else
{
my $n_line;
my ($my_hash2,$n_line)=txt2arr3($in,$exp_hash);
# while (($k,$v)=each(%my_hash2))
# {
# print FD_hash "$k,$v\t";
# }
# print FD_hash "\n";
$n=comp_hash($d,$my_hash2);
print FD_log "totle$n_line=$n\n";
$thash->{$n_line}=$n;
$kk++;
}
$f++;
}
unlink("/tmp/$rand.sctxt2");
#close(FD_hash);
close(FD_arr);
my @key =sort {$thash->{$b} <=> $thash->{$a}} keys %{$thash};
my @m_res;
#print @key;
my $o_n;
foreach $i (@key)
{
if($o_n >200){ last;}
else
{
if( $thash->{$i}>0)
{
push (@m_res,$thash->{$i}."-$i-".$input_ar[$i]);
$o_n++;
}
}
}
print FD_log "return ",scalar(@m_res),"\n";
@m_res;
}
# }
# );
#$poe_kernel->run();
sub txt2arr
{
my $input_t=$_[0];
my $exp_h=$_[1];
my $tmp_hash=();
# chomp($input_t);
my @list=split(" ",$input_t);
#print "txt2arr input =$input_t\n";
#$rand=rand();
#open(FD_r,">/tmp/$rand");
print FD_log "$input_t\n";
#close(FD_r);
#$dir="/home/wyb/stanford-parser-2012-07-09";
@res=();
#@res=`$dir/parser-10.pl /tmp/$rand 0 50`;
#$part=0;
#unlink("/tmp/$rand");
#print @res;
my $n_a=0;
my $n_b=0;
my $v_a=0;
my $v_b=0;
my $n1;
my $n2;
my $v1;
my $v2;
my $nline;
my $tmp_first=0;
foreach $tmp_res (@list)
{
if ($tmp_first==0)
{
$nline=$tmp_res;
$tmp_first++;
next;
}
if(exists($hash_st->{$tmp_res}))
{next;}
my $l=length($tmp_res);
# if($l<=3)
# {
#if($hash_sc->{$tmp_res}->{'a'} !~/n|v|a/)
#{
#print FD_log "seek $tmp_res\n";
#next;}
if($l<=3)
{next;}
#主题放大
if(exists($hash_sc->{$tmp_res}))
{
if($hash_sc->{$tmp_res}->{'a'}=~/n/)
{
if($hash_sc->{$tmp_res}->{'h'}> $n_a)
{
$n_a=$hash_sc->{$tmp_res}->{'h'};
$n1=$tmp_res;
print FD_log "放大主题-$tmp_res-","$n_a n\n";
if($n_a >$n_b)
{
my $swtmpn=$n_b;
my $swtmpnw=$n2;
$n_b=$n_a;$n2=$n1;
$n_a=$swtmpn;$n1=$swtmpnw;
}
}
}
elsif($hash_sc->{$tmp_res}->{'a'}=~/v/)
{
if($hash_sc->{$tmp_res}->{'h'}> $v_a)
{
$v_a=$hash_sc->{$tmp_res}->{'h'};
$v1=$tmp_res;
print FD_log "放大主题-$tmp_res-","$v_a v \n";
if($v_a >$v_b)
{
my $swtmpn=$v_b;
my $swtmpnw=$v2;
$v_b=$v_a;$v2=$v1;
$v_a=$swtmpn;$v1=$swtmpnw;
}
}
}
}
# }
# }
# print $tmp_res,"---",length($tmp_res),"\n";
#if (($part==1)&&($tmp_res=~/(.*?),(.*)/))
#{
#print "切词1=$1,2=$2\n";
my $w=$tmp_res;
# my $value=$2;
# if(exists($exp_h->{$w}))
# {
# print "$w\n";
# next;}
if(exists($hash->{$w}))
{
my $t_n=0;
while(($k,$v)=each($hash->{$w}))
{
if(exists($hash_st->{$k}))
{next;}
if(exists($d->{$k}))
{
if(exists( $tmp_hash->{$k})){$tmp_hash->{$k}=$tmp_hash->{$k}+0.1;}
else{$tmp_hash->{$k}=$d->{$k}+1;}
}
else
{
$tmp_hash->{$k}=1;
}
#$tmp_hash->{$k}=$v;
#$tmp_hash->{$k}=$hash_sc->{$w}->{'h'};
print FD_log "正常词 $w 转换",$hash_sc->{$w}->{'h'},"--",$tmp_hash->{$k}," $k\n";
#正常词 交 转换0.00--0.00 相互
#if(exists($sytmp->{$k}))
#{
# print "归一 $k ",$sytmp->{$k},"\n";
# $k=$sytmp->{$k};
#}
# if($value>20)
# $heap->{client}->put("$k $v\n");
}
$tmp_hash->{$k}=$v;
}
else
{
if(exists($hash_sc->{$w}))
{
if(exists($d->{$w}))
{
if(exists( $tmp_hash->{$w})){$tmp_hash->{$k}=$tmp_hash->{$k}+0.2;}
else{$tmp_hash->{$w}=$d->{$w} +1;}
}
else
{
$tmp_hash->{$w}=1;
}
#$tmp_hash->{$w}=$hash_sc->{$w}->{'h'};
}
else
{
if(exists($d->{$w}))
{
if(exists( $tmp_hash->{$w})){$tmp_hash->{$w}=$tmp_hash->{$w}+0.2;}
else{$tmp_hash->{$w}=$d->{$w} +1;}
}
else
{
$tmp_hash->{$w}=1;
}
$tmp_hash->{$w}=0.1;
}
# print FD_log "no exists wordnet 切词 k=$w,v=",$hash_sc->{$w}->{'h'},"\n";
}
}
while(($k,$v)=each(%{$tmp_hash}))
{
if(($k eq $n1)||($k eq $n2)||($k eq $v1)||($k eq $v2))
{$tmp_hash->{$k}=$v*3;
print FD_log "放大主题-$k-",$tmp_hash->{$k}," \n";
}
}
return (\%{$tmp_hash},$nline);
}
#print "output = $res\n";
# $heap->{client}->stop();
# $session->stop();
# Start the server.
sub comp_hash
{
my $h1=@_[0];
my $h2=@_[1];
my $akv=0;
my $all;
my %all;
my $vall=0.01;
my $vall2=0.01;
my $k,$v;
while(($k,$v)=each(%{$h1}))
{
$all->{$k}++;
$vall+=$v;
print FD_log " hash 1$k $v\n";
}
my $k,$v;
while(($k,$v)=each(%{$h2}))
{
$all->{$k}++;
print FD_log " hash 2$k $v\n";
$vall2+=$v;
}
my $k,$v;
while(($k,$v)=each(%{$all}))
{
if($v >= 2)
{
$kv=($h1->{$k}+$h2->{$k})/($vall+$vall2);
$kv2=((($h1->{$k})/$vall) + ($h2->{$k}/$vall2))/2;
$kv3=($kv+$kv2)/2;
#if($h1->{$k}<$h2->{$k})
#{
#}
print FD_log "double $k $v kv $kv kv2 $kv2\n";
$akv+=$kv3;
}
}
# print "---------------\n";
$akv;
}
sub ndate
{
($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst)=localtime(time());
$year=~s/1(.*)/20$1/;
$mon+=1;
if($mon=~/^\d$/)
{
$mon="0"."$mon";
}
if($mday=~/^\d$/)
{
$mday="0"."$mday";
}
if($hour=~/^\d$/)
{
$hour="0"."$hour";
}
if($min=~/^\d$/)
{
$min="0"."$min";
}
if($sec=~/^\d$/)
{
$sec="0"."$sec";
}
my $res=$year."-".$mon."-".$mday." ".$hour.":"."$min".":"."$sec";
#print $res,"\n";
$res;
}
sub txt2arr_one
{
my $w=$_[0];
my $exp_h=$_[1];
$tmp_hash=();
$rand=rand();
@res=();
$part=0;
#print @res;
my @w_arr=split(",",$w);
foreach my $w_tmp (@w_arr)
{
if($w_tmp=~/(.*?) (.*)/)
{
$w=$1;
$vh=$2;
#print "$w -- $vh\n";
if(exists($hash->{$w}))
{
my $t_n=0;
while(($k,$v)=each($hash->{$w}))
{
if($v==0)
{
$v=1;
}
$tmp_hash->{$k}=$vh;
}
}
$tmp_hash->{$w}=$w;
}
else
{
print "input error\n";
}
}
# while(($k,$v)=each(%{$tmp_hash}))
#
# {
#print "meaning $k $v\n";
# }
# }
return (%{$tmp_hash});
}
sub txt2arr3
{
my $input_t=$_[0];
my $exp_h=$_[1];
my $tmp_hash=();
my @list=split(" ",$input_t);
print FD_log "$input_t\n";
@res=();
my $n_a=0;
my $n_b=0;
my $v_a=0;
my $v_b=0;
my $n1;
my $n2;
my $v1;
my $v2;
my $nline;
my $tmp_first=0;
foreach $tmp_res (@list)
{
if ($tmp_first==0)
{
$nline=$tmp_res;
$tmp_first++;
next;
}
if(exists($hash_st->{$tmp_res}))
{next;}
my $l=length($tmp_res);
if($l<=3)
{next;}
#主题放大
if(exists($hash_sc->{$tmp_res}))
{
if($hash_sc->{$tmp_res}->{'a'}=~/n/)
{
if($hash_sc->{$tmp_res}->{'h'}> $n_a)
{
$n_a=$hash_sc->{$tmp_res}->{'h'};
$n1=$tmp_res;
print FD_log "放大主题-$tmp_res-","$n_a n\n";
if($n_a >$n_b)
{
my $swtmpn=$n_b;
my $swtmpnw=$n2;
$n_b=$n_a;$n2=$n1;
$n_a=$swtmpn;$n1=$swtmpnw;
}
}
}
elsif($hash_sc->{$tmp_res}->{'a'}=~/v/)
{
if($hash_sc->{$tmp_res}->{'h'}> $v_a)
{
$v_a=$hash_sc->{$tmp_res}->{'h'};
$v1=$tmp_res;
print FD_log "放大主题-$tmp_res-","$v_a v \n";
if($v_a >$v_b)
{
my $swtmpn=$v_b;
my $swtmpnw=$v2;
$v_b=$v_a;$v2=$v1;
$v_a=$swtmpn;$v1=$swtmpnw;
}
}
}
}
my $w=$tmp_res;
if(exists($hash->{$w}))
{
my $t_n=0;
while(($k,$v)=each($hash->{$w}))
{
if(exists($hash_st->{$k}))
{next;}
$tmp_hash->{$k}+=0.1;
print FD_log "正常词 $w 转换",$hash_sc->{$w}->{'h'},"--",$tmp_hash->{$k}," $k\n";
}
}
else
{
if(exists($hash_sc->{$w}))
{
$tmp_hash->{$w}+=0.2;
}
else
{
$tmp_hash->{$w}=0.1;
}
}
if(exists($hash_zhonto->{$w}))
{
while(($k,$v)=each($hash_zhonto->{$w}))
{
$tmp_hash->{$k}+=0.1;
print FD_log "正常词 $w zhonto转换",$hash_sc->{$w}->{'h'},"--",$tmp_hash->{$k}," $k\n";
}
}
}
#是否采用scws放大文章中心词
while(($k,$v)=each(%{$tmp_hash}))
{
if(($k eq $n1)||($k eq $n2)||($k eq $v1)||($k eq $v2))
{$tmp_hash->{$k}=$v*3;
print FD_log "放大主题-$k-",$tmp_hash->{$k}," \n";
}
}
return (\%{$tmp_hash},$nline);
}
sub search_article2
{
my $file=$_[0];
chomp($file);
#print "file=$file\n";
$expword=$ARGV[1];
$exp_hash=();
@exp_w=split(",",$expword);
foreach $exp (@exp_w)
{
$exp_hash->{$exp}=1;
}
my $f=0;
my $n=0;
my $kk;
my $thash=();
my @input_ar=`cat $file`;
my %my_hash;
my %my_hash2;
my %thash;
my $j;
#my $rand2=rand();
#open(FD_hash,">$file.txt");
#print "openfile $file\n";
open(FD_arr,"<$file");
my $first=0;
my $rand=rand();
open(FD_aro,">/tmp/$rand.sctxt");
my $j=1;
while(<FD_arr>)
{
my $in=$_;
chomp($in);
print FD_aro "$j $in\n";
$j++;
}
close FD_arr;
close FD_aro;
system("/home/wyb/xs/bin/scws -I -E -N -d /home/wyb/xs/etc/dict.utf8.xdb:/home/wyb/stanford/dict_user.txt -c utf-8 /tmp/$rand.sctxt -o /tmp/$rand.sctxt2");
unlink("/tmp/$rand.sctxt");
open(FD_scws,"/tmp/$rand.sctxt2");
my $hashleft;
my $alln=1;
while(<FD_scws>)
{
my $in=$_;
print FD_log "line ttt $kk $in \n";
chomp($in);
my $n_line;
my ($my_hash2,$n_line)=txt2arr($in,$exp_hash);
$hashleft[$n_line]=clone($my_hash2);
$alln++;
}
for( 1 .. $alln-1)
{
my $nn=$_;
for ($nn .. $alln-1)
{
$nn2=$_;
if($nn==$nn2)
{next;}
$n=comp_hash($hashleft[$nn],$hashleft[$nn2]);
$n=$n * 100;
print "$n $nn $nn2 \n";
print FD_log "$n $nn $nn2 \n";
}
}
unlink("/tmp/$rand.sctxt2");
close(FD_arr);
@m_res;
}