文本處理求助

peawseshc

1. 源文本：
   
2. 
   
3. open_sep01activity001 19
   
4. open_sep01activity002 0
   
5. open_sep01activity003  0
   
6. open_sep01activity004   10
   
7. open_sep01activity005 0
   
8. 
   
9. sep01activity005    0
   
10. sep01activity001    2
    
11. sep01activity003 0
    
12. sep01activity004  2
    
13. 
    
14. 
    
15. sep01activity_success004  2
    
16. sep01activity_success003 0
    
17. sep01activity_success001    2
    
18. sep01activity_success005    0
    
19. 
    
20. 
    
21. 
    
22. 
    
23. 目標(biāo)文本：
    
24. 
    
25. 001      19   2   2
    
26. 002      0    0   0
    
27. 003      0    0   0
    
28. 004      10   2   2
    
29. 005      0    0   0

復(fù)制代碼

hz_oracle

1. sed 's/^.*00/00/' file | awk '$0~/00/{a[$1]=a[$1]" "$2}END{for(i in a)print i a[i]}'

復(fù)制代碼

001 19 2 2
002 0
003 0 0 0
004 10 2 2
005 0 0 0

peawseshc

回復(fù) 2# hz_oracle

謝謝，基本符合要求。002后面兩列空的能補(bǔ)成0嗎。

hz_oracle

1. awk '$1~/00/{gsub("^.*00","00",$1);a[$1]=a[$1]" "$2}END{for(i in a)print i a[i]}' file | awk 'NF<4{a=NF+1;for(i=a;i<=4;i++)$i=0}{print}'

復(fù)制代碼

001 19 2 2
002 0 0 0
003 0 0 0
004 10 2 2
005 0 0 0

peawseshc

回復(fù) 4# hz_oracle

謝謝~

zxy877298415

回復(fù) 1# peawseshc

1. awk -F '[a-z ]+' '!/^$/{a[$(NF-1)]=a[$(NF-1)]" "$NF}END{for(i in a) {t=split(a[i],c," ");if(t>=m) m=t;}for(i in a) {printf i a[i];for(j=1;j<=m-split(a[i],c," ");j++) printf " 0 ";print ""}}' file
   

復(fù)制代碼

moperyblue

1. awk 'function p(){for(i in a)a[i]=a[i] FS (b[i]?b[i]:0)};{sub(/.*[[:alpha:]]/,"",$1)}NR==1,NF==1{if(NF==2)a[$1]=$2;next}NF>1{b[$1]=$2;f=1}NF==1&&f{p();delete b;f=0}END{p();for(i in a)print i,a[i]|"sort"}' file

復(fù)制代碼

peawseshc

回復(fù) 6# zxy877298415

可用，謝謝~

fh21_xuejinlian

awk  '{l=split($1,a,"");b[a[l-2]a[l-1]a[l]]=b[a[l-2]a[l-1]a[l]]" "$2}END{for(i in b)print i,b[i]}' file

sunzhiguolu

1. #!/usr/bin/perl
   
2. use strict;
   
3. use warnings;
   
4. 
   
5. local $/ = '';
   
6. my %hData = map {my @aT = /\d++\b/g; $aT[0] => [$aT[-1]]} split (/\n/, scalar <>);
   
7. 
   
8. while (<>){
   
9.     my %hT = /\d++\b/g;
   
10.     push (@{$hData{$_}}, $hT{$_} // 0) for keys %hData;
    
11. }
    
12. 
    
13. print join ("\t", $_, @{$hData{$_}}), "\n" for sort keys %hData;

復(fù)制代碼