文本处理有难度

chaogooler8888

文本
1001 yuwen 90
1001 yingyu 70
1001 shuxue 60
1002 shuxue 65
1002 huaxue 100
1003 yuwen 80
1003 yingyu 30
1003 shuxue 60
1003 huaxue 80
转化为
num yuwen shuxue yingyu huaxue
1001,90,60,70,
1002,,65,,100
1003,80,60,30,80

分数对应相应的科目，num没有该科目的默认为空，请大家帮帮忙，谢谢！

ziyunfei

awk '{a[$1]=a[$1]"\n"$2$3;b[$2]}END{printf "num";for(i in b){printf " "i}print "";for(i in a){printf i;for(j in b){if(match(a[i],j"[0-9]*")){printf ","substr(a[i],RSTART+length(j),RLENGTH-length(j))}else{printf ","}}print ""}}' a.txt

chaogooler8888

回复 2# ziyunfei


    ok thanks!

chaogooler8888

看到match,substr,RSTART,RLENGTH使用 学习了  {}

yinyuemi

1. echo "1001 yuwen 90
   
2. 1001 yingyu 70
   
3. 1001 shuxue 60
   
4. 1002 shuxue 65
   
5. 1002 huaxue 100
   
6. 1003 yuwen 80
   
7. 1003 yingyu 30
   
8. 1003 shuxue 60
   
9. 1003 huaxue 80" |awk 'NR==FNR{print;for(i=2;i<=NF;i++) a[i]=$i;l=NF;next}{for(j=2;j<=l;j++) if($0~a[j]) {b[$1 FS j]=$3};p=$1==v?p:++p;c[p]=$1;v=$1}END{for(m=1;m<=p;m++) {printf c[m] FS; for(n=2;n<=l;n++) printf b[c[m] FS n]?b[c[m] FS n] FS:"-" FS;print ""}}' <(echo "num yuwen shuxue yingyu huaxue") - |column -t
   
10. num   yuwen  shuxue  yingyu  huaxue
    
11. 1001  90     60      70      -
    
12. 1002  -      65      -       100
    
13. 1003  80     60      30      80
    

复制代码

rdcwayx

1. awk '{a[$1 FS $2]=$3;b[$1];c[$2]}
   
2. END {   printf "num" OFS;for (i in c) printf i OFS;printf RS;
   
3.         for (i in b)
   
4.              {  printf i OFS;
   
5.                 for (j in c) printf a[i FS j] OFS; printf ORS
   
6.              }
   
7.     }' OFS=","  infile |sort -n

复制代码

chaogooler8888

有没有高手用perl来实现

ziyunfei

perl -ane '{$a{$F[0]}{$F[1]}=$F[2];$b{$F[1]}=0}END{print "num";for(keys %b){print " $_"}print "\n";for(sort keys %a){print;for $s (keys %b){print ",$a{$_}{$s}"}print "\n"}}' a.txt

yinyuemi

回复 8# ziyunfei


    收藏这个Perl one liner

benteke

很强大