Question

输入文件示例：

LINE1 LINE1 M) W1 W2 W3}N) REST
LINE2 LINE2 SOME OTHER WORDS
LINE3 LINE3 LINE3 M) Z1 Z2 Z3 Z4}N) REST

在模式M）和N）之间的匹配线中我需要替换所有空格但第一个空格，预期输出为：

LINE1 LINE1 M) W1W2W3}N) REST
LINE2 LINE2 SOME OTHER WORDS
LINE3 LINE3 LINE3 M) Z1Z2Z3Z4}N) REST

Answer 1

GNU awk 解决方案：

awk 'match($0, /(M\) )(.+N\))/, a){
         gsub(/[[:space:]]+/, "", a[2]);
         sub(/M\) .+N\)/, a[1] a[2])
     }1' file

输出：

LINE1 LINE1 M) W1W2W3}N) REST
LINE2 LINE2 SOME OTHER WORDS
LINE3 LINE3 LINE3 M) Z1Z2Z3Z4}N) REST

Answer 2

关注awk可能对您有帮助。

awk 'match($0,/M) .*N)/){val=substr($0,RSTART,RLENGTH);gsub(/ +/,"",val);sub("M)","M) ",val);print substr($0,1,RSTART-1) val substr($0,RSTART+RLENGTH+1);next} 1'  Input_file

现在也添加非单线形式的解决方案。

awk '
match($0,/M) .*N)/){
  val=substr($0,RSTART,RLENGTH);
  gsub(/ +/,"",val);
  sub("M)","M) ",val);
  print substr($0,1,RSTART-1) val substr($0,RSTART+RLENGTH+1);
  next}
1
'   Input_file

输出如下。

LINE1 LINE1 M) W1W2W3}N)REST
LINE2 LINE2 SOME OTHER WORDS
LINE3 LINE3 LINE3 M) Z1Z2Z3Z4}N)REST

Answer 3

你可以尝试这个sed：

sed -E ':A;s/(.*M\) )([^ ]*)( )(.*N\).*)/\1\2\4/;tA' infile

替换除线条中2个图案之间的第一个空格

3 个答案: