大数据分析

Question

# 1.0
#=GF ID   45651
#=GF AC   CD7.8

我使用的命令是

awk '{print $0 > $2 NR}' RS='//' assignment.txt

这是生成名为file0、file1等的文件。

Answer 1

使用GNU awk进行多字符RS和打开文件处理：

awk -v RS='\n//\n' -F'\n' '{print > ($3 ".txt")}' file

这当然会产生糟糕的输出文件名，因为它们将同时包含空格和=符号，并以#开头，但这正是您所要求的.

Answer 2

考虑到您不想在输出文件中使用//，请您尝试以下操作。

awk '
/^\/\//{
  close(file".txt")
  flag=val=""
}
/#=GF AC   PF.*/{
  flag=1
  file=$0
  sub(/^#=/,"",file)
  print val ORS $0 > file".txt"
  next
}
flag{
  print > file".txt"
}
!flag{
  val=val?val ORS $0:$0
}' Input_file

上面的代码将从文件名中删除#=，以防您在输出文件名中也需要这样做，然后也从上面的代码中删除sub(/^#=/,"",file)语句。

解释：现在也添加了对上述代码的解释。

awk '                               ##awk program starts here.
/^\/\//{                            ##Checking condition if a line has // then do following.
  close(file".txt")                 ##Using close command to close file with file named file".txt" here to avoid TOO MANY FILES OPENED.
  flag=val=""                       ##Nullifying variables flag and val here.
}
/#=GF AC   PF.*/{                   ##Checking condition if a line starts from #=GF AC   PF .* will match anything here, if yes then do following.
  flag=1                            ##Setting variable flag as 1 here.
  file=$0                           ##Setting variable file value to current line here.
  sub(/^#=/,"",file)                ##Substituting #= from start of the line with variable file.
  print val ORS $0 > file".txt"     ##Printing variable val ORS(output record separator) current line and printing it to file".txt"
  next                              ##next will skip all further statements from here onward.
}
flag{                               ##Checking condition if variable flag is SET then do following.
  print > file".txt"                ##Printing current lines to file".txt" file name.
}
!flag{                              ##Checking condition here is variable flag is NOT SET then do following.
  val=val?val ORS $0:$0             ##Creating variable val and concatenating to its own value.
}' Input_file                       ##Mentioning Input_file name here.