用字符串替换一行或创建一个不存在的新字符串-sed

Question

I want to replace 'DUMP=' line or add new line if doesn't exist within a certain block. When block define by from [text] until the next [text] , for example:

[name]
  sometext=...
  moretext=...
  DUMP=
  moretext2=...    
[name2]
  ...

For replacing existing line I will do:

sed -i '/^\['name'\]$/,/^\[/!b
    /DUMP=/cDUMP=example' $file

I don't know if it is possible in one liner or two, but how do I add the option to add DUMP=... if it doesn't exists in that block?

Example 1:

file1.txt :

[name]
  A=text1
  B=text2
  DUMP=x
  C=text3

[name2]
  A=text12
  B=text22
  DUMP=x2
  C=text32

sed '/^\\['name'\\]$/,/^\\[/!b /DUMP=/cDUMP=example' file1.txt output:

[name]
  A=text1
  B=text2
  DUMP=example
  C=text3

[name2]
  A=text12
  B=text22
  DUMP=x2
  C=text32

Example 2:

file1.txt :

[name]
  A=text1
  B=text2
  C=text3

[name2]
  A=text12
  B=text22
  DUMP=x2
  C=text32

sed '/^\\['name'\\]$/,/^\\[/!b /DUMP=/cDUMP=example <add DUMP incase it doesn't exist>' file1.txt output:

[name]
  A=text1
  B=text2
  C=text3
  DUMP=example

[name2]
  A=text12
  B=text22
  DUMP=x2
  C=text32

Answer 1

awk '/^\[/&&/]$/&&s&&!f{print "  DUMP=example"; s=0}
     $0 ~ "^\["blockname"]$"{s=1;f=0}
     s&&/^ *DUMP=/{$0="  DUMP=example"; f=1}
     1
     END{if(s&&!f)  print "  DUMP=example"; } blockname="name" <file>

Here, we did the following:

• The variable s checks if the correct block is found
• The variable f checks if in the correct block, the entry DUMP= is found
• /^\\[/&&/]$/&&s&&!f{print " DUMP=example"; s=0} /^\\[/&&/]$/&&s&&!f{print " DUMP=example"; s=0} :: If we enter a new block, and the previous block was the correct one ( s==1 ) and that block did not have a DUMP= entry ( f==0 ), write one.
• s&&/^ *DUMP=/{$0=" DUMP=example"; f=1} s&&/^ *DUMP=/{$0=" DUMP=example"; f=1} :: If we are in the correct block ( s==1 ), and we find a DUMP= entry, update it.