Find the every nth occurrence of a string

Despite the many similar questions, I am still confused.

qx|o1|md|4SAK9H7DQ876CAJ943,SJT8642H2DKT2CQ76,SQ5HKT653DAJCKT52|rh||ah|Board 1|sv|0|pg||
qx|o2|md|4SKQJ7642H9DQJ8CK8,STHAK8762D74CQJ65,SA85HQJTDAK96CT92|rh||ah|Board 2|sv|0|pg||
qx|o3|md|4ST3HAT9DAK96CA983,S76HK864D732CJT74,SAQJ82HJDQJT85CKQ|rh||ah|Board 3|sv|0|pg||
qx|o4|md|4SAQ4HT65432DAJ4CJ,SJT765HAKDT982C43,SK98HQJ9DK75CAQ92|rh||ah|Board 4|sv|0|pg||

In the above text, how do I find the every second occurrence of 4S, and replace it with 3S? (Or, in every other line, 4S becomes 3S.) (It is a given that the string 4S occurs only once in every line.) I working in Sublime Text, or Notepad++ on Windows.

2 Answers

Using Notepad++:

• Ctrl+H
• Find what: .*?4S.*?\K4S
• Replace with: 3S
• CHECK Match case
• CHECK Wrap around
• CHECK Regular expression
• CHECK . matches newline
• Replace all

Explanation:

.*? # 0 or more any character, not greedy
4S # literally 4S, first one
.*? # 0 or more any character, not greedy
\K # forget all we have seen until this position
4S # literally 4S, second one

Screenshot (before):

Screenshot (after):

I guess you'll need something like sed or awk to pull that off. On a Bash shell in Linux I'd use sed with the following commads:

sed -e 'N; s/4s/3s/2' < file.txt

whereby file.txt should contain your input lines.

When started, sed will
- pull the first line into the pattern space of sed,
- N will pull the next (second) line into the pattern space and
- `s/4s/3s/2' will then replace the second occurrence of "4s" in the pattern space with "3s".
After that the two lines are written to the standard output and a new cycle of sed begins.

Obviously, you also can pipe the output of a previous command through that.

'command' | sed -e 'N; s/4s/3s/2'