Using sed to replace only the first two regex catches with different strings

Question

So I'm trying to change two instances of a regex pattern to two different strings with sed like:

sed -E "s/\\t([A-Za-z]|[-])+$/\\t`get_addr.py -1`/" /path/to/file.ext
sed -E "s/\\t([A-Za-z]|[-])+$/\\t`get_addr.py -2`/" /path/to/file.ext

I've tried a lot of things, and done a solid amount of manpage scouring, but I just can't quite figure out how make it happen.

Is there a way I can tell sed to replace each match with a different string with something like

sed -E "s/\\t([A-Za-z]|[-])+$/[1st replacement],[2nd replacement]/" /path/to/file.ext

and/or is there a way I can specify in the regex that I'm looking for the nth match, so that I can just call sed twice?

Answer 1

At least in GNU sed, I think you can do something like

sed -e '0,/pattern/ s//replacement1/' -e '0,/pattern/ s//replacement2/' file

e.g. given a file

abc
def
abc
ghi
abc
jkl

then

$ sed -e '0,/abc/ s//XYZ/' -e '0,/abc/ s//UVW/' file
XYZ
def
UVW
ghi
abc
jkl