field ordering with sed

Question

I have a bash script with a sed command that I want to run on a csv file to change the order of some fields. This is what I tried:

sed -r '{
   s/(.*),(.*),(.*),(.*),(.*),(.*),(.*),(.*),(.*),(.*),(.*)/\1,\2,\3,\4,\5,\6,\8,\9,\10,\11,\7/
}' $1

The problem comes because I have 11 fields (i.e. two digit number) so when I specify the desired order i.e.

"\1,\2,\3,\4,\5,\6,\8,\9,\10,\11,\7/"

10 and 11 are taken as literals and this ruins my attempt. I have tried the obvious alternatives like:

"\1,\2,\3,\4,\5,\6,\8,\9,(10),(11),\7/"

or

"\1,\2,\3,\4,\5,\6,\8,\9,{10},{11},\7/"

or

"\1,\2,\3,\4,\5,\6,\8,\9,\(10),\(11),\7/"

or

"\1,\2,\3,\4,\5,\6,\8,\9,\{10},\{11},\7/"

But none of these work, they are also treated as literals. I am running out of imagination, do not know what else to try, any ideas?

I know there are other ways of going about this problem (like awk etc), but I would appreciate if your answers are focused in sed since the rest of my code is done using sed.

Wish to thank you all in advance!

Answer 1

I see three blocks: fields 1-6, field 7 and the rest. So you can use

sed -r '{s/^(([^,]*,){6})([^,]*),(.*)/\1\4,\3/}'