Custom Line Number/Prefix for text file using awk/sed

Before raising the flag I tried couple of things but didn't get the desired result.

I have many text files which I want prefixed with custom numbering;

   the first 2 lines should be prefixed  00A:  , 00B:  
   and remaining lines should be incremental, like 001:  ,002:  ,003:  , and so on

Presently, I use this command for incremental numbering

awk '{printf("%03d:  %s\r\n", NR,$0)}' file1.txt > file2.txt

*which does the incremental nos. ok for the whole file; but not the multiple types needed.

Example Input file:

136725A6449C5279 933FB466C9CD699B
8FFBBA87E9D3209A AB41FBDC5E281A92
FF80DA7B0054FB29 006BF1C82C75C341
FA118264221B02A7 81E9A1FEB75FFB3D
31AA9FC566C3ADE0 70DDFD6DED2BF29C
F0B39014DA7FA6B1 77401108A81E33E1
74EF54060BC2B72F B5518D896DDC266F
DE10C97F9FBDA5A6 6C79566CA1BDC06E

Desired Output:

00A:  136725A6449C5279 933FB466C9CD699B
00B:  8FFBBA87E9D3209A AB41FBDC5E281A92
001:  FA118264221B02A7 81E9A1FEB75FFB3D
002:  31AA9FC566C3ADE0 70DDFD6DED2BF29C
003:  F0B39014DA7FA6B1 77401108A81E33E1
004:  74EF54060BC2B72F B5518D896DDC266F
005:  DE10C97F9FBDA5A6 6C79566CA1BDC06E

Solution

Neither Awk nor sed does this very well, but Perl has it built in.

perl -pe 'BEGIN { $prefix = "A"; }
  $prefix = "1" if ($. == 3);
  printf "%03s:  ", $prefix++;' file

The crucial feature here is that in Perl, "A"++ produces "B", natively. It doesn't work so well with leading zeros, though; so I resorted to padding here.

Your question is rather unclear as to what should happen after 00Z or after 009 so I had to guess. For what it's worth, in Perl, "Z"++ is "AA".

If you really insist on an Awk solution, it can be done with something like this:

awk '{ printf("%03s:  %s\n", (NR == 1 ? "A" : \
    (NR == 2 ? "B" : NR-2)), $0)}' file

I took out the fugly \r; if you are on Windows, maybe put it back (or consider your options).

As noted in comments, this works on MacOS / nawk but might not on other Awks.