I need a pattern for sed/grep/perl, any will do, for...
0001 Host Software. S. Crocker. April 1969. (Format: TXT=21088 bytes) (Status: UNKNOWN) 0002 Host software. B. Duvall. April 1969. (Format: TXT=17145 bytes) (Status: UNKNOWN) 0003 Documentation conventions. S.D. Crocker. April 1969. (Format: TXT=2323 bytes) (Obsoleted by RFC0010) (Status: UNKNOWN)
...where each description after the four digit number can be extracted. I've tried removing duplicate newlines, and other white space tricks with tr but no luck so far. If everything were on the same line it would be a snap but I'm unsure how to continue with this multiline pattern. Couldn't find a pattern that worked searching every character up to "\n\n", the next entry.