Find all lines between two patterns, EXCLUSIVE of the second pattern?

Question

Consider that I have the file listed below. I need to select all lines from every instance of the regex pattern Word A to before the regex pattern Word D.

Word A
Word B
Word C
Word D
Word E
Word F
Word G
Word A
Word H
Word I
Word D
Word J
Word A
Word K
Word D
Word L
Word M
Word A
Word D

Note the variable number of rows between A and D. Sometimes, D is the very next row. Here's what I need the output to be:

Word A
Word B
Word C
Word A
Word H
Word I
Word A
Word K
Word A

Can be done with awk, perl, python, or sed. Doesn't matter as long as it's installed on the RHEL6 server where the file is.

Answer 1

Using AWK:

awk '/Word A/ { m = 1 } /Word D/ { m = 0 } m'

Answer 2

Here's an awk solution

awk \
  -vstart='Word A' \
  -vend='Word D' \
  '{
     if ($0==end  ) {flag=0;next};
     if ($0==start) {flag=1};
     if (flag==1) {print $0};
  }'

Only a minor change required for regex handling

awk \
  -vstart='Word[ ]A' \
  -vend='Word[ ]D' \
  '{
     if ($0 ~ end  ) {flag=0;next};
     if ($0 ~ start) {flag=1};
     if (flag==1) {print $0};
  }'

Answer 3

Using Raku (formerly known as Perl_6)

~$ raku -ne '.put if / Word \h A / fff^ / Word \h D /;'  file

Raku is a programming language in the Perl-family. It's an "operator-rich" language that features a powerful Regex engine. Above, the -ne non-autoprinting linewise flags are used, in conjunction with Raku's sed-like fff "Flip-flop" operator.

Raku includes various 'flavors' of its sed-like fff infix operator, including fff^, ^fff and even ^fff^. While each Regex is recognized, the ^ caret indicates that recognized line should be dropped from the output:

Sample Input:

Word A
Word B
Word C
Word D
Word E
Word F
Word G
Word A
Word H
Word I
Word D
Word J
Word A
Word K
Word D
Word L
Word M
Word A
Word D

Sample Output:

Word A
Word B
Word C
Word A
Word H
Word I
Word A
Word K
Word A

The above code solves the OP's test case. But what if the /start/ and /stop/ Regexes are actually on the same line? For that problem you could try Raku's awk-like ff operator:

~$ echo 'AB\nCD\nEF' | raku -ne 'say $_ if /A/ ff /B/;'
AB
~$ echo 'AB\nCD\nEF' | raku -ne 'say $_ if /A/ ff /C/;'
AB
CD

As compared to Raku's sed-like fff operator:

~$ echo 'AB\nCD\nEF' | raku -ne 'say $_ if /A/ fff /B/;'
AB
CD
EF
~$ echo 'AB\nCD\nEF' | raku -ne 'say $_ if /A/ fff /C/;'
AB
CD

https://docs.raku.org/routine/fff
https://docs.raku.org/routine/ff
https://raku.org

Answer 4

GNU sed only:

sed '/Word A/!d;:1;n;/Word D/d;b1' file

In more complex cases - invalid blocks:

sed -n '/Word A/!b;:1;/Word A/h;n;/Word D/{g;p;d};H;b1' file

Answer 5

TXR Lisp's awk macro supports this directly; the rng (range) operator has nine variants for various ways of excluding records from the start or end of a range:

$ txr -e '(awk ((rng- #/Word A/ #/Word D/)))' data
Word A
Word B
Word C
Word A
Word H
Word I
Word A
Word K
Word A

Also, unlike Awk's range operator, it combines with other operators. E.g. suppose you wanted to print records which are simultaneously in a foo to bar range, and in a start to end range, no matter how those kinds of ranges overlap in the data:

(awk ((and (rng #/foo/ #/bar/)
           (rng #/start/ #/end/))))

Answer 6

Using awk:

$ awk '
    $0 == "Word A" { f=1; rec=$0; next }
    { if ( $0 == "Word D" ) { print rec; f=0 } }
    f{rec = rec ORS $0}'

# For regex pattern
$ awk '          
    (/Word A/ && !/Word D/) { f=1; rec=$0; next }
    (/Word D/ && rec){ print rec; f=0; rec="" }
    f{rec = rec ORS $0}
'

If Word D matches Word A everytime, then the following command may be used.

$ awk '/Word A/,/Word D/ { if (!/Word D/) print }'

Answer 7

sed lets one do arithmetic on line specifications:

sed -n -e '/Word A/,/Word D/-1p' The_File

Read man sed.