Atom regexp: discarding multiline text around blocks - regex

Question

Assume I have this text:
blah blah Bob Loblaw Law blah
keep1 { i want this } blop
blah blob keep2 { and
this too } blaw blat
etc...
And I want to end up with
keep1 { i want this }
keep2 { and
this too }
or perhaps:
keep1 { i want this }
keep2 { and this too }
I haven't figured out how to get Atom's regexp find/replace mechanism to discard everything across multiple lines outside of a specific matching string. Hints?
update:
Of the many things I've tried, this gets me closest:
[\S\s]+?(keep\d\s+\{[\S\s]+?\})
which results in:
keep1 { i want this }
keep2 { and
this too }
blaw blat
etc...
This is probably good enough -- I can edit the trailing shards -- but it would be useful to know how to trim those as well.

Answer 1

You may use this simple regex replace in Atom for this task:
\b(keep\d+\s*{[^}]*})|.+?
Replace it with: $1
RegEx Demo
RegEx Details:
\b: Word boundary
(keep\d+\s*{[^}]*}): In capture group #1 match a string that starts with keep followed by 1+ digits followed by 0+ whitespaces followed by any text that is inside {...} spanning across the lines as well. This assumes { and } are balanced and there is no escaping of { and }.
|: OR
.+?: Lazily match 1+ of anything
PS: If you want to remove leading line break then use:
\n?\b(keep\d+\s*{[^}]*})|.+?
Atom Editor Demo
Before replacement:
After replacement:

Answer 2

Use
[\s\S]*?(keep\d\s+\{[^{}]*\})|(?:(?!keep\d\s+\{[^{}]*\})[\s\S])+$
See proof.
EXPLANATION
--------------------------------------------------------------------------------
[\s\S]*? any character of: whitespace (\n, \r, \t,
\f, and " "), non-whitespace (all but \n,
\r, \t, \f, and " ") (0 or more times
(matching the least amount possible))
--------------------------------------------------------------------------------
( group and capture to \1:
--------------------------------------------------------------------------------
keep 'keep'
--------------------------------------------------------------------------------
\d digits (0-9)
--------------------------------------------------------------------------------
\s+ whitespace (\n, \r, \t, \f, and " ") (1
or more times (matching the most amount
possible))
--------------------------------------------------------------------------------
\{ '{'
--------------------------------------------------------------------------------
[^{}]* any character except: '{', '}' (0 or
more times (matching the most amount
possible))
--------------------------------------------------------------------------------
\} '}'
--------------------------------------------------------------------------------
) end of \1
--------------------------------------------------------------------------------
| OR
--------------------------------------------------------------------------------
(?: group, but do not capture (1 or more times
(matching the most amount possible)):
--------------------------------------------------------------------------------
(?! look ahead to see if there is not:
--------------------------------------------------------------------------------
keep 'keep'
--------------------------------------------------------------------------------
\d digits (0-9)
--------------------------------------------------------------------------------
\s+ whitespace (\n, \r, \t, \f, and " ")
(1 or more times (matching the most
amount possible))
--------------------------------------------------------------------------------
\{ '{'
--------------------------------------------------------------------------------
[^{}]* any character except: '{', '}' (0 or
more times (matching the most amount
possible))
--------------------------------------------------------------------------------
\} '}'
--------------------------------------------------------------------------------
) end of look-ahead
--------------------------------------------------------------------------------
[\s\S] any character of: whitespace (\n, \r,
\t, \f, and " "), non-whitespace (all
but \n, \r, \t, \f, and " ")
--------------------------------------------------------------------------------
)+ end of grouping
--------------------------------------------------------------------------------
$ before an optional \n, and the end of the
string