2

My File is:

"abc"..,,xyz 123 "New york".."END" <- # Print this entire line as starts with " and ends with "END"
"mnh".....blahblah                 <- # dont print this line and hold til "END" is
"rew"..?/.."1324.75 United            # found and once it is found, merge multi-
 States"??..."END"                    # lines to single line without changes

Desired output:

"abc"..,,xyz 123 "New york".."END"
"mnh".....blahblah"rew"..?/.."1324.75 United States"??..."END"

Found the below sed which does everything with start_pattern as " and end_pattern as "END", but it breaks when intermediate line starts with "; here when it starts with "rew", before the end_pattern is found.

sed -n '/^"/,/^"END"/{
        //!{H;/{x;s/\n\([^\n]*\)$/\1/;x}};
        /^"/{h};/"END"/{x;p};d
        }' file

Link: How to remove newlines between data of each record which are located between 2 patterns?

Improve this question
0

2 Answers 2

Reset to default
3 
awk -v RS='"END"\n' -v FS='\n' -v ORS='"END"\n' -v OFS=' ' '{ $1=$1 };1' file

or the equivalent

awk 'BEGIN { ORS=RS="\"END\"\n"; FS="\n" } { $1 = $1 }; 1' file

(which uses the fact that OFS=' ' is the default anyway).

This is a really short awk command that simply reformats your input from one set of record and field separators to another set of record and field separators. Most of the magic is how the input record and field separators (RS and FS) are changed into new output separators (ORS and OFS).

Your input document,

"abc"..,,xyz 123 "New york".."END"
"mnh".....blahblah
"rew"..?/.."1324.75 United
States"??..."END"

uses "END"\n as the record separator, and \n (newline) as field separator.

Changing the field separator to (a space) would be enough to solve your issue.

The actual code just makes awk reformat the record so that the new field separator (OFS) is used in place of the old (FS).

The result would be

"abc"..,,xyz 123 "New york".."END"
"mnh".....blahblah "rew"..?/.."1324.75 United States"??..."END"

Technically, using a multi-character value for RS invokes unspecified behavior according to the POSIX specification for the awk utility. Most (all?) current awk implementations treat a multi-character value in RS as a regular expression.

Improve this answer
0
1

You can join a line to the following one if it does not end in "END" using sed as follows:

sed -e :a -e '/"END"$/!N; s/\n//; ta' file

Ex. given file as

"abc"..,,xyz 123 "New york".."END"
"mnh".....blahblah
"rew"..?/.."1324.75 United
States"??..."END"

then

$ sed -e :a -e '/"END"$/!N; s/\n//; ta' file
"abc"..,,xyz 123 "New york".."END"
"mnh".....blahblah"rew"..?/.."1324.75 UnitedStates"??..."END"

Note that (in line with your desired output) this does not add a space in place of the newline when it makes the join - if you actually want the space, then simply change s/\n// to s/\n/ /

Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.