AWK - converting a paragraphs into comma delimited text tables

Author

converting a paragraphs into comma delimited text tables

ari.rennt@gmail.com

2005-11-12, 3:55 am

I would like to convert text files with the following format:

title1: paragraph1.
(space)
title2: paraphraph2

e.g.
ALLERGIES: He has no known drug allergies.

IMMUNIZATIONS: up to date.

BIRTH HISTORY: He was born via cesarean section secondary to repeat.
This is usually a paragraph of data with multiple lines. Sometimes the
paragraph text contains colons and numbers as well.

to something that looks like this

"ALLERGIES" "IMMUNIZATIONS" "BIRTH HISTORY"
"He has no known drug allergies. " "up to date." "He was born via
cesarean section secondary to repeat."

I've had very little success and would sincerely appreciate any
help/advice you fine folks could provide!

thanks
ari

Ed Morton

2005-11-12, 6:55 pm

ari.rennt@gmail.com wrote:
> I would like to convert text files with the following format:
>
> title1: paragraph1.
> (space)
> title2: paraphraph2
>
> e.g.
> ALLERGIES: He has no known drug allergies.
>
> IMMUNIZATIONS: up to date.
>
> BIRTH HISTORY: He was born via cesarean section secondary to repeat.
> This is usually a paragraph of data with multiple lines. Sometimes the
> paragraph text contains colons and numbers as well.
>
> to something that looks like this
>
> "ALLERGIES" "IMMUNIZATIONS" "BIRTH HISTORY"
> "He has no known drug allergies. " "up to date." "He was born via
> cesarean section secondary to repeat."
>
> I've had very little success and would sincerely appreciate any
> help/advice you fine folks could provide!
>
> thanks
> ari
>

You can tweak where the spaces appear on the printf and/or strip leading
blanks if you like:

$ awk -vRS="" -F: '{t[NR]=$1;$1="";r[NR]=$0}END{for(i=1;i<=NR;i++)printf
"\"%s\" ",t[i];print "";for(i=1;i<=NR;i++)printf "\"%s\" ",r[i];print
""}' file
"ALLERGIES" "IMMUNIZATIONS" "BIRTH HISTORY"
" He has no known drug allergies." " up to date." " He was born via
cesarean section secondary to repeat."

Regards,

Ed.

William James

2005-11-12, 6:55 pm

ari.rennt@gmail.com wrote:
> I would like to convert text files with the following format:
>
> title1: paragraph1.
> (space)
> title2: paraphraph2
>
> e.g.
> ALLERGIES: He has no known drug allergies.
>
> IMMUNIZATIONS: up to date.
>
> BIRTH HISTORY: He was born via cesarean section secondary to repeat.
> This is usually a paragraph of data with multiple lines. Sometimes the
> paragraph text contains colons and numbers as well.
>
> to something that looks like this
>
> "ALLERGIES" "IMMUNIZATIONS" "BIRTH HISTORY"
> "He has no known drug allergies. " "up to date." "He was born via
> cesarean section secondary to repeat."
>
> I've had very little success and would sincerely appreciate any
> help/advice you fine folks could provide!
>
> thanks
> ari

Save as "convert.awk" and run with
gawk --re-interval -f convert.awk myfile.txt >outfile.txt

BEGIN {
Width = 70
if ( "j" !~ /j{1}/ )
{ print "Must run with 'gawk --re-interval'."
exit
}
# A blank line may contain blanks.
RS = "\n([ \t]*\n)+"
}

match($0, /: */) {
header = header " \"" substr($0,1,RSTART-1) "\""
text = text " \"" strip( substr($0,RSTART+RLENGTH)) "\""
}

END {
display( header )
display( text )
}

function display( s )
{ s = strip( s )
gsub( /[ \t\n]+/, " ", s )
while ( length(s) > Width && match( s, /^.{1,70}[ \t]+/ ) )
{ print strip( substr( s, 1, RLENGTH ) )
s = substr( s, RLENGTH + 1 )
}
if ( s ) print s
}

function strip( s )
{ gsub( /^[ \t]+|[ \t]+$/, "", s )
return s
}

Sponsored Links