awk合并基于第一个字符的行

Question

I am trying to write a command to convert a texter bundle file into an autokey file. The texter file is formatted like so:

.abbr
This would be the replacement text for the line above in texter.
.day
Have a GREAt day!
.but
But some of the 
information takes up multiple lines
.how
However all the abbreviations 
start with a "." and 
then x lines of text to be

I need to put each one in a separate file or print the abbrevation and its replacement text into the appropriate fields in the following format:

  "items": [
{
    "usageCount": 0, 
    "omitTrigger": false, 
    "prompt": false, 
    "description": "description", 
    "abbreviation": {
        "ignoreCase": false, 
        "wordChars": "[\\w]", 
        "immediate": false, 
        "abbreviation": "ABBREVIATION GOES HERE", 
        "backspace": true, 
        "triggerInside": false
    }, 
    "hotkey": {
        "hotKey": null, 
        "modifiers": []
    }, 
    "phrase": "REPLACEMENT TEXT GOES HERE", 
    "modes": [
        1
    ], 
    "showInTrayMenu": false, 
    "matchCase": false, 
    "filter": null, 
    "type": "phrase", 
    "sendMode": "kb"

I am pretty sure I can handle the formatting part with awk but How could I grab the proper abbrevation and its appropriate replacement text. Heres what I came up with for the printing it to individual files that I could them format but its giving me an error about close f:

awk '/^\./ {f="$title\."++d} f{print > f} /^\./ {close f; f=""}' 

This script wont work either that I wrote:

#!/bin/sh

while read line
do
   if echo $line | grep "^\."; then
       line_no=`echo $line | awk -F "\." '{ print $2 }'
   elif echo $line | grep "^\."; then
       : #ignore
   else
       echo "$line" >> t2ak.$line_no
   fi
done < Default.texter

Answer 1

This might work for you:

csplit -n5 -z file '/^\./' '{*}'

This will place each abbreviation in a separate file, then use sed or awk to fill in a template using the resulting files.

Answer 2

Try using "\\n\\." as awk 's recordseparator regex.