Split.sh

#!/usr/bin/env bash

# http://davidbosman.fr || david@davidbosman.fr || août 2012.
# Feel free to modify and repost.
#
# This script is the first of two that will import a bunch of text files 
# formated in Markdown as Day One entries.
# In order to work, each file name must start with AAAA-MMDD (wich will be as 
# the date of the entries) and contain only one entry. So if, like me, you 
# used to write a day worth of entries in a single file, like this:
#
# 2012-0816-J-my daily diary.txt:
#
# ### 17H32
# Some text. 
#
# ### 23h41
# Some more text.
#
# ![alt text](link to an image)
#
# Some more text.
#
# ### some title but no time stamp
# Some text, again.
#
# this script will look for occurences of ## or ### (or none, at the first 
# line of the file) and, for each one consisting of a time stamp, will create 
# a new file. (other ### will be added to the previous time stamp). 
# So the example will look like this after spliting:
#
# 2012-0816-J-my daily diary1.txt:
#
# ### 17H32
# Some text. 
#
# 2012-0816-J-my daily diary2.txt:
#
# ### 23h41
# Some more text.
#
# ![alt text](link to an image)
#
# Some more text.
#
# ### some title but not a time stamp
# Some text, again.
#
# Each new file will be created in a temporay export folder 
# ("~/Desktop/output"). This folder *must* exist before running the script.
# 
# The second script, Import2DayOne.sh, will deal with the import process 
# itself, using Day One CLI.


for i in *
	do
		time=`awk '/^##[#]?[ ]?[0-9][0-9]?[h:]/ {print $2}' "$i"`
		if [ "$time" != "" ]			
		then
		# CHANGE PATH if you want to. 
		# Please create the "ouptput" folder before running the script:		
			perl -n -e "\$p++; if(\$p==1 || /^##[#]?[ ]?[0-9][0-9]?[h:]/) { open FH, \">$HOME/Desktop/output/$i\".++\$n.\".txt\"; } print FH;" "$i"
		else
			# CHANGE PATH if you want to. 
			# Please create the "ouptput" folder before running the script:
			cp "$i" "$HOME/Desktop/output/$i";
		fi
		
done