Simple bash script to get the summed run time of all titles on an imdb.com user list with help of the oMDB API and some open source software

summate-imdb-list-runtime.sh

#!/bin/bash

# Dependencies (you must have these already installed on your system)
#  - jq, JSON processor written in C
#  - bc, arbitrary precision numer processor
#  - curl, for transferring data given an URL

if [ -z $1 ] ; then 
	echo 'Pass an argument of the imdb.com list ID i.e. ls123456789 found in the URL'
	exit 1
fi

SourceOut=./imdb-source
IdsOut=./imdb-ids
ImdbListId=$1
StartOfList=1
Furl="http://www.imdb.com/list/${ImdbListId}/?start=${StartOfList}&view=compact&sort=listorian:asc&defaults=1"

if [ -e $IdsOut ] ; then
	echo 'Capture file already exists'
	exit 1
fi

echo 'Getting index to figure out number of pages on the IMDB list'
curl -L -s "$Furl" > $SourceOut

echo 'Collecting IDs of titles on IMDB from all list pages'
Pages=$(cat $SourceOut | grep -E 'Page 1 of [0-9]' | tail -n 1 | sed -r 's/Page 1 of ([0-9])/\1/')
CurPage=1
while [ $CurPage -le $Pages ] ; do
	if [ $CurPage -eq 1 ] ; then
		cat $SourceOut | grep -Eo 'href="\/title\/tt[0-9]{7}\/"' | sed -r 's/.*(tt[0-9]{7})\/"/\1/' | sort | uniq >> $IdsOut
	else
		StartOfList=$[$StartOfList + 250]
		Furl="http://www.imdb.com/list/${ImdbListId}/?start=${StartOfList}&view=compact&sort=listorian:asc&defaults=1"
		curl -L -s $Furl | grep -Eo 'href="\/title\/tt[0-9]{7}\/"' | sed -r 's/.*(tt[0-9]{7})\/"/\1/' | sort | uniq >> $IdsOut
	fi

	CurPage=$[$CurPage + 1]
done

echo 'Querying oMDB API for title metadata (this will take a long time for long lists)'
TotalRunTime=0
for ImdbId in `cat $IdsOut` ; do
	JsonResponse=$(curl -s "http://www.omdbapi.com/?i=${ImdbId}&plot=short&r=json")
	RunTime=$(echo $JsonResponse | jq -r '.Runtime' | cut -d" " -f1)
	TotalRunTime=$[$TotalRunTime + $RunTime]
done

TotalDays=`echo "scale=2; ${TotalRunTime}/60/24" | bc`
echo -e "\nYou have spent $TotalDays days watching movies!"