DataHoarder_scripts/lego_dl.sh

 #!/bin/bash
 ##############################
 #
 # This scripts uses the brickset instructions file to get links and set numbers.
 # In order to get themes, we get themes from rebrickable. 
 # If a booklet isn't available on lego.com, then brickinstructions.com is tested. 
 # 
 # Files are saved as:
 # $tID""$setDESC""${setNAME// /_}"_("$setTHEME"_"$setYEAR").pdf"
 # 1190-1_Retro_Buggy_(Town_1999).pdf
 # 
 # If a set contains multiple filesm the file name will be:
 # 2520-1_(1_of_2)_Battle_Arena_(Ninjago_2011).pdf
 # 2520-1_(2_of_2)_Battle_Arena_(Ninjago_2011).pdf
 # 
 # Req:
 # bash, awk, grep, sed, curl, wget
 #
 # Get themes.csv and sets.cvs from https://rebrickable.com/downloads
 # 
 # error_level: 
 #	0 no errors reported
 #	1 Download errors reported
 #	2 Existing files and download errors reported
 #
 ##############################

 echo "----> Starting Download of all LEGO instructions from https://brickset.com/exportscripts/instructions"
 firstline=0

 ##### CHANGE HERE #####
 logfile="lego_errors.log"
 downloadFolder="../Instructions"
 error_level=1
 ##### STOP CHANGE #####

 echo "" > $logfile

 THEMES="themes.csv"
 SETS="sets.csv"

 if [ -f "instructions" ]; then
	 echo "---> Using existing instructions file. Manually delete it to redownload..."
 else 
	 echo -ne "---> Downloading csv from Brickset..."
	 wget https://brickset.com/exportscripts/instructions &> /dev/null  
	 if [[ "$?" != 0 ]]; then
		 echo "Error... Exiting!"
		 exit; 
	 else
		 echo "Done!" 
	 fi
 fi


 echo "----> Converting csv with ',' to '|'..."
 head -1 instructions | sed 's/,/|/g' > instructions.csv
 sed 1,1d instructions | sed -r 's/("[^",]+),([^",]+")/\1###\2/g' | awk -F, '{print $1,$2,$3,$4,$5,$6}' | sed 's/###/,/g' | sed 's/" "/"|"/g' >> instructions.csv

 LINES=$(wc -l instructions | awk '{ print $1 }')
 CURRENT=0

 function validate_url(){
	if [[ `wget -S --spider $1  2>&1 | grep 'HTTP/1.1 200 OK'` ]]; then
		return 0
	else
		return 1
	fi 
 }  

 while IFS='|' read -r ID LINK NAME DESC ADDED RETRIVED
 do

	#$PC=$((200*$CURRENT/$LINES % 2 + 100*$CURRENT/$LINES))
	#CURRENT=$((CURRENT + 1))
	if [ "$firstline" = 0 ]; then
		firstline=1
	else			
		tID=$(sed -e 's/^"//' -e 's/"$//' <<<"$ID")
		tLINK=$(sed -e 's/^"//' -e 's/"$//' <<<"$LINK")
		#tNAME=$(sed -e 's/^"//' -e 's/"$//' <<<"$NAME")
		ttNAME=$(cut -d, -f1-2 sets.csv | grep -w $tID | cut -d, -f2)
		setYEAR=$(grep -w $tID sets.csv | cut -d, -f3)
		setNAME=$(sed -e 's/[^A-Za-z0-9._-]/_/g' <<< $ttNAME)
		themeID=$(grep -w $tID sets.csv | cut -d, -f4)
		themeName=$(awk -F',' -v id="$themeID" '$1 == id' themes.csv)
		IFS=',' read -r -a array <<< "$themeName"  
		tempID=${array[2]}
		if [[ $tempID != "" ]]; then
			while [[ $tempID != "" ]]  
			do 	 
				tthemeName=$(awk -F',' -v id="$tempID" '$1 == id' themes.csv)
				IFS=',' read -r -a tArray <<< "$tthemeName"
				tempID=${tArray[2]}
				themeName=${tArray[1]}
			done 
		else  
			themeName=${array[1]} 
		fi

		setTHEME=$(sed -e 's/[^A-Za-z0-9._-]/_/g' <<< $themeName)
		 
		tADDED=$(sed -e 's/^"//' -e 's/"$//' <<<"$ADDED")
		tDESC=$(sed -e 's/^"//' -e 's/"$//' <<<"$DESC")

		ttDESC=$(echo $tDESC | grep -Eo '[^0-9][0-9]{1}\s?\/\s?[0-9]{1,2}' | sed 's/[^A-Za-z0-9/]//g' | sed 's/\//_of_/g')
		#ttDESC=$(echo $tDESC | grep -Eo '\s[0-9]{1}\s?\/\s?[0-9]{1,2}' | sed 's/ //g' | sed 's/\//_of_/g')

		if [ -z "$ttDESC" ]; then
			etDESC="_"
		else
			tDESC="_("$ttDESC")_"
		fi
		#PC=$(echo $CURRENT $LINES | awk '{print 100*$1/$2}')
		tFilename=""$tID"_"$setDESC""${setNAME// /_}"_("$setTHEME"_"$setYEAR").pdf"
		filename=$tFilename

		if [ -f "$downloadFolder/$filename" ]; then
			if [[ $error_level = 2 ]]; then
				echo "-> $tID exists. Skipping..."
				echo "$filename exists." >> $logfile
			fi 	
		else
			if [[ "$tDESC" = "{No longer listed at LEGO.com}" ]] ; then
				echo -ne "-> $tID testing links..."
				if validate_url $tLINK; then     
					echo -ne "Found on LEGO.com... Downloading..."   
					curl -H "Mozilla/5.0 (platform; rv:75.0) Gecko/20100101 Firefox/75.0" -L $tLINK --silent --output "$downloadFolder/$filename"
					if [ -f "$downloadFolder/$filename" ]; then
						echo "Done! > $filename"
					else
						echo "ERROR!"
						if [[ $error_level = 1 || $error_level = 2 ]]; then  
							echo "--> Not downloaded. Try again manually..."
							echo "$filename was not downloaded. Check CURL" >> $logfile
						fi
					fi
					#random sleep in order to not look like a script
					sleep $(( ( RANDOM % 5 )  + 1 ))
				else
					#test

					biID=$(sed -e 's/[^0-9_]/_/g' <<< $tID)
					biLink="https://lego.brickinstructions.com/pdfdrop/"
					if validate_url "$biLink$biID.pdf"; then
						echo -ne "Found on BrickInstructions.com... Downloading..."
						curl -H "Mozilla/5.0 (platform; rv:75.0) Gecko/20100101 Firefox/75.0" -L "$biLink$tID.pdf" --silent --output "$downloadFolder/$filename"
						if [ -f "$downloadFolder/$filename" ]; then
							echo "Done! > $filename"
						else
							echo "ERROR!"
							if [[ $error_level = 1 || $error_level = 2 ]]; then  
								echo "--> Not downloaded. Try again manually..."
								echo "$filename was not downloaded. Check CURL" >> $logfile
							fi
						fi
						#random sleep in order to not look like a script
						sleep $(( ( RANDOM % 5 )  + 1 ))
					else
						if [[ $error_level = 1 || $error_level = 2 ]]; then
							echo "-> $tID is not available. Skipping..."
							echo "$filename is not available." >> $logfile
						fi
					fi
				fi
				
			else  
				echo -ne "--> $tID downloading now..."
				curl -H "Mozilla/5.0 (platform; rv:75.0) Gecko/20100101 Firefox/75.0" -L $tLINK --silent --output "$downloadFolder/$filename"
				if [ -f "$downloadFolder/$filename" ]; then
					echo "Done! > $filename"
				else
					echo "ERROR!"
					if [[ $error_level = 1 || $error_level = 2 ]]; then  
						echo "--> Not downloaded. Try again manually..."
						echo "$filename was not downloaded. Check CURL" >> $logfile
					fi
					#random sleep in order to not look like a script
					sleep $(( ( RANDOM % 5 )  + 1 ))
				fi
			fi
		fi
	fi
done < instructions.csv