This bash script takes links from an rss feed and sends .mobi files to an address (usually your kindle email) for easy reading. It has some Kippt specific tuneups, so might not easily work with other feeds.

send2kindle

#!/bin/bash

#this bash script takes links from an rss feed and sends
#.mobi files to an address (usually your kindle email)
#for easy reading.

#Usage is to set the rss url below and run the script

#Dependencies
# https://launchpad.net/ubuntu/precise/+package/html-xml-utils
# Calibre
# possibly more

# this script has Kippt specific tuneups. Might not work for other feeds
# creates files in the same directory. Leave them be.


### Define some constants
rssUrl=""

rssFile="rssFile"
newLinksFile="newLinks"
processListFile="processList"
processedListFile="processedList"
rawLinksFile="rawLinks"
downloadFolder="html"


### Define some functions

getRss () {
# Gets an rss feed and writes to a file
 rssUrl=$1
 rssFile=$2
 wget --no-verbose --output-document=$rssFile $rssUrl
}

extractLinks () {
# Extracts links from an rss file in a clean format and writes to file
rssFile=$1
linksFile=$2
rawLinksFile=$3

 hxextract link $rssFile > $rawLinksFile
 # 1. remove kippt.com link 2. remove the closing tag and put on new lines 3. remove the opening tag
 sed 's/<link>http:\/\/kippt.com<\/link>//' $rawLinksFile | sed 's/<\/link>/\n/g' | sed 's/<link>//g'  > $linksFile
}

isProcessed () {
# Checks if the url is present in the file
 url=$1
 linksFile=$2

 alreadyProcessed="0"
  while read aurl; do
   if [ $url == $aurl ]
    then
     alreadyProcessed="1"
   fi
  done < $linksFile
}

addProcesses () {
# Takes urls in newlinks file and adds them to the processList file
# if not already not in the processedList file. 
 newLinksFile=$1
 processListFile=$2
 processedListFile=$3
 while read newurl; do
  isProcessed $newurl $processListFile
  if [ $alreadyProcessed == "0" ]
   then
      isProcessed $newurl $processedListFile
    if [ $alreadyProcessed == "0" ]
     then
      echo "$newurl" >> $processListFile
    fi
  fi
 done < $newLinksFile
}

processUrl () {
# Takes a url and tries to send it to the kindle
# exit code is 0 only if sent
 url=$1
 #Get the page off the web
 echo "Getting url $url"
 wget -e robots=off --no-verbose --page-requisites --convert-links --directory-prefix=html --adjust-extension --html-extension --no-directories  --ignore-tags=a,area,iframe,script $url
# if [ $? != 0 ]
#  then
#   echo "Failed to get link $url"
#   exit 1
# fi

 #Get the html name
 htmlFileName=$(ls $downloadFolder | grep .htm)
 oneFileCheck=$(ls $downloadFolder | grep $htmlFileName)
 if [ ! $oneFileCheck ]
  then
   echo "Leads to multiple html files $url"
   exit 2
 fi
 htmlFilePath="./$downloadFolder/$htmlFileName"
 echo "Path to html file is $htmlFilePath"

 #Convert it to .mobi
 echo "Begin conversion to .mobi"
 title=$(hxextract title $htmlFilePath | sed 's/<[^>]*>//g' | sed 's/[!@#\$%^&*()".:?;]//g' | sed "s/'//g" |  sed ':a;N;$!ba;s/\n/ /g' | sed 's/ //g')
 mobiFile="$title.mobi"
 echo "Mobi file will be name $mobiFile"
 ebook-convert $htmlFilePath $mobiFile --output-profile=kindle --dont-compress --no-inline-toc --pretty-print --max-levels=0

 if [ $? != 0 ] 
  then
   echo "ebook-convert failed on $url"
   exit 3
 fi

 #Send to kindle
 echo "Sending to your Kindle"
 calibre-smtp --relay smtp.gmail.com \
              --port 587  \
              --username abd.kindle \
              --password testable \
              --attachment $mobiFile \
              --subject "" \
              [email protected] \
              [email protected] \
              "."

 if [ $? != 0 ]
  then
   echo "Could not send by email $url"
   exit 4
 fi
 echo "Done"

 # fix this
 rm -r html
# rm $mobiFile
}

runProcesses () {
# Takes a process list and processes all links in it
 processListFile=$1
 processedListFile=$2
 while read url; do
  processUrl $url
  # if successful, add to processed list
#  echo "$url"
  if [ $? == 0 ]
   then
    echo "$url" >> $processedListFile
    sed -i "\;$url;d" $processListFile
  fi
 done < $processListFile
}

### Begin script
echo "Let's start..."

# Make sure files exist
echo "Creating files needed"

if [ ! -f $processListFile ]
 then
  printf "%s" "" >> $processListFile
fi

if [ ! -f $processedListFile ]
 then
  printf "%s" "" >> $processedListFile
fi

rm -r $downloadFolder

# Get rss file
echo "Getting rss"
getRss $rssUrl $rssFile

# if fail to get file, exit
if [ $? != 0 ]
 then
  echo "Failed to get rss file"
  exit 1
fi

# Process the rss to extract the links
echo "Extracting links"
extractLinks $rssFile $newLinksFile $rawLinksFile

# Add links to to be processed list
echo "Finding new links"
addProcesses $newLinksFile $processListFile $processedListFile

# Now start processing
echo "Begin run processes"
runProcesses $processListFile $processedListFile