Hacks Push files to Kindle (unattended, untethered)

[kfried]

I have a Calibre instance running in my home server that's downloading my various subscriptions into local .mobi files for local consumption. I am then able to fire up my kindle browser and download these through the Kindle's browser and read on the Kindle. (I drop them on my http server that's accessible)

What I really want is the "push" functionality from this server to the Kindle and have it appear in the list automagically. I know the public IP of the Kindle (through looking at my server logs). Can I push to a running service on the Kindle (sshd?) from my home servers? Can I set up a daily job on the Kindle to download a specified ftp or http source? Can I set up a pop3 polling job on the Kindle to get these files? (would need to email them to a box) etc etc

[BarryTX]

You can set up Calibre to email them to your kindle email account, which then syncs them to the Kindle. Calibre can be set up to download news and automatically email to the kindle email account as well. Of course, sending anything through the kindle email account is not free, Amazon charges for the service of syncing it wirelessly to the device. I'm not aware of any way to send them around the USB or kindle account, but I'd love to see someone come up with a hack to do so.

[NiLuJe]

That may be more or less doable, but that would probably break Amazon's terms, because you'd effectively be 'hijacking' the 3G connection/amazon's e-mail delivery service.

[susan_cassidy]

If you are pushing these files to an http server, why not have the http server just provide the list of available files. Then, when you go to that web page, the files show up.

The only other way to "magically" get the files onto the Kindle would be via 3G.

If you connected via USB, your processing job could just copy the files onto the Kindle.

[mike_bike_kite]

Using Calibre:
You'd need Calibre to be running permanently on your PC and you'd need to schedule various rss news feeds ie just enter the url of the feed and how often you want Calibre to read it. Then under "Calibre > preferences > email delivery" just enter your Kindle email and the feeds should appear magically on your Kindle.

Using Feedbooks:
feedbooks allows you to create a newspaper of various rss feeds (your subscriptions) and it then packages them into a mobi document. You initially download the first copy via usb cable but you can have the newspaper emailed to your Kindle email account on a regular basis. If you don't want the charge then there is a link in the mobi file to download the latest copy.

I've tried it with mixed results but essentially it works. The creator of the site is a regular on the forum.

Note : Amazon imposes a small charge to send data to your device via email.

[kfried]

Thanks -- but I'm looking for the files to just magically appear in my list (no fees ); I currently already hitting my web server with the Kindle browser to download the files manually... but there's too many of them and it's tedious. So here's what I've done thus far, but have one hitch (at the end).

----------------
# Path to the destination for Books (.mobi)
DESTROOTPATH="/mnt/us/documents"
SOURCEROOTPATH="<MY PUBLIC URL TO MOBI FILES>"

# Get today's date (for file retrieval)
# DON'T RELY ON KINDLE DATE, USE A SERVER HACK FILE WITH "MY" DATE...
# TODAY=`date +%C%y-%m-%d`
wget -q "$SOURCEROOTPATH/today" -O today_file
TODAY=`cat today_file`
rm today_file

# Get the WSJ
FILENAME="The%20Wall%20Street%20Journal%20$TODAY.m obi"
SOURCEPATH="$SOURCEROOTPATH/$FILENAME"
DESTPATH="$DESTROOTPATH/$FILENAME"
if [ ! -f "$DESTPATH" ]
then
echo "Getting $FILENAME..."
wget -q "$SOURCEPATH" -P "$DESTROOTPATH/"
echo "DONE."
fi
----------------
I've put this in the /etc/crontab/root file to run every half hour and restarted the service of course. So I'm now able to get all of these files PULLED onto the Kindle.

PROBLEM: they don't appear in the Kindle Home even though they're in the right place. Browsing that dir it looks like .MBP files are generated alonside .MOBI files when they're downloaded through the browser. Any idea where the script / job is that does this generation work so I can add that to my script?

[mike_bike_kite]

Originally Posted by kfried

Thanks -- but I'm looking for the files to just magically appear in my list (no fees

I believe Amazon charges 10c per Mb on downloads but you'd have to check this. Each RSS feed is small so it should take a while to add up to a Mb and that 10c charge. I'm not sure how long you're planning on developing this application but you should keep in mind how much money you're saving - it will be cents rather than dollars

Of course if you're doing it for the challenge then I quite understand.

PS There is also a kindle development forum on the site.

[susan_cassidy]

I have never found that an .mbp file is required. If you copy a .mobi file to the Kindle, there is no .mbp file until you open the file. Maybe a permissions thing is causing them to not appear on your Home Page?

Edited to add: Are the files the right size?

I would remove the "echo" command from the script.

The charge is rounded up, per download, too, not cumulative. So, if they decide to charge you (since you are abusing the Whispernet privilege), every upload would be 15 cents (it's 15 cents per megabyte now, not 10).

[yifanlu]

Originally Posted by kfried

Thanks -- but I'm looking for the files to just magically appear in my list (no fees ); I currently already hitting my web server with the Kindle browser to download the files manually... but there's too many of them and it's tedious. So here's what I've done thus far, but have one hitch (at the end).

----------------
# Path to the destination for Books (.mobi)
DESTROOTPATH="/mnt/us/documents"
SOURCEROOTPATH="<MY PUBLIC URL TO MOBI FILES>"

# Get today's date (for file retrieval)
# DON'T RELY ON KINDLE DATE, USE A SERVER HACK FILE WITH "MY" DATE...
# TODAY=`date +%C%y-%m-%d`
wget -q "$SOURCEROOTPATH/today" -O today_file
TODAY=`cat today_file`
rm today_file

# Get the WSJ
FILENAME="The%20Wall%20Street%20Journal%20$TODAY.m obi"
SOURCEPATH="$SOURCEROOTPATH/$FILENAME"
DESTPATH="$DESTROOTPATH/$FILENAME"
if [ ! -f "$DESTPATH" ]
then
echo "Getting $FILENAME..."
wget -q "$SOURCEPATH" -P "$DESTROOTPATH/"
echo "DONE."
fi
----------------
I've put this in the /etc/crontab/root file to run every half hour and restarted the service of course. So I'm now able to get all of these files PULLED onto the Kindle.

PROBLEM: they don't appear in the Kindle Home even though they're in the right place. Browsing that dir it looks like .MBP files are generated alonside .MOBI files when they're downloaded through the browser. Any idea where the script / job is that does this generation work so I can add that to my script?

Does that actually work at downloading files? I thought the Kindle can only connect to Amazon's servers, anything else is denied (web browser is through amazon proxy)? And the reason you're not seeing the files is because the list doesn't auto-refresh. You need to either manually send a command to the framework (don't know if possible), or do something like buy a book from amazon or connect and eject the USB storage.

[duluoz]

Why don't you use calibre-smtp to email the mobi files to your free kindle address?

[daglitch]

Originally Posted by duluoz

Why don't you use calibre-smtp to email the mobi files to your free kindle address?

This is what I ended up using. Simple script and cron job = easy solution. Did a write up of this awhile back over at my site if anyone was interested. My deliveries are also free of charge using the kindles free email address.
http://getglitched.com/?p=723

[tomsem]

Um, and with the Personal Documents service, you can just download all of your stuff from Archived Items, no need to run calibre server at all unless you are restricted to 3G for some reason.