Sunday, December 15, 2013

A Christmas Gift to Myself : Raspberry Pi


Here are some ideas me and my friend Mohit are going to implement when our Raspberry Pi's reach us -

  1. Personal search engine : I have a big collection of ebooks and publications I constantly need to query. So, putting its search index on a webserver would make it accessible to me everywhere and thus fasten up my research quite a lot. I'm going to use recoll for the PSE.
  2. Bit Torrent Sync - Personal data back-up server.
  3. Social network data mining - Using facebook and gmail API this could be easily implemented. Although, I need to find more open source algorithms to run. I am also open to implementing research algorithm [1]. Looking for suggestions here.
  4. Mint Server (mint.com) - My bank sends me alerts after every successful transaction. This makes it easy to implement my own mint.com server and also a splitwise.com.
  5. Bitcoin mining 
  6. Tor server - This was Mohit's idea. This reminds me of  Oscar Wilde's quote.
  7. A personal home music server (Mohit's idea)
  8. A live twitter feed monitoring and LED lighting - something on these lines.

I'm also going to get a Raspberry Pi camera module. It would be nice to create a compressed time-lapse video. I need more suggestions here too.

PS : comments and suggestions required.




Friday, December 13, 2013

Hacking Dropbox

Having spent a lot of time finding tools for file syncing without a cloud and those that would work behind proxies, I decided to write my own Dropbox app to synchronize data across computers.

But Dropbox already does that. The point is to be able to do that using the free 2GB space that Dropbox provides. The motivation is that this way the synchronization should work behind every network routers (which must not block dropbox) that blocks torrent data. (ofcourse you can also avoid that using SSH tunneling).

So I decided to register a Dropbox developer app and use the Dropbox API to synchronize files using only the free 2GB storage. The only constraint now is that the individual files cannot be more than 2GB, but that is easily solved by breaking up the file using rar. Note that, the same could be done using Google Drive which provides 15GB of free space.

The Dropbox API documentation is very neat. I used the core Dropbox API with full access to user data for this app. The code below just authenticates the app to allow access to the user account. Once provided the authentication is then saved to a file, which when detected is used again.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
import dropbox
import email, imaplib, os
import urllib2

##sign in to DROPBOX
app_key = '<key>'
app_secret = '<secret>'

flow = dropbox.client.DropboxOAuth2FlowNoRedirect(app_key, app_secret)
app_auth = False # set to false initially
if os.path.exists('accesstoken'):
 app_auth = True

if (app_auth == False):
 ### authorize server
 authorize_url = flow.start()
 print '1. Go to: ' + authorize_url
 print '2. Click "Allow" (you might have to log in first)'
 print '3. Copy the authorization code.'
 code = raw_input("Enter the authorization code here: ").strip()
 access_token, user_id = flow.finish(code)
 ## save access token once
 f = open( 'accesstoken', 'w' )
 f.write( access_token )
 f.close()
else:
 f = open( 'accesstoken', 'r' )
 access_token = f.read()
 f.close()

client = dropbox.client.DropboxClient(access_token)
print 'linked account: ', client.account_info()


The server and client communicate uses gmail. When the server uploads a file it emails the client the share link for the data. The client continuously monitors it's mailbox. As soon as it receives an email from the server, it downloads the data and sends back an acknowledgement mail. The server receives the acknowledgement, deletes the previous file, uploads another and sends the share link to the client again.


1
2
3
4
5
6
7
8
##sign in to GMAIL
user = "<server>"
pwd = "<password>"

m = imaplib.IMAP4_SSL("imap.gmail.com")
m.login(user,pwd)
m.list()
m.select("inbox")


Here's how I implemented the rest of it (thanks to these webpages 1, 2, 3) . The script for the client side would be similar

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
##main for loop
while True:
 resp, items = m.search(None, 'TO', '"<server>+python@gmail.com"')
 items = items[0].split()
 print items

 emailid=items[-1]
 resp, data = m.fetch(emailid, "(RFC822)")
 email_body = data[0][1]
 mail = email.message_from_string(email_body)

 if mail['Subject'] == "done"
  ##delete file
  client.file_delete('/'+os.path.basename(uploadfilepath))
  
  ##ask for another file (or just read from a file list) to upload directory in dropbox = main
  uploadfilepath = raw_input("enter (absolute) path to file").strip()
  uploadfile = open(uploadfilepath)
  response = client.put_file('/'+os.path.basename(uploadfilepath), uploadfile)
  print "uploaded:", response

  ##create share link
  sharelink = client.share("/"+os.path.basename(uploadfilepath), False)
  print sharelink['url']+"   "+sharelink['expires']

  ##send email
  from send_email import mail
  mail("client+python@gmail.com",   sharelink['url'],    "")
  print "email sent"
  
 time.sleep(25000)


The send_email script called above is the following :


 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
import smtplib
from email.MIMEMultipart import MIMEMultipart
from email.MIMEBase import MIMEBase
from email.MIMEText import MIMEText
from email import Encoders
import os

gmail_user = "<id>@gmail.com"
gmail_pwd = "<passwd>"

def mail(to, subject, text):
 msg = MIMEMultipart()

 msg['From'] = gmail_user
 msg['To'] = to
 msg['Subject'] = subject

 msg.attach(MIMEText(text))

 #part = MIMEBase('application', 'octet-stream')
 #part.set_payload(open(attach, 'rb').read())
 #Encoders.encode_base64(part)
 #part.add_header('Content-Disposition','attachment; filename="%s"' % os.path.basename(attach))
 #msg.attach(part)

 mailServer = smtplib.SMTP("smtp.gmail.com", 587)
 mailServer.ehlo()
 mailServer.starttls()
 mailServer.ehlo()
 mailServer.login(gmail_user, gmail_pwd)
 mailServer.sendmail(gmail_user, to, msg.as_string())
 # Should be mailServer.quit(), but that crashes...
 mailServer.close()

Wednesday, December 11, 2013

File Sync without the Cloud

I had to share some large data files with a friend recently. As I was going through the list of usual options :
1. Opera Unite
2. Torrent sharing
3. Dropbox
4. Google Drive
5. wetransfer.com
6. justbeamit.com

I found no fast and cheap option. The painful thing about the good (fast) services is that they do not provide only sync across computers, but require you to upload data to their servers, which is both expensive and insecure.

I found an awesome software, Bitsync, that just works on any device. Yes, even my phone running android. No need to setup port forwarding or go through any tech hassles. And the best feature of all, it's FREE. It's also very fast. Usually when I had to resort to torrent-sharing the speed of download was very slow, but bitsync somehow finds the optimal path.

There are other companies that provide the same service. The best among them is aerofs . You wouldn't require it's paid services for most personal use. Here's a list of all such softwares on how-to-geek .