Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Simple GET-PUT-DELETE service to (temporary) store your files as an alternative to NFS shared folders.
Python
branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
.gitignore
MANIFEST.in
README.rst
gpds.py
setup.cfg
setup.py

README.rst

GPDS version 0.4.1

Simple GET-PUT-DELETE service (GPDS) to (temporary) store your files as an alternative to NFS shared folders.

It uses Gunicorn WSGI server as base and pass through all it's options. So you can simply start it using next command gpds WORKING-DIRECTORY, where working directory is path to store and read files (by default, it'll be started on 8000 port).

This server can process only three request types, so try them with curl from your console:

curl -i -X PUT -T somefile.jpg http://localhost:8000/

HTTP/1.1 201 Created
Server: gpds/0.3.1
Date: Mon, 04 Feb 2013 16:06:27 GMT
Connection: close
Location: /55/1c/551ced9e9d54658155ae8b4197afd32087a1c551.jpg
Content-Length: 0

So, what can we see here. GPDS created new file on yours file system (201 Created HTTP response) and this file is accessible on next address http://localhost:8000/55/1c/551ced9e9d54658155ae8b4197afd32087a1c551.jpg. This big string 551...551 is SHA1 hash from uploaded file content. So we can do next trick:

curl -i -X PUT -T somefile.txt http://localhost:8000/

HTTP/1.1 200 OK
Server: gpds/0.3.1
Date: Mon, 04 Feb 2013 16:17:12 GMT
Connection: close
Location: /55/1c/551ced9e9d54658155ae8b4197afd32087a1c551.jpg
Content-Length: 0

As you can see, Location is the same as in previous request. GPDS won't rewrite uploaded file.

OK, let's try to access this file:

curl -i http://localhost:8000/55/1c/551ced9e9d54658155ae8b4197afd32087a1c551.jpg

HTTP/1.1 200 OK
Server: gpds/0.3.1
Date: Mon, 04 Feb 2013 16:38:26 GMT
Connection: close
Content-Length: 56737
Content-Type: image/jpeg

...file content...

Cool, our service sent our image for us properly. Now we'll delete this file and try to access it one more time:

curl -i -X DELETE http://localhost:8000/55/1c/551ced9e9d54658155ae8b4197afd32087a1c551.jpg

HTTP/1.1 204 No Content
Server: gpds/0.3.1
Date: Mon, 04 Feb 2013 16:40:49 GMT
Connection: close
Content-Length: 0

curl -i http://localhost:8001/55/1c/551ced9e9d54658155ae8b4197afd32087a1c551.jpg

HTTP/1.1 404 Not Found
Server: gpds/0.3.1
Date: Mon, 04 Feb 2013 16:41:09 GMT
Connection: close
Content-Length: 0

As you can see GPDS successfully erased your file from disk and next time sent not found response for you.

Ok, now we'll try to do previous work with Python:

from requests import put, get, delete

filename = 'somefile.jpg'
server = 'http://localhost:8000'
r = put('%s/%s' % (server, filename), data=open(filename, 'rb'))
result = ''.join([server, r.headers['Location']])
print result # http://localhost:8000/2c/f3/2cf3af58d2416d40e18ca5e38f5baabc8a092765.jpg

r = get(result)
print r.headers['Content-Type'] # image/jpeg (our image sent for us)

for _ in range(2):
    r = delete(result)
    print r.status_code # 1st time - 204 (No Content = deleted),
                        # 2nd time - 404 (Not Found = already deleted)

As you can see, we used requests library, best way to do HTTP request in Python.

Something went wrong with that request. Please try again.