straup / py-ticketserver Goto Github PK

This is a bare-bones ticket server that runs as a stand-alone HTTP service.

It allows applications to request a new ticket, check the status of an existing
ticket and mark a ticket as completed. It is not sexy. It does not make
ponies. It makes tickets. What you do with them is your business but presumably
you'd use to keep track of long(ish) running tasks so that clients can
periodically poll to see if they've been completed.

It uses the Shove module to provide a variety of backends for storing tickets;
as of this writing only in-memory and Amazon S3 backends are supported.

The interface looks like this:

	http://localhost:8003/get_ticket

	http://localhost:8003/check_ticket?ticket=TICKETID

	http://localhost:8003/mark_ticket_completed?ticket=TICKETID

All calls return plain text. The 'get_ticket' method returns a ticket. The check
ticket returns (UNKNOWN, PENDING, COMPLETED). The 'mark_ticket_completed' method
returns (OK, FAIL).

Ticket IDs are generated as UUIDs and, by default, are purged the first time
they are checked after having been marked as completed.

There is (currently) no authentication model. It's not expected that you'll be
running this on a public-facing host. Buyer beware.

--

# EXAMPLE (CLI usage)

$> ticketserver/__init__.py --scheme s3 --port 9999 --s3_config ~/.api/aws.cfg --s3_bucket ticketserver

# EXAMPLE (S3 config file)

[aws]
access_key=AMAZON_WEBSERVICES_ACCESSKEY
secret_key=AMAZON_WEBSERVICES_SECRETKEY

# EXAMPLE (from your code)

ts = ticketserver('s3://key:secret@bucket')

ticket = ts.get_ticket()

ts.mark_ticket_completed(ticket)

status = ts.check_ticket(ticket)

--

# IMPORTANT

This package also uses the Servable WSGI wrapper library which is great except
for the part where you will currently need to comment out line 104 of
servable.py per this thread:

http://code.google.com/p/servable/issues/detail?id=1

--

# REQUIREMENTS

http://pypi.python.org/pypi/shove
http://pypi.python.org/pypi/servable/