mediafire

Python MediaFire client library


Keywords
mediafire, cloud, files, sdk, storage, api, upload
License
BSD-3-Clause
Install
pip install mediafire==0.6.1

Documentation

MediaFire Python Open SDK

This is a Python implementation of MediaFire Core API client.

https://travis-ci.org/roman-yepishev/mediafire-python-open-sdk.svg?branch=master

Installing

$ pip install mediafire

What should I use?

If you are in a hurry, use MediaFireClient - it contains enough functions to get your uploads/downloads and file listings working. It does not have a stable API, and has rudimentary test coverage only.

You may want to stick to MediaFireApi and MediaFireUploader to have as much control as possible over your application flow.

mediafire.MediaFireApi

API Client library provides an interface to MediaFire API. It handles requests, responses, signatures and errors.

Usage:

from mediafire import MediaFireApi

api = MediaFireApi()
session = api.user_get_session_token(
    email='your.email@example.net',
    password='password',
    app_id='42511')

# API client does not know about the token
# until explicitly told about it:
api.session = session

response = api.user_get_info()
print(response['user_info']['display_name'])

# Or directly for methods that are not yet wrapped
response = api.request("upload/add_web_upload", {
    "url": "http://forum.mediafiredev.com/images/mfforumlogo.png",
    "filename": "mfforumlogo.png"})

response = api.request("upload/get_web_uploads",
                       {"key": response['upload_key']})

API Client library supports operation w/o session_token. In this case all operations that do require session_token will fail with Access denied error:

from mediafire import MediaFireApi

api = MediaFireApi()
response = api.system_get_info()
print(response)  # prints system info

response = api.user_get_info()  # fails with "Session token is missing"

Once set, session token can be unset:

api.session = None
# or
del api.session

For information on wrapped methods, see pydoc mediafire.api. For documentation on actual values expected, see MediaFire Core API documentation.

All wrapped methods follow the same naming convention, category_action, so upload/instant is upload_instant.

You can construct the call yourself easily:

response = api.request("user/set_avatar",
                       {"quick_key": "123456789012345"})

Downloading

API client does not handle regular file downloads because these are simple HTTP requests to URLs returned by "file/get_links". Here's how you can do that yourself:

response = api.file_get_links('c94lcpx3vax6xp3')
normal_download_url = response['links'][0]['normal_download']

response = requests.get(normal_download_url, stream=True)
with io.open("/tmp/green.jpg", 'wb') as fd:
    for chunk in response.iter_content(chunk_size=4096):
        fd.write(chunk)

In case response is a file download, e.g. file/zip, the response returned is a requests.Response object, which you can read from:

...
response = api.request("file/zip", {"keys": "c94lcpx3vax6xp3"})
with io.open("/tmp/green.zip", 'wb') as fd:
    for chunk in response.iter_content(chunk_size=4096):
        fd.write(chunk)
...

See Download documentation for more information.

mediafire.MediaFireUploader

MediaFire supports several upload methods and MediaFireUploader exposes a single upload method to make things easier:

from mediafire import (MediaFireApi, MediaFireUploader)

api = MediaFireApi()
uploader = MediaFireUploader(api)

# ... authenticate ...

fd = open('/path/to/file', 'rb')

result = uploader.upload(fd, 'Some filename.txt',
                         folder_key='1234567890123')

pprint(api.file_get_info(result.quickkey))

result is a mediafire.uploader.UploadResult instance.

FileDrop

For FileDrop uploads (i.e. when filedrop_key is used) only upload/instant result has quickkey. upload/instant and upload/resumable return None for all the fields, since upload/poll does not support encrypted upload key.

MediaFireHashInfo

MediaFire API requires file hash information to be provided during upload. For resumable uploads, the hash needs to be calculated for every upload unit.

If you are uploading a large object, you may want to calculate and store hash information, so that you can re-use it if upload fails.

  1. Check whether the file size is larger than mediafire.uploader.UPLOAD_SIMPLE_LIMIT_BYTES
    • If the file is larger, get the unit_size from mediafire.uploader.compute_resumable_upload_unit_size()
    • If the file is smaller, unit_size is None
  2. Get MediaFireHashInfo from mediafire.uploader.compute_hash_info(fd, unit_size)
  3. Store the received object for future (hash_info._asdict() will return a dict-like object)
  4. Receate the hash info structure (e.g. MediaFireHashInfo(**hash_info_dict))
  5. Provide the received object to MediaFireUploader.upload() via hash_info parameter.

MediaFireHashInfo fields:

  • file - sha256 hexdigest of the whole file
  • units[] - list of sha256 hexdigest of the separate file units
  • size - size of the file at the time when hash was computed. In case of mismatch, the file has definitely changed and the hash_info structure is no longer valid

See examples/hashing-upload.py for a working example.

mediafire.media.ConversionServerClient

This API is subject to change

This is a very thin layer on top of Image and Document conversion API.

from mediafire.media import ConversionServerClient

conv = ConversionServerClient()

response = conv.request('2004', 'm8d6blce79xhxl5', 'i', size_id='1')
with open('/tmp/example.jpg', 'rb') as fd:
    fd.write(response.content)

mediafire.client.MediaFireClient

This API is subject to change

High-level client library wraps API calls and presents simplified interface.

Supported operations:

  • File upload
  • File download (direct download link)
  • Listing directories
  • Creating directories
  • Removing files and directories
  • Getting info about files and directories

MediaFire resources can be referenced by path or by quickkey/folderkey.

  • path: mf:/Pictures/Sample.jpg or /Pictures/Sample.jpg
  • folder_key: mf:6302u1a9p0a9x (folder_key is 13 chars long)
  • quick_key: mf:46d3y4p8542kiyp (quick_key is 15 chars long)
from mediafire.client import (MediaFireClient, File, Folder)

client = MediaFireClient()
client.login(email='your.email@example.net',
    password='password',
    app_id='42511')

client.upload_file("flower.jpg", "mf:/Pictures/")
client.download_file("mf:/Pictures/flower.jpg",
                     "flower-from-mediafire.jpg")

for item in client.get_folder_contents_iter("mf:/Pictures"):
    if type(item) is File:
        print("File: {}".format(item['filename']))
    elif type(item) is Folder:
        print("Folder: {}".format(item['name']))

See examples/mediafire-cli.py for high-level client usage.

Requirements

  • python 2.7 or 3.4
  • six
  • requests
  • requests_toolbelt
  • responses (for testing)

Tests

Test suite is located under tests/

git clone https://github.com/MediaFire/mediafire-python-open-sdk.git
cd mediafire-python-open-sdk
# Run tests with python 3 interpreter
PYTHONPATH=. python3 -munittest
# Run tests with python 2 interpreter
PYTHONPATH=. python -munittest discover

Reporting issues

Please use the MediaFire/mediafire-python-open-sdk project issue tracker to report issues with the implementation.

Note that MediaFire server API is evolving as well, so you may to check MediaFire Developers Forum / REST API section for known API issues.

About and License

Copyright (c) 2014, Roman Yepishev. All rights reserved. Website: http://www.keypressure.com

This project was forked by MediaFire with explicit permission from Roman Yepishev on 10.24.2014

This project is made under BSD license. See LICENSE file for more information.

MediaFire® is a registered trademark of the MediaFire, LLC.