This is a no-bullshit file hosting and URL shortening service that also runs 0x0.st. Use with uWSGI.

mia

Last update: Dec 31, 2022

Related tags

URL Manipulation 0x0

Overview

The Null Pointer

This is a no-bullshit file hosting and URL shortening service that also runs 0x0.st. Use with uWSGI.

Configuration

To change settings, modify instance/config.py. For more information on instance configuration, see the Flask documentation.

To customize the home and error pages, simply create a templates directory in your instance directory and copy any templates you want to modify there.

If you are running nginx, you should use the X-Accel-Redirect header. To make it work, include this in your nginx config’s server block:

location /up {
    internal;
}

where /up is whatever you’ve configured as FHOST_STORAGE_PATH.

For all other servers, set FHOST_USE_X_ACCEL_REDIRECT to False and USE_X_SENDFILE to True, assuming your server supports this. Otherwise, Flask will serve the file with chunked encoding, which sucks and should be avoided at all costs.

To make files expire, simply create a cronjob that runs cleanup.py every now and then.

Before running the service for the first time, run FLASK_APP=fhost flask db upgrade.

NSFW Detection

0x0 supports classification of NSFW content via Yahoo’s open_nsfw Caffe neural network model. This works for images and video files and requires the following:

Caffe Python module (built for Python 3)
ffmpegthumbnailer executable in $PATH

Comments

fhost: allow configuration at runtime

This reads configuration from a file called fhost.cfg if it exists, allowing users to configure the server without having to maintain a fork.

I also used this opportunity to remove the gitignore entry for .py[cod]. It's only needed for python2, as python3 always places bytecode files inside __pycache__, which was already ignored.

opened by ioistired 2

Show 404 message instead of crashing when requesting a file that isn't there

When you request a file that doesn't exist, 0x0 crashes:

thestranjer@monolith:~$ curl http://0x0.st/non_existent_file.txt
<pre>Process 16 stopped
* thread #1: tid = 16, 0x00007f9e6f4d84e0, name = 'fhost'
    frame #0:
Process 16 stopped
* thread #8: tid = 16, 0x00007f9e51d0ad08 fhost`get(path='/non_existent_file.txt') + 27 at fhost.c:139, name = 'fhost/responder', stop reason = invalid address (fault address: 0x30)
    frame #0: 0x00007f9e51d0ad08 fhost`get(path='/non_existent_file.txt') + 27 at fhost.c:139
   136   get(SrvContext *ctx, const char *path)
   137   {
   138       StoredObj *obj = ctx->store->query(shurl_debase(path));
-> 139       switch (obj->type) {
   140           case ObjTypeFile:
   141               ctx->serve_file_id(obj->id);
   142               break;
(lldb) q</pre>

I fix this error with this commit. It returns an answer that is human comprehensible:

thestranjer@monolith:~$ curl http://localhost:5000/non_existent_file.txt
404 File Not Found: /non_existent_file.txt

It does not use the abort() call in the interest of returning human-comprehensible text.

opened by TheStranjer 2

Actually send file contents when using FHOST_USE_X_ACCEL_REDIRECT
When I try to run 0x0 with the default settings, by going python3 fhost.py runserver, and using FHOST_USE_X_ACCEL_REDIRECT in the state of True, it allows me to upload files. However, when I try to curl, those files, it throws errors such as this:

curl: (18) transfer closed with 34767 bytes remaining to read

This is because it isn't actually sending the file contents! The edit to the code I make causes the server to function normally and actually send the file over without issue.
opened by TheStranjer 2
remove short_url and add in-tree URLencoder

This PR removes the short_url dependency as per issue #41. This implementation is pretty much the same as in short_url except I've rewritten the enbase() function to be iterative instead of recursive. The only functions of the class are enbase() and debase() since those were the only functions being used by fhost. My python is very rusty so if anything could be better I'll gladly modify this PR :)

opened by 7415963987456321 1
Refactor is_fhost_url to use regex for url matching

I figured it was a bit unnecessary to have two functions for this so I refactored it into one function. I thought about introducing fhost_url as a global variable in order to avoid checking for every function call, but I'll await your input on that. This PR introduces re as a dependency, I don't know if you consider that to be excessive, but it certainly simplifies things.

opened by 7415963987456321 0