blob: 0aff7e1f8b4dcd511f93233bb1a3909ed566cc06 (
plain)
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
34
35
36
37
38
39
40
41
42
43
44
45
46
|
# send
I refuse to use WeTransfer or FileMail or any of that stuff.
The idea of using a third party service to do something so stupidly simple as to transfer files is insulting.
And, because nobody I work with is going to set up an FTP server for me to drop files into,
and managing an FTP server with a bunch of different accounts for each client sounds like a hassle,
I'm doing this instead.
Writing my own version of these services, so I can host it myself.
## NOTE:
This is not finished software, don't use it.
## Planning
I need a dir where I put a bunch of zip files.
Probably the home directory of a new user named "send"
Each filename will get md5sum'ed, and that will become the link.
send.sparkburst.net/f38eba6dbbe1965bc4869621d5a6fed3/test.zip
Is this amazing for security? No, you could brute force this to find the files, but honestly, who cares?
Any production important for security and leaks to be a concern will have a dedicated DIT, and this won't be needed.
Maybe I could salt it... Hmmm... A thought for later.
To generate the webpage, all I need is three pices of HTML to cat together.
1. Header and down to an opening `<h1>`
2. Title of File
3. `<\h1>` and on to half of the `<a>`
4. filename.zip
5. Download and down the rest of the page.
6. And then a bunch of file info, from these commands:
stat something.zip
zipinfo -1 something.zip | tree --fromfile .
`>>` these into index.html, and you've got a webpage.
Save that at
send.sparkburst.net/f38eba6dbbe1965bc4869621d5a6fed3/index.html
Et voila! Download page done.
Maybe I'll grab the style.css that is already in use for the rest of https://sparkburst.net/
|
