gudarzi / savehere Goto Github PK

There's no input sanitization. The app might be vulnerable to malicious user inputs.

Also, there are no checks on the downloaded files. Requested files could probably overwrite sensitive files on the host!

The app must keep all the logs for each download queue item and show them to the user in frontend

Fix the packaging of the app and create 1 single docker image to ease deployment (no need to include filebrowser)

Add an option for users to add custom headers to their download requests

Add reporting download status to the user (useful for big files)

The app must show current/average download speed to the user.
Also, use better units in frontend for showing file sizes!

Add a button to frontend that checks the github repo and if there's an update to the repo, notifies the user to update to the latest version

Add an option for using proxy servers for downloading files
This helps with accessing region-blocked files, proxy-protected files, etc.

Add an option for selecting the network interface that the app uses for downloading files
a webserver could be connected to multiple networks using different network interfaces, each with their own ipv4/ipv6
using different ip/interface for downloading files could help with rate limiting, network-level access management, etc.

Add user management and set their access policies, so that multiple users can use the app at the same time

Download percentage goes above 100 after resuming a paused download.

The app can't properly handle downloading big files. It would be better to use a download engine like aria2 instead of relying on .net classes for downloading files!

Add some user-defined limits for downloading files. For example:

• a speed limit (maxspeed)
• a max retry count
• a max request per second (some servers might block you otherwise)

Add a collapsable speed graph for each download queue item and show current/average download speed on the graph

For now, the frontend app sends a request to the backend every 1 second, to get the status of all queue items. There must be a better way of doing this!

For the backend api to work properly, you must create an empty file at db/database.sqlite3.db. Despite using EnsureCreated and Migrate in code, the file is not created automatically by asp.net. Please find a way to fix this.

There's an issue on linux where folder ownership/permissions for downloads folder keeps changing to an irrelavant user. Doing this fixes it temporarily, but we need a better solution.

sudo chown -R 1000:1000 downloads/
sudo chmod -R 777 downloads/

Add an option so that users can use their own custom filename for each queue item (before downloading)

There's still no proper way of cancelling a download after it has started!

Add an option to frontend for renaming downloaded files

Add an option for using filenames that the target server sends, instead of using the url which can end up creating files with invalid names!

Add a button to the frontend that runs a script for updating the app to the latest version. it's probably something like this:

• stop the app (docker compose down)
• pull the latest version (git pull)
• clear some junk and data (everything under db/ and downloads/)
• run the app using docker

Add an option so that users can stream downloaded video files instead of downloading them

Add pause/resume to ongoing downloads (and if the target link doesn't support resuming, the file download must start from the beginning)

There are still some links that end up creating problematic filenames. For example this link:

• https://avatars.githubusercontent.com/u/30085894?v=4

Show more info about a file, right after adding a new link to the queue. Also, show errors if the server can't access the file. Things to consider:

• file size (otherwise show proper message)
• chosen filename
• chosen file extension
• whether the file is protected by user:password
• any errors or exceptions

Add settings/methods for increasing download speed. For example:

• using parallel connections for downloading (let users decide how many connections to use)
• using HTTP_v2 and its features like multiplexing (multiple requests over 1 connection)
• increasing buffer size (it's 80kb for now)

There are some files that are either protected by a basic auth password, or need some kind of user authentication (cookies or something else). The app currently can't download these files.