valkryst / schillsaver Goto Github PK

All of the file icons used on the main screen of the program should be redesigned.

• Move the symbol inside of the file.
• Complete the file's border.
• Ensure all file icons are the same height and width when viewed in the program.
• Adjust minimum/default window width/height if necessary.

One or two users have requested that the splash screen allow for transparency on the images. See the Photoshop splash screen for an example of this.

To do this, the code that shows the splash screen (located in Driver.java) would need to be rewritten for JavaFX and make use of the top answer at the link below.

http://stackoverflow.com/questions/14972199/how-to-create-splash-screen-with-transparent-background-in-javafx

The buttons look a bit off. Get some feedback from the users and adjust them until they look good.

Couldn't colored pixels make it possible to store more bytes in a pixel? At the moment, it's just 0 or 1, but a color RGB value consists of 4 bytes. Wouldn't that make more sense?

• Add badges, if relevant.
• Add new information if required.
• Improve layout.
• Review and adjust existing information.

I think your program looks very neat :) Will have to have a proper play with it soon!

I expect you've probably seen this, https://motherboard.vice.com/en_us/article/j558k3/how-to-save-files-from-a-pc-to-a-vhs-tape but in case you haven't just thought I'd link you to it.

A few users have suggested their requirement for an option to disable update checks. Add an option through the settings manager to enable/disable the check.

If an en/decoding job is running, then Schillsaver should delete the unfinished files.

Change the button color-gradient and the text color of all components so they're easier to read and so the buttons stand out more.

The user should be able to enter two parts to both the encoding and decoding settings. The first part will come before the input file is specified and the second part will come afterwards.

The program will do the work of appending the path to ffmpeg, the first part, the path to the input file, the second part, and the path to the output file on it's own.

The fully-custom FFMPEG options WILL NOT WORK until this is resolved.

The Settings Manager's layout is alright for now, but it's going to need some sort of redesign. My current idea for the redesign is as follows.

On the main screen there will be a toolbar at the top of the screen. From this you'll be able to select, from a dropdown menu, which section of the settings you wish to edit. From that, a specific window for the specific settings selected, will be shown to the user.

Although this idea might just end up producing a lot more code with a worse looking interface.

Acceptance Criteria

• User Requirements:
  • User can open a menu to add files to the job.
  • User can a file from the job.
  • User can delete all files from the job.
  • User can specify whether the job is for Encoding or Decoding.
  • User can specify a name for the job.
    • The name must be an auto-generated UUID if the user does not specify a name.
  • User can specify a description for the job.
  • User can view a time estimate for how long the Encoding or Decoding will take.
  • User can select an output folder.

Rewrite the GUI using JavaFX instead of swing to improve GUI performance, allow for more easy GUI development, and a newer look.

Acceptance Criteria

• User Requirements:
  • User can open a menu to edit the program settings.
  • User can open a menu to create a Job.
  • User can open a menu to edit an existing Job.
  • User can delete an existing Job.
  • User can delete all existing Jobs.
  • User can process existing Jobs.
  • For each Job being processed, the user must be able to view the FFMPEG/status output in real time.

I decided to mess around and run the ffmpeg command directly through ffmpeg without using Schillsaver.

I am able to convert an image into the pixels at my desired block size and resolution and successfully convert the image back.

The issue is that the original image is always a couple KB more than the converted image, and I'm not sure why.

Schillsaver doesn't do this but the direct command does for some reason. Am I perhaps doing something wrong?

I came up with a new way to handle encoding/decoding in the program.

Instead of adding files to the list of files to encode or decode, you will add Jobs.

A Job, at it's base, is either an encode or decode Job.

Each Job, as it's added, will use the base settings of the program along with a set of settings set as the Job is added.

I might also add in a feature to save/load Jobs from a file.

Prevent a Job from running before a certain date, or time.

Etc...

This will allow for the program to encode and decode at the same time as I'll add the option to run more than one Job at a time.

Basically, just create another GUI menu to facilitate downloading any YouTube video from the users own channel (you'd need to save the user login into securely and load it when needed if you go that route) or from some input link or set of input links.

https://rg3.github.io/youtube-dl/

Program continues to run after initial settings setup is essentially exited.

When the user clicks the "Close" button on the "Settings Manager" GUI on the initial settings setup, before the config file is generated, the program continues running. The program should close.

• Remove the separate en/decode buttons.
• Allow for one or more encode and one or more decode jobs to run at once.
  • Add settings to set the number of jobs that can run at once.
  • Create a tabbed pane of FFMPEG output on the right-hand side of the screen.

Basically just rewrite everything. The code was done in a hurry, so it's a complete mess right now.

Create some form of diagram to map the program's states, what happens when each button is pressed, what happens when a job finishes, etc...

Use this info to simplify the implementation when rewriting the program.

Downloaded the lastest release, marked it as exectuable and ran it from the terminal (java -jar Schillsaver.jar).
Running Ubuntu 17.04

Encoding

1. Load in file.
2. Pad file.
3. Process file.
   1. Load X amount of data.
   2. Draw a BufferedImage containing black/white squares to represent the data.
   3. Add BufferedImage to MP4 video file.

Decoding

1. Load in file.
2. Process file.
   1. Load one frame of video file.
   2. Determine size of first square and assume all squares share the same size.
   3. Load data from squares.
   4. Write data to output archive file.

Basically, what needs to be done is that the settings menu needs to be expanded to include an option to specify the location of the YouTube-Upload python (script?), then see if you can redirect the output of FFMPEG into the python(script?) as it's running using some sort of OutputStream.

https://github.com/tokland/youtube-upload

Might not even be possible.