gordon-cs / appbuilder-storybuilder Goto Github PK

As a user, I want to be able to switch the language of the title and credit slide

I believer you are running the script by using go as an interpreter based on the current directory. For deployment in SAB, we will need to compile to GO code to a platform-specific binary.

I asked colleagues who release go applications. They said to use https://github.com/goplus/gox.

Something like:

targets=( "darwin/amd64" "linux/386" "linux/amd64" "linux/arm" "windows/386" )
distPath="../../dist"

for target in "${targets[@]}"
do
    # Build binary using gox
    gox -osarch="${target}" -output="${distPath}/${target}/my-project"
...

As a user, I want to be able to give a .slideshow file with all the information, then run the program that will produce a video with the given timing file.

As a user, I want to be able to specify whether the program displays detailed information on each step of the process as it proceeds.

Software should be able to check system core count and run production in parallel if able to increase runtime speeds

The transitions should happen at the specified time and the duration should be split between the two slides.

The audio should not fade in/out.

If template is not provided at runtime, code should search current folder for any available .slideshow templates and use the first one it finds

Instead of labeling the final output video "final.mp4", video title should use the .slideshow template name to avoid confusion when producing multiple videos

1/31 - 2/4

Added because too many temporary files generated by the process

Please rename repo to appbuilder-storybuilder. This will affect the generated binaries with #45.

As a user, I want to be able to add selected filters to each image such as zoom-in/out

Final product can include an option for generating an overlay video for comparison between a previously generated video and current iteration version to spot differences or bugs between the two.

1/30

As a user, I want to be able to see a log of how long the video creation process took

When -o or a more specific option is used, the temp folder where all the temporary videos generated are placed should be created separate from the current folder with the .slideshow. This is to avoid read-only folder issues

2/1 -

Could you add a license to the project? I would recommend the MIT License.

Here is the documentation on how to add a license from a template.
https://docs.github.com/en/communities/setting-up-your-project-for-healthy-contributions/adding-a-license-to-a-repository

As a user, I want to be able to add selected filters to each image such as zoom-in/out

As a user, once a video is finished, the program will upload the video and subtitles to my YouTube channel

Software should include a flag to add a filepath to an output directory, defaulting to local folder

After testing, figured out that this happens when video contains empty audio (silent audio) at the end of the video. Then, it cannot add audio transitions between the two videos.

Solution:

1. Maybe removing the empty audio, creating new video with empty audio, and then concatenating the videos.

Solved: Came up with new method that merges videos in one single ffmpeg command.

Update old fade function (used for older ffmpeg version) to first create individual videos and then merge them using just traditional fade.

Added after user study

1/24 - 1/30

Added because traditional transitions only do fade. xfade has a wide range of filters available, but only works for recent versions of ffmpeg

In the new slideshow, start/stop attributes were removed from the element. The values were from the original audio, not the resulting narration mp3 (the reason the start was 4360 was that it was the position in the original source full John 1 mp3). So it was not correct to include them in the .slideshow output.

Determine when you should start and stop the narration audio based on the durations of the slides.

As a user, I want to be able to see the log when the program is generating videos. So, when exceptions happen, I can deal with it immediately.

As a user, I want to be able to select an audio + timing file and a slideshow, then click a button/key that will combine both into a .mp4 video.

When creating a video our scale images function modifies the original images used which can cause issues if a lower quality video is being created. We need to alter the scaling to create separate copies of the images and store those in the temp folder to ensure we don't affect the template

As a user, I want to be able to run the same program multiple times at once.

As a user, I want to be able to produce a lower quality video to be uploaded to Scripture Earth

When code is run and .slideshow is parsed, all the template images/audios should be accessed based on the location of the .slideshow template, not the location of the executable.

Merging all the temporary videos in one single command is much faster than the previous merge method. It is also simpler to debug and understand.

As a user, I want to be able to share and run this process as an executable on any OS (Mac, Linux, Windows) even in a docker container.

As a user, I want to be able to add appropriate transitions between images

Code should include a "-h" option flag to display list of possible flags and their uses.

This is only partially parallel since it only parallelizes the diving part of the method.
Merging is not parallel.

Software should include a flag to prevent deletion of temporary files created during video production for debugging purposes

1/31 - 2/4

Added because result video is different than test videohttps://github.com//issues/39