Hey Yakubin,
Co-creator of Javis.jl here!
As adgjlsfhk1 said, there is an implicit global scope here!
This comes from a historic artifact from Cairo (which is what Javis is using deep within its tooling stack) which utilizes this sort of approach.
Here is an ELI5 that may help about the flow of a Javis animation: imagine Javis as a magical artist sitting at a table.
The artist is a bit peculiar as they can draw anything you want but will not draw anything until you finish telling them everything you want to see.
When you create a `Video` object, the artist grabs an empty notebook and opens it up.
If you create a `Background` object, basically you are telling the artist that you want the background of each page in the notebook to be, in this case, black for pages 1 - 200.
As you keep telling the artist what you want, the artist listens to how you want it ordered, how objects should appear and animate, etc.
Finally, when you tell the artist to `render` this animation, Javis springs to work and flips through all the different pages, drawing backgrounds, objects, etc. to match what you asked.
Then, when Javis is all done, the artist gives you the notebook that when you open it up, it will magically flip through all the pages (i.e. at a given frame rate) showing you your requested animation.
I know this ELI5 doesn't really get into the internals, but at a high level, that is what is happening with the start of the `Video` object.
Happy to give more details if desired.
We are trying to move to a v1.0.0 release very soon and in v2.0.0 we want to experiment with different syntax - perhaps a more implicit syntax similar to manim and Processing.
Out of curiosity, for you, do you find it necessary to know more about Javis's approach to use Javis or was this more of a question that you had?
Reason why I ask is perhaps we could add a "How Javis Works" section to the documentation if that could help more folks.
Thanks!
Co-creator of Javis.jl here! As adgjlsfhk1 said, there is an implicit global scope here! This comes from a historic artifact from Cairo (which is what Javis is using deep within its tooling stack) which utilizes this sort of approach.
Here is an ELI5 that may help about the flow of a Javis animation: imagine Javis as a magical artist sitting at a table. The artist is a bit peculiar as they can draw anything you want but will not draw anything until you finish telling them everything you want to see. When you create a `Video` object, the artist grabs an empty notebook and opens it up. If you create a `Background` object, basically you are telling the artist that you want the background of each page in the notebook to be, in this case, black for pages 1 - 200. As you keep telling the artist what you want, the artist listens to how you want it ordered, how objects should appear and animate, etc. Finally, when you tell the artist to `render` this animation, Javis springs to work and flips through all the different pages, drawing backgrounds, objects, etc. to match what you asked. Then, when Javis is all done, the artist gives you the notebook that when you open it up, it will magically flip through all the pages (i.e. at a given frame rate) showing you your requested animation.
I know this ELI5 doesn't really get into the internals, but at a high level, that is what is happening with the start of the `Video` object. Happy to give more details if desired. We are trying to move to a v1.0.0 release very soon and in v2.0.0 we want to experiment with different syntax - perhaps a more implicit syntax similar to manim and Processing. Out of curiosity, for you, do you find it necessary to know more about Javis's approach to use Javis or was this more of a question that you had? Reason why I ask is perhaps we could add a "How Javis Works" section to the documentation if that could help more folks.
Thanks!