No offense taken. On the contrary, I appreciate the critique because it will make the script better. And second, I am a scientist, critique is fuel and chisel.
There are a couple of things that you mention that have already addressed, in the third iteration, such as the parent folder. I haven’t published yet; about to. I know having a static folder name was the uttermost ugly thing.
If something doesn’t make sense, I will add more comments, specially that date regex that doesn’t seem important but after tests, it is.
The tp.file.rename issue is unavoidable. Sometimes it runs too fast in my machine that doesn’t do the renaming, I had to delay it a bit. That’s why. Maybe you could test it your machine with and without the timeout.
The tp.system.suggester is not needed in this case, as this script is trying to replicate the original “Meta Template Picker” (MTP) by Jenks and Wang. The MTP is just a part that does something specific: automatic location of notes based on a prefix in the name. I am trying to make tp.system.suggester work to select between three more template pickers, when this MTP, is at the top of the choices, but I am struggling to execute the template pickers. In the use cases I’ve seen it is mostly used to select a template to run. In my case I want to select a template that contains other templates. I was about to ask for help in the forum.
The treatment of the properties entered as lines between backticks is probably a side-effect of me being influenced by my use of R’s Rmarkdown, knitr, and other R markdown packages. To me it seems neater that way.
One thing that comes obvious after using/working this template script so many times is that are clearly three parts identified: the properties declaration for many classes in a dictionary; the process of acting on the instructions in that dictionary; and the utility functions. I was thinking what if I were able to insert any dictionary, instead of being stuck at one, at the top, at will?