The followings are on my wish list, but most of them might be not realisable, I think:
Folding and unfolding the code lines like Mathematica notebook and Jupyter notebook. This has been wished many times but not accepted because it needs too much time and effort. However, shortly this feature was supported under Windows. Was the version 3.5.x?
HyperLinks to a position in an SCD and another SCD.
Displaying multimedia files like pictures and video in SCD similar to old SC under macOS.
Changing font, font size and style similar to old SC under macOS.
Default font which uniformly looks in all OSs.
Help documentation colours corresponding to the IDE colour themes. It is not precisely the IDE wishlist.
Indicating code line numbers in the code Mirrors in the help documentation. It is also not precisely the IDE wishlist.
inverting helps with dark themes, but i found it too distracting to have the help differ so much from the code. I ended up copying the actual color values for the theme i’m using into the help source files and regenerating them using some SC commands. unfortunately i don’t remember the actual details of the procedure
ok, i think i found the style of the theme on codemirror and used/modified the codemirror.css file
Help browser recognize Back/Forward commands from the OS (like mulitple finger swiping on the trackpad on a macbook to go back forward in history).
Some way of hiding the IDE and only focus on the GUI, not having the IDE and GUI represented as two different applications. How it is now looks and feels like a development interface and not something for an “end user”. It would be great to have the option to open SC as one application and be able to hide/unhide the IDE at will. This is one of the really nice things about the old Cocoa SC 3.6 and a reason I still use it for standalones.
I miss the keybind to clear the post window (cmd+shift+c).
I think this is a simple one: after executing a block of code, move the cursor to the end of that block so that the next block of code is easy to select and execute (as it was in pre IDE times). As it is now, the cursor stays at the beginning of the block and, if it is very long, it can take a while to find the next block, thus obstructing the flow, especially in a live situation.
I would like it if it was possible to switch between tabs by using Cmd + numbers like one does in a browser. I see students try this and become confused when it does not work this way.
I am aware that Cmd + Alt + Arrow Keys move to next tab left and right, but this is not as precise. Maybe it would be nice to be able to reach the rightmost tab by wrapping around from the leftmost tab, maybe.
I need to state that I do not use tens of tabs in scide, but I have seen students do this.
“cmd + shift + b” selects the block of the cursor’s current position. After evaluating the block, the cursor moves one line below the block by pressing the down-arrow key.
However, I think it would be convenient if there were a shortcut “move the cursor to the next block or the next line below after evaluating”.
I use this bridge to make it possible to have a GUI/interactions in the dockedBrowser.
This works very well but need to install somes dependencise, run a nodejs program and make html pages.
this can be better/easier to load Views directly on the dockedBrowser Window.
have a var/id/class for splits to automate somthing like : split[0].open(document[3]). Maybe for sessions too… Session(“abc”).open
In the days of old, when DOS was a thing, 3D Studio had a little + button on the menu bar. With that you’d save the file and a version number, or date?, was added. Very useful and independent of undo - redo.
A more complex way would be to hook up SQLite and every time one hits ctr+enter save the file (or a diff) to it. You could make a textual representation of a live session that way. Kind of like a project file, independent of the text scdfile.
If it makes it independent of the editor, maybe even better. Had a quick look for preprocessor, but only found an example turning sines into saws. Is there more?
The preprocessor is a user defined function that receives the user’s code string as input.
Since it is a user defined function, it can do anything that is possible to do with a string.
It’s reasonable for a help example to keep it simple, but it shouldn’t be inferred from this that the help example reflects an outer limit of what is possible.
For instance, the screen image in the linked post shows a dialect that is compiled by a preprocessor hook into SC code.