Improve scroll behavior by studying Surfingkeys

[AdamWagner]

Hi Alex. I'm really impressed by Krabby – amazing work!
Of all the navigate-by-keyboard extensions, the Krabby codebase is the easiest to understand (even easier than Saka Key, IMO).

IMO, Krabby's weakest functionality is scrolling behavior: it's laggy and twitchy compared to Surfingkeys (I can provide a proper report if you can't reproduce, but I think it will be easy to see the shortcomings)

The Surfingkeys codebase may be convoluted, but the scrolling behavior is the smoothest of all the options (Vimium, C-vim, Triadactyl, Saka Key, etc). Porting the smoothness of Surfingkeys' smooth scrolling to Krabby would make Krabby my preferred extension for keyboard navigation.

The scrolling functionality in Surfingkeys resides here:
https://github.com/brookhong/Surfingkeys/blob/master/content_scripts/normal.js#L513

[alexherbo2]

Saka Key is god-like to me, in terms of architecture. It misses so much features, but its codebase is the cleanest I have seen.

I’ll take a look.

[alexherbo2]

@AdamWagner I do it, or do you want to make a PR?

[AdamWagner]

@alexherbo2 Thanks for the offer – I will try to untangle Surfingkeys scroll functionality and make a PR. 🤞

[alexherbo2]

@AdamWagner I’m really happy with that change, thanks for the report. 🦀

[AdamWagner]

@alexherbo2 I re-installed Krabby on a fresh Chrome profile and started to dig into the scrolling behavior.

Your last two commits dramatically improved scrolling via j and k, with a couple of very minor changes! Really nicely done.

Big-step scrolling with J and K is still sticky during key repeat, so I'll try to smooth that out.

[alexherbo2]

J and K, for pageDown and pageUp, use the built-in scrollBy() function with behavior option being set to smooth.

It’s just a matter of making them using animate().

[AdamWagner]

@alexherbo2 Thanks. As I played with it, I realized pageDown and pageUp weren't intended to be repeated.

I (mis)use pageDown and pageUp as "scroll faster". For my use, I'm pretty happy with these tweaks: AdamWagner@2959b42

If you have a chance, let me know what you think.

[alexherbo2]

It looks awesome, really, and the implementation looks clean.

• I noticed a regression of the small scroll down and up functionalities: there is a tiny pause after the first scroll, before the smooth scrolling continues.

• Is there a way to preserve pageDown and pageUp methods when a single key is released?

• fastFactor could be removed if we can preserve the functionality.

The user interface Customizability should be achieved via programmability, not an enormous inventory of options that interact with each other in obscure ways.

• I find the idea of friction and longThrow really awesome.

• Can you detail a bit how they work (by documenting the code)?

• What does step → amt name change is for?

• I prefer real world variable names when possible, instead of abbreviations.

• I prefer to not change style (as for _event and once), at least not in the same commit and / or PR.

• For coding style, I prefer to not add ; at the end of statements and , after the last field element (as in directions).

• Public static fields are not supported in Firefox (static directions), you will have to use a similar approach than in Modal:

class Modal {
  // To do: Use public static fields when supported by Firefox
  // https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Classes/Class_fields#Public_static_fields
  static MODIFIER_KEYS() {
    return ['Shift', 'Control', 'Alt', 'Meta']
  }
}

• I’m not sure if directions should be capitalized or kept lowercase, as it’s static. 🤔

• Finally, I’m all in to embrace \o/ smooth scrolling only.

Configurability is the root of all evil Every configuration option in a program is a place where the program is too stupid to figure out for itself what the user really wants, and should be considered a failure of both the program and the programmer who implemented it.

[alexherbo2]

(While experimenting with the new changes, have you updated krabby.js to pass the repeat argument to J and K?)

[alexherbo2]

@AdamWagner I added your work and gave you credit in the README page. I hope that is ok.