Integrating features from typer-mode into speed-type

[lordnik22]

Hi there

I recently opened a PR at melpa to add the repo: https://github.com/lordnik22/typer-mode

typer-mode is an old snipped I once found on emacswiki which I now thought I could maintain. It's also meant to train typing speed and accuracy. So it's similar to speed-type.

The two main differences are:

• words are arranged in random order
• when making an error, a new line is added

I now want to integrate these two feature with two additions to speed-type:

• adding an custom parameter (speed-type-add-extra-line-on-mistake, either t or nil, default: nil) which adds "extra-line" when non-nil for each speed-type-setup. I planned to check this new custom parameter in speed-type--diff.
• new speed-type-setup: speed-type-random which arranges the words randomly from the content-source (gutenberg, buffer)

Existing functionality wouldn't change when not explicitly making changes to the custom variable.

• What do you think, would this two features fit in to this package?
• Any advice or tips :)?

Best regards
lordnik22

[dakra]

Hi!

Thanks for posting.

Did you see the speed-type-top-x/100/1000 commands? These show random top X (or top 100, 1000) words for each language.
So I think this functionality is there.

I don't mind adding a feature that adds a new line (or maybe just a new word?) when you make a typo.
Personally I think I would get annoyed by it but maybe I just have to try it first :D
But if it's customizable I don't think there's any harm in adding it as I think it will not add much code.

I'm happy to help more or give some tips if you need any.

[lordnik22]

Hi

Yes, I saw the speed-type-top-x commands. I run into this because speed-type-buffer prepares the buffer content differently than typer-mode. I would like to use a speed-type-command which takes the text from my org-buffer. This worked but I ended up typing org-tables by hand :D. This is why I though a new setup command which prepares the text a bit differently (more like typer-mode) would be nice.

I'm working on the line-adding on the fork: https://github.com/lordnik22/speed-type

I like the idea just adding words instead of whole lines. I therefore made a custom variable speed-type-add-extra-words-on-mistake with which you can specify how many words should be added on each mistake.

Setting the variable: speed-type-add-extra-words-on-mistake to a number greater than 0 and calling M-x speed-type-top-100 inserts "asdf" on each mistake. So the concept more or less works, but I had to hack some things: remove hooks and adding them again, adjust variables: speed-type--mod-str, speed-type--original-content.

The biggest problem I currently facing is that there isn't a "content-buffer" from which the typing-session takes his text from. The text is prepared in the setup-method and afterwards discarded. What do you think about the idea of an general "speed-type-content-buffer<#> for each speed-type session. This content buffer would be killed as soon as the user stops his typing-session (kills the speed-type-buffer). This content-buffer could be usefull in an concrete add-extra-word-content-fn which I supply to the setup-method.

Would be glad about your thoughts!

[dakra]

This worked but I ended up typing org-tables by hand

there is also speed-type-replace-strings but I guess you saw it and isn't enough.

I'm working on the line-adding on the fork: https://github.com/lordnik22/speed-type

feel free to create a draft/WIP PR, then it's easier to talk about concrete changes.

What do you think about the idea of an general "speed-type-content-buffer<#> for each speed-type session

For the random words I don't think it's needed, but for the books (or buffer text) I would
even like this. Then, after you finished a session, instead of pressing (n)ext and getting a complete new book/text
maybe we can add an option to simply continue where you left off.