Wrong unicode is used for special characters with UPPERCASE/lowercase plugins

[mesqueeb]

I write this with my keyboard:

À

then I write this, and I convert it uppercase of PopClip:

à

it becomes

À

now I copy past both and log the charCode in JS:

'À'.charCodeAt(0) // 192
'À'.charCodeAt(0) // 65

Github comments seem to normalise this so I can't properly paste the difference...

but the top one is written with my keyboard, and the bottom one with popclip's uppercase conversion:

[mesqueeb]

followup:
you can see in VSCode it's not recognised to be the same string any more... (look closely at the highlighting)

[mesqueeb]

Ok it's about this:

• "NFC"
  Canonical Decomposition, followed by Canonical Composition.
• "NFD"
  Canonical Decomposition.

there is no right or wrong answer...

[mesqueeb]

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/normalize

[pilotmoon]

Uppercase etc fails with unicode chars

[mesqueeb]

@invariant FYI, I think it makes most sense that you keep the original status of special characters:

• ①　special characters that are using "canonical decomposition" (under the hood 2 characters)
  • should be made capital / lower case by using the "canonical decomposition" version of its capital / lower case version.
• ②　special characters that are using "canonical composition" (under the hood 1 character)
  • should be made capital / lower case by using the "canonical composition" version of its capital / lower case version.

[pilotmoon]

Just going through some issues. I wonder, did this get resolved with the shift to JavaScript?

[mesqueeb]

@pilotmoon I just double checked and it seems to be fixed! 🎉

NFD keeps NFD no matter how many times we convert,
and same for NFC