Add POJ marks, please

[cathree3]

Cormorant and Ysabeau are so beautiful, I hope they can be used in this Taiwanese language. To-siā (Thank you).

About the POJ: https://en.wikipedia.org/wiki/Pe%CC%8Dh-%C5%8De-j%C4%AB POJ marks: https://tauhu.tw/gua-ji-pio/

E.g https://github.com/justfont/open-huninn-font https://github.com/ButTaiwan/genyo-font https://github.com/aiongg/POJFonts

[CatharsisFonts]

I'll look into it. Do you have a list of Unicode ranges that need to be covered? Are combining marks OK or do you also need precomposed characters?

[cathree3]

need precomposed characters. Is this list ok? https://docs.google.com/spreadsheets/d/1RPN_4sVAYlHaMchak1wwf2dpNKRcLil7WFHkB1ZGVuE/edit#gid=1375294313

AIUEOO͘MNaiueoo͘mn ÁÍÚÉÓÓ͘ḾŃáíúéóó͘ḿń ÀÌÙÈÒÒ͘M̀Ǹàìùèòò͘m̀ǹ ÂÎÛÊÔÔ͘M̂N̂âîûêôô͘m̂n̂ ǍǏǓĚǑǑ͘M̌Ňǎǐǔěǒǒ͘m̌ň ĀĪŪĒŌŌ͘M̄N̄āīūēōō͘m̄n̄ A̍I̍U̍E̍O̍O̍͘M̍N̍a̍i̍u̍e̍o̍o̍͘m̍n̍ A̋I̋ŰE̋ŐŐ͘M̋N̋a̋i̋űe̋őő͘m̋n̋ ĂĬŬĔŎŎ͘M̆N̆ăĭŭĕŏŏ͘m̆n̆ ⁿᶰ

Here is a paper, hope it can help you https://books.google.com/books?id=3W2f-4VpuSEC&printsec=frontcover&hl=zh-TW&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false

https://en.wikipedia.org/wiki/Taigi_Unicode

[CatharsisFonts]

I implemented the «Latin» unicode range specified by your Google Doc in the Roman cut (no italics or stylistic alternates etc. so far). It's in the new commit. Does that work? I noticed there are no precomposed characters with the toprightdotcomb; hope it still does its job. I did add ligatures with o and O, but that might be counterproductive...?

[cathree3]

Letters and symbols have collides, need more ligatures, Thanks

Like this font: https://github.com/justfont/open-huninn-font/blob/master/font/jf-openhuninn-1.1.ttf Huninn has many POJ ligatures

[CatharsisFonts]

Better now?

BTW, I'll still have to propagate those changes to Italic, make the smallcaps, and implement the stylistic alternates. I've got a lot of other things on my plate right now, so that's going to take a while to happen.

[cathree3]

Great. Only a little problem, They are a little close o͘ ó͘ ò͘ ô͘ ǒ͘ ō͘ o̍͘ ő͘ ŏ͘

Could you please add the Superscript small cap ᶰ? code: U+1DB0

[CatharsisFonts]

I added the cap version of nmod and regularized hungarumlaut to fit in with the other accents (there is a dedicated loclHUN feature to adapt it to Hungarian tastes anyway). As for the o-dot combinations: Should I just raise the macron in that particular glyph, or is there some other subtle change between the two lines? Hard to tell by eye.

[cathree3]

This version looks great

[CatharsisFonts]

That link doesn't work for me.

[cathree3]

sorry, I re-uploaded, These glyphs looks great

This is the text of the POJ, Can test typesetting effect. https://zh-min-nan.wikipedia.org/wiki/Th%C3%A2u-ia%CC%8Dh

[CatharsisFonts]

Great! I’ll propagate the coverage to the Italics and to Ysabeau.

Just noticed the alignment of cap accents is wobbly, especially comparing breve and tilde... will have to fix.

[CatharsisFonts]

I've propagated the POJ characters and ligatures through to smallcaps and the stylistic sets. Do they work for you?

BTW, that Wikipedia page doesn't seem to include any toprightdot combining accents. I'm assuming the periodcentered that appears here and there is a fallback solution to simulate it?

[CatharsisFonts]

I just added the Italics as well. Do they work as intended?

[cathree3]

I temporarily found these problems Not finished yet Please give me some more time to test

Reply to you on October 10th Thank you for your work

[CatharsisFonts]

Both of those look correct on my side:

I assume the problem is that the ligatures are not triggered and the combining accents are used as-is. I do have the requisite OpenType code in LIGA for both of these cases, though.

Can you try again with the newest commit (fixed export problem), and otherwise send me the exact string you're using for those samples?

[cathree3]

text: ji̍t Pài-gō͘

Sorry, I tried to export fonts with Glyphys, But still not working I test the POJ typography in Indesign.

Can't liga (One is ok, the other is not)

I 預組：i í ì î ī ǐ ĭ I Í Ì Î Ī Ǐ Ĭ I 分離：i ì í î ī i̍ ǐ ĭ i̋ I Ì Í Î Ī I̍ Ǐ Ĭ I̋ O + 點 + 調號：o͘ ò͘ ó͘ ô͘ ō͘ o̍͘ ǒ͘ ŏ͘ ő͘ O͘ Ò͘ Ó͘ Ô͘ Ō͘ O̍͘ Ǒ͘ Ŏ͘ Ő͘ O + 調號 + 點：o͘ ò͘ ó͘ ô͘ ō͘ o̍͘ ǒ͘ ŏ͘ ő͘ O͘ Ò͘ Ó͘ Ô͘ Ō͘ O̍͘ Ǒ͘ Ŏ͘ Ő͘ O 預組 + 點（o ó ò ô ō ǒ ŏ ő O Ó Ò Ô Ō Ǒ Ŏ Ő+_̇）：o͘ ó͘ ò͘ ô͘ ō͘ ǒ͘ ŏ͘ ő͘ O͘ Ó͘ Ò͘ Ô͘ Ō͘ Ǒ͘ Ŏ͘ Ő͘ https://blog.justfont.com/2019/01/jinxuan-taiwan-letters/

[CatharsisFonts]

I think I found the problem. Your test string contained /omacron/dotaboverightcomb/ rather than /o/macroncomb/dotaboverightcomb/. I added those ligatures, too. Did it work...?

And I hope the SC and Unicase fonts work too, now...

[cathree3]

I installed them all. thank you very much, I'm sorry I didn't find the problem before, they can work to liga (except for Upright) The point of o̍͘ , compared with others, is somewhat lower (All weights)

The point of o̍͘ , compared with others, is somewhat lower (All weights)

[CatharsisFonts]

When I paste your character into Glyphs, it looks right:

But when I look at it in Font Book, I can reproduce your error:

It looks to me like the smallcaps version of the verticallineabove is used, and possibly the smallcaps version of the dotaboveright as well. This probably has to do with the support for smallcaps ligatures that I just recently added. I'll check.

EDIT: Interestingly, the exported OTF file Cormorant-Regular.otf contains the correctly shaped ligatures:

I wonder whether it could be a font cache issue.

[CatharsisFonts]

Huh, my latest export now seems to work in Font Book too. Not sure what I did to help. Maybe it was a caching issue after all?

[cathree3]

Yes, they can work. (in cormorant, garamond, infant) Except Unicase and SC, their points should be higher.

Glyphs of Unicase are not uniform. m m̀ ḿ m̂ m̄ m̍ m̌ m̆ n ǹ ń n̂ n̄ n̍ ň n̆ n̋ ⁿᶰ

[CatharsisFonts]

Upon recommendation from the Glyphs team, I moved all those precomposed characters from LIGA to CCMP, where they feel more at home. I found another issue that was messing up some SC ligatures (decomposition happening after rather than before glyph replacement). Hopefully this finally finishes the job? Please tell me it does. 🙄

[cathree3]

Yes, Ligatures are fine, the dot is ok. thank you

Could the ᴺ add to the Cormorant? (Modifier Letter Capital N [U+1D3A])

[schriftgestalt]

I don't think it needs most of the "verticallineabovecomb" ligatures. Most of them should be handled by mark positioning. And the to remove the dot in "i" for "i_verticallineabovecomb" there should be code in the ccmp already that handles this by removing the dot whenever there is a top comping in mark next to it. And after looking at it a bit more, I think i doesn’t need any of the precomposed glyphs.

[CatharsisFonts]

I figure the point of the precomposed ligatures is to enable the script in applications that don’t support combining accents?

[schriftgestalt]

There are no apps that support ligature feature but no mark positioning. At least non less than 10 years old.