Some Combining Diacritical Marks not rendered correctly

[LuoZijun]

U+0304 ◌̄
U+0301 ◌́
U+030C ◌̌
U+0300 ◌̀
U+00B7 ·◌

Character set: ◌̄ ◌́ ◌̌ ◌̀ ·◌

These characters don't render correctly in Source Serif 4 font family.

[frankrolf]

I think the real issue is that U+25CC DOTTED CIRCLE is not supported in Source Serif (same in adobe-fonts/source-code-pro#313 ).

code point	character	name	supported
U+25CC	◌	DOTTED CIRCLE	❌
U+0304	̄	COMBINING MACRON	✅
U+0301	́	COMBINING ACUTE ACCENT	✅
U+030C	̌	COMBINING CARON	✅
U+0300	̀	COMBINING GRAVE ACCENT	✅
U+00B7	·	MIDDLE DOT	✅

Would you mind to explain your usage scenario for the dotted circle?

Additional background info:
https://www.unicode.org/versions/Unicode6.2.0/ch17.pdf

[LuoZijun]

In chinese Hanyu Pinyin (romanization system for Standard Mandarin Chinese), each vowel will have 5 tones: ◌̄ ◌́ ◌̌ ◌̀ ·◌，The last tone (·◌) is generally not need marked.

Hanyu Pinyin tone marks: https://en.wikipedia.org/wiki/Pinyin#Numerals_in_place_of_tone_marks
Vowel with tone mark: https://en.wikipedia.org/wiki/Pinyin#Unicode_code_points

The dotted circle with tone mark can show students the relationship between vowels and tones in Chinese teaching, and how to mark them.

here is a list for vowel with tone marks ( Note: maybe you need Noto Serif or Noto Sans font):

A Ā Á Ǎ À  a ā á ǎ à
E Ē É Ě È  e ē é ě è
I Ī Í Ǐ Ì  i ī í ǐ ì
O Ō Ó Ǒ Ò  o ō ó ǒ ò
U Ū Ú Ǔ Ù  u ū ú ǔ ù
Ü Ǖ Ǘ Ǚ Ǜ  ü ǖ ǘ ǚ ǜ

Ê Ê̄ Ế Ê̌ Ề  ê ê̄ ế ê̌ ề
M M̄ Ḿ M̌  M̀   m m̄ ḿ m̌  m̀
N N̄ Ń Ň Ǹ  n n̄ ń ň ǹ

[frankrolf]

Thank you.
It makes sense to add the dotted circle for use in a dictionary etc. – I will add it in a future release.

FWIW, here’s the current level of support for these combinations in SS4:

I remember adding the e with double circumflex to the project, which means it might be supported in a current version of Source Han Serif. Of course, I can make those glyphs available in the (non-Han) project as well, in the same update.

[LuoZijun]

Thanks a lot.

I made an HTML page to render the characters inside Pinyin, which should better identify which characters are not well supported.

HTML Page: https://gist.github.com/LuoZijun/0a1686fd5d1d57f06432ce8f7f4ba1a5

The following three pictures are the rendering results , Noto Sans Mono looks is good (the last picture).

[LuoZijun]

Noto Sans Mono ( looks good ):

Source Serif 4:

Source Sans 3:

Source Code Pro:

Source Han Serif SC:

Source Han Sans CN:

Source Han Mono SC:

[frankrolf]

Thanks for this overview.
FYI – the glyphs marked as “looks weird“ glyphs mostly are combinations as expected in Vietnamese. Since those glyphs exist as pre-composed variants, it may make sense to have the mark feature produce the stacked variants expected for Pinyin instead.

[LuoZijun]

@frankrolf

I checked the accented letters in Vietnamese, and I also asked a Vietnamese person, and they told me that the accented letters in Vietnamese and Pinyin below have the same shape. Other similar Vietnamese accented letters have similar shape rules.

Single Unicode Point: Ề Ế ề ế
Combining Diacritical Marks: Ề Ế ề ế

also i opened a new issue #119 about accented letters in Vietnamese.