Topic on Project:Village Pump

done; download missing .ttf

Tracked in Phabricator
Task T250677

Non latin texts display correctly (/he /my /zh /ja /ko ...) but /kjp text doesnt display correclty showing one small box for each char; for example on page Extension:Scribunto/Lua_reference_manual/kjp . I dont know if it is a problem but it is strange aspect. Where to correct ? Thanks.

မီဒီယာဝီကီသုံး Scribunto ဖွင့်ထားသည့် ဝီကီတစ်ခုတွင် "Module:" ဆိုသည့်စာစုဖြင့် စတင်သည့် စာမျက်နှာတစ်ခုကို ဖန်တီးပါ၊ ဥပမာ "Module:ငှက်ပျောသီးများ"။
Into this new page, copy the following text:

.

Christian FR (talk) 10:36, 19 April 2020 (UTC)

I see squares in front of some characters, but not all.

For example I see only one square on မြန်မာဘာသာ (the language name), at the beginning. Are you sure it's not part of this language?

Yes the language id is nok in the language box, but inside the page I see what ive printed. Nevertheless in anycase the issue is that characters are not displayable in the selected font.

https://www.mediawiki.org/wiki/Content_translation/Infographic/mni <=(task T198132) Added language support for Manipuri (mni). MediaWiki 1.32

(task T203908) Added language support for Eastern Pwo (kjp).MediaWiki 1.33

This seems like a font issue to me; all characters appear without any of the mentioned boxes on my machine. It might help to take a screenshot of what you're seeing so we can see it as well.

It is definitely a font issue. I copy-pasted the string that @Ciencia Al Poder said contained a box into Emacs and used and describe-char to find the code point and font being used to show the character.

             position: 1 of 10 (0%), column: 0
            character: မ (displayed as မ) (codepoint 4121, #o10031, #x1019)
              charset: unicode-bmp (Unicode Basic Multilingual Plane (U+0000..U+FFFF))
code point in charset: 0x1019
               script: burmese
               syntax: w 	which means: word
             category: .:Base, L:Left-to-right (strong)
             to input: type "C-x 8 RET 1019" or "C-x 8 RET MYANMAR LETTER MA"
          buffer code: #xE1 #x80 #x99
            file code: #xE1 #x80 #x99 (encoded by coding system utf-8)
              display: composed to form "မြ" (see below)

Composed with the following character(s) "ြ" using this font:
  ftcrhb:-SIL -Padauk Book-normal-normal-normal-*-13-*-*-*-*-0-iso10646-1
by these glyphs:
  [0 1 4121 237 2 0 10 12 6 [0 0 10]]
  [0 1 4156 172 7 0 8 6 0 [-7 0 0]]

Character code properties: customize what to show
  name: MYANMAR LETTER MA
  general-category: Lo (Letter, Other)
  decomposition: (4121) ('မ')

Padauk Book is a font for Myanmar text from SIL International. In fact, the same font is used to show that whole word, so I'm wondering if Ciencia's problem is because his browser or OS is not displaying composed characters correctly.

@Wladek92: I think installing Padauk Book will let you display the characters correctly.

Cannot find a way to attach my snapshots. But you mean clients need to install specific fonts ? I thought browser has to be transparent. Aren t they loaded as google fonts instead ? or may be browser doesnt display these values.... a lot of questions for me.

Also found chart => http://www.unicode.org/charts/PDF/U1000.pdf and for latin =>http://www.unicode.org/charts/PDF/U0000.pdf

Christian FR (talk) 23:53, 19 April 2020 (UTC)

I thought browser has to be transparent.

The browser could be configured to install fonts transparently, but in this case it is not.

Aren't they loaded as google fonts instead?

Here, you've triggered my inner free software pedant. I'm sure Google supplies a lot of fonts, but the company is not the font of all fonts.

I think you mean the woff standard. SIL has a WOFF font available for download that would work, but making it work seamlessly for people without a pre-installed font to handle the Burmese script would require some work.

You could create a task in phabricator to request that a WOFF be provided for the Burmese script. Or you could install the TTF file and your computer would probably use it.

I have copied a burmese .ttf in my /font directory under windows and restarted the laptop. My text hereabove is now showing the characters. Sounds sufficient.

Similar problem mni characters => with https://www.mediawiki.org/wiki/Reading/Web/Advanced_mobile_contributions/Navigation_prototype_feedback/mni

=>http://www.unicode.org/charts/PDF/UABC0.pdf

Emacs tells me the following information for ꯄꯥ (the first character in the title):

             position: 146 of 206 (70%), column: 0
            character: ꯄ (displayed as ꯄ) (codepoint 43972, #o125704, #xabc4)
              charset: unicode-bmp (Unicode Basic Multilingual Plane (U+0000..U+FFFF))
code point in charset: 0xABC4
               script: meetei-mayek
               syntax: w 	which means: word
             category: .:Base, L:Left-to-right (strong)
             to input: type "C-x 8 RET abc4" or "C-x 8 RET MEETEI MAYEK LETTER PA"
          buffer code: #xEA #xAF #x84
            file code: #xEA #xAF #x84 (encoded by coding system utf-8)
              display: composed to form "ꯄꯥ" (see below)

Composed with the following character(s) "ꯥ" using this font:
  ftcrhb:-GOOG-Noto Sans Meetei Mayek-normal-normal-normal-*-13-*-*-*-*-0-iso10646-1
by these glyphs:
  [0 1 43972 32 13 0 13 9 0 nil]
  [0 1 44005 65 0 -5 -1 13 -9 nil]

Character code properties: customize what to show
  name: MEETEI MAYEK LETTER PA
  general-category: Lo (Letter, Other)
  decomposition: (43972) ('ꯄ')

There are text properties here:
  fontified            t

This time the font is from Google. For me that is the NotoSansMeeteiMayek-Regular.ttf font file.

Thanks. I downloaded & installed MEETEI MAYEK + restart laptop => ok, both fonts are working now. Standard users must know what to do and where to load the .ttf so following is on => Phab task T250677