mainUnifont - Support: sr #109546, Many Language Alphabets are not...

 
 

You are not allowed to post comments on this tracker with your current authentication level.

sr #109546: Many Language Alphabets are not there.

Submitter:  Daryl Rhoades <gilgalad>
Submitted:  Sun 26 Aug 2018 07:41:26 AM UTC
   
 
Category:  None Priority:  5 - Normal
Severity:  3 - Normal Status:  None
Privacy:  Public Assigned to:  unifoundry
Open/Closed:  Closed Operating System:  Microsoft Windows

Sun 09 Dec 2018 12:43:40 AM UTC, comment #4: 

Daryl,

I have not heard from you following my September 1st post, so I am hoping that you understand the limitations of Unifont and its binding to the Unicode Standard. In your case, you would benefit most from a full OpenType font that provides alternate glyphs for the same character, which is beyond the capabilities of Unifont.

I also hope that the pointers I gave you to those who submitted Unicode proposals for the scripts that you use were helpful.

As I do not think there is anything more I can do on this with Unifont (given how limited its construction is), I am closing this entry. Thank you for reaching out to me though, and hopefully the information I provided has helped.

Sincerely,


Paul Hardy

Paul Hardy <unifoundry>
Group administrator
Sat 01 Sep 2018 05:10:42 PM UTC, comment #3: 

Daryl,

Unifont would only display glyphs that had code points (the numeric location of a glyph in a font) assigned in the Unicode Standard. The only exception I have made with that is with the ConScript Unicode Registry (CSUR) version, which has Private Use Area glyphs for things like Klingon and Elvish. But even that follows a generally accepted standard begun by Michael Everson, who has been involved with the Unicode Standard from the beginning.

So it would be preferable if you thought there were glyphs missing from a script to try to get someone interested in adding them to the Unicode Standard.

With Kharosthi you might be in luck. Two people put forth a proposal in 2017 to add some additional Kharosthi glyphs. You can see the proposal here:

https://www.unicode.org/L2/L2017/17012-kharosthi-add.pdf

They give their email addresses at the very beginning of that document. If the glyph you want is not included, I would contact them and be able to provide them with scans of the glyphs that you have, on coins or elsewhere.

For Phoenecian, if there are two variations of a letter you might have a more difficult time, but then again you might not. The Unicode Standard allows what are called "variation selectors". They are frequently used for Chinese characters that have many different forms. Unifont is not smart enough to handle variation selectors. Michael Everson wrote a proposal for adding Phoenecian to Unicode 20 years ago:

http://unicode.org/wg2/docs/n1932.pdf

You could try contacting him at his website, http://www.evertype.com/, but he has become extremely busy with other endeavors and he might not have time to reply.


Unifont is also not smart enough to have substitution pairs, where a two or more single glyphs might replace two separate glyphs (for example in the Latin script, replacing "fi" of "ffi" with a ligature). Likewise, Unifont does not support that capability in Indic scripts. It also cannot handle reshaping of Arabic letters where (as with the Kharosthi examples you give) a letter's shape can change depending on its position within a word.

Unifont only has one glyph per Unicode code point, with no substitution capability. Part of that is an artifact of the hard limit of TrueType fonts to only allow 64k glyphs per font, but another reason is it would be hard to know where to begin and when to stop; the Indic and Arabic scripts would be logical places to begin, but that would have to be part of a more complex font. Unifont is very simple. A full-fledged OpenType font with intelligent glyph substitution would allow the things you want, but I do not know of such fonts for the cases you mention.

Theoretically I could add any glyph I wanted in the Private Use Areas in Unicode. But then you will be tied to only being able to use one font that does not conform to the Unicode Standard. Then if missing glyphs are added to Unicode, you will have incompatible encodings. A better long term approach would be to seek to have missing glyphs for a script added to the Unicode Standard. At least with Kharosthi, your timing might be perfect if you contact the people who are proposing revisions.

Do you think this is enough information for you to continue from here? I did not give any status to this issue because I do not think there is anything that would be appropriate to do by modifying Unifont. Does this make sense, and would you be okay with me closing this request?

Thank you,


Paul Hardy

Paul Hardy <unifoundry>
Group administrator
Wed 29 Aug 2018 06:46:49 PM UTC, comment #2: 

Thanks a lot for the response. I didn't realize that Unifont is a two font set. As such I labeled them on my website as Unifont (lower), Unifont(upper). So that people using the keyboard know which font to use or switch to.

I understand and know what you mean that low resolution fonts can't support the merging of the isolated characters to vowels. There are some characters that are already merged with the vowel and vowel diacritic marks.

I'm wondering if there is a way to merge these characters manually and then make a character that can be displayed. This has already been done on this Wikipedia page but only for a few, not all.

final | isolated | isolated |

𐨀 +  𐨁  = 𐨀𐨁
𐨐 =  𐨁  + 𐨐𐨁

There are many more that are displayed on this wikipedia page.

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

I hope you don't mind me asking. I'm just being persistent in trying to display all the Kharosthi characters.

There are also characters that are missing in other sets. I'm wondering if there is anything that can be done about that. In Phoenician, there are two letter forms for "He". 𐤄‬

The other form doesn't have a character or unicode number. Not sure what to do about that. I can show you the letter on the coin(s) too.


Daryl Rhoades <gilgalad>
Sun 26 Aug 2018 02:55:23 PM UTC, comment #1: 

Daryl,

You also should download unifont_upper-11.0.02.ttf and its gpg digital signature file, unifont_upper-11.0.02.ttf.sig.  That has the Plane 1 Unicode glyphs.

Because of the large number of glyphs involved, Unifont has to be split into two fonts: Unifont covers Unicode Plane 0, and Unifont Upper covers higher planes (except Plane 2, which I leave for the Wen Quan Yi Project).

The limitation is that the TrueType font format has a hard limit of 2^16 glyphs, which is 64k or 65,536 glyphs.  I expect that sometime in the future Chinese fonts will push for a newer standard (there are over 100,000 Unicode Chinese characters, both modern and historical), but that has not happened yet.

Unifont of course is a very low-resolution font, and it does not have sophisticated glyph formation for complex scripts such as Kharosthi, Arabic, and any Indic script in general. A full vector-based TrueType font is necessary to properly render those scripts. I think that is what you are referring to in saying "the vowel characters don't transform the isolated character". That is beyond the capability of Unifont, which only has one glyph per Unicode code point.  Again, that is to some extent based on the very wide coverage of Unifont and the hard limit of 64k glyphs per TrueType font, but mainly because Unifont is not sophisticated enough to transform letter pairs into a new glyph or glyphs, for example transforming an isolated vowel in Indic languages into its combining form. If that is the issue, you will need to find an intelligent OpenType font with substitution glyphs for letter pairs, and I do not know of such a font for Kharosthi.

Thank you,


Paul Hardy

Paul Hardy <unifoundry>
Group administrator
Sun 26 Aug 2018 07:41:26 AM UTC, original submission:  

I downloaded Unifont, the latest version unifont-11.0.02.ttf. All the fonts for Nabataean, Sogdian, Old Sogdian and many others do not display in any browser or word processing document when the Unifont is selected. The version I have supports a lot of alphabet sets, but not any of the newer ones.

I also opened the Unifont in FontForge, the glyphs are not there at all for many alphabet sets. If you hold your cursor over that character or empty box, it will tell you the number.

Nabataean - Not showing.
Aegean Numerals - Not showing


Yeah, my project is working on a keyboard to help read ancient coins. Many scripts are supported. I'd like to add more but most of the newer sets I can't see.

Kharosthi has a big problem that I would like to explain later. Basically, the vowel characters don't transform the isolated character, inherent with A.

http://www.forumancientcoins.com/ancientwhitesheet/AC-VK-Greek-Latin.html

I'm using Windows 10 right now. Most browsers, including Chrome, Opera and Firefox.

If you need any more information, let me know. I'd appreciate some assistance.

Daryl Rhoades <gilgalad>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by unifoundry (Posted a comment)
  • -email is unavailable- added by gilgalad (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

     

    Follow 2 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2018-12-09 unifoundry Assigned toNone unifoundry
        Open/ClosedOpen Closed

    Back to the top

    Powered by Savane 3.13-4448.
    Corresponding source code