Wrong Position of Accents for Sequences of DIN SPEC 91379

[vk-github18]

The accents in the following sequences are at wrong positions: Latin Letters (normative) Sequences A̋ C̀ C̄ C̆ C̈ C̕ C̣ C̦ C̨̆ D̂ F̀ F̄ G̀ H̄ H̦ H̱ J́ J̌ K̀ K̂ K̄ K̇ K̕ K̛ K̦ K͟H K͟h L̂ L̥ L̥̄ L̦ M̀ M̂ M̆ M̐ N̂ N̄ N̆ N̦ P̀ P̄ P̕ P̣ R̆ R̥ R̥̄ S̀ S̄ S̛̄ S̱ T̀ T̄ T̈ T̕ T̛ U̇ Z̀ Z̄ Z̆ Z̈ Z̧ a̋ c̀ c̄ c̆ c̈ c̕ c̣ c̦ c̨̆ d̂ f̀ f̄ g̀ h̄ h̦ j́ k̀ k̂ k̄ k̇ k̕ k̛ k̦ k͟h l̂ l̥ l̥̄ l̦ m̀ m̂ m̆ m̐ n̂ n̄ n̆ n̦ p̀ p̄ p̕ p̣ r̆ r̥ r̥̄ s̀ s̄ s̛̄ s̱ t̀ t̄ t̕ t̛ u̇ z̀ z̄ z̆ z̈ z̧ Ç̆ Û̄ ç̆ û̄ ÿ́ Č̕ Č̣ č̕ č̣ Ī́ ī́ Ž̦ Ž̧ ž̦ ž̧ Ḳ̄ ḳ̄ Ṣ̄ ṣ̄ Ṭ̄ ṭ̄ Ạ̈ ạ̈ Ọ̈ ọ̈ Ụ̄ Ụ̈ ụ̄ ụ̈

Rendering with Liberation Sans and MS Word:

Rendering with Noto Sans and MS Word (mostly correct except combining comma above right):

The tested sequences are used in the following specification(DIN SPEC and draft DIN):

DIN SPEC 91379: Characters in Unicode for the electronic processing of names and dataexchange in Europe; with digital attachment

https://www.xoev.de/downloads-2316#StringLatin https://www.din.de/de/wdc-beuth:din21:301228458 https://www.din.de/de/mitwirken/normenausschuesse/nia/entwuerfe/wdc-beuth:din21:339812645 See also: https://github.com/googlefonts/noto-fonts/issues/1882

[vishalvvr]

Can you please confirm which liberation-fonts release/version are you using.

if you are unaware of how to get it on windows 10, Windows+R, type fonts in the empty box and click OK find liberation-sans font file right click and choose preview.

[vk-github18]

I downloaded the tar.gz-file of release 2.1.4 from github.

[vk-github18]

The combinations with "Comma above right" look better with Featherpad/Linux The position with Windows10/Word is different, and the accent is placed too low.

See also: googlefonts/noto-fonts#1882

[vk-github18]

Rendering with hb-view 2.6.8 under Linux looks different from rendering with Word/Win10

Input file:
din91379-sequences.txt

Fonts: https://github.com/liberationfonts/liberation-fonts/files/6418984/liberation-fonts-ttf-2.1.4.tar.gz

Command line: hb-view -o din91379-sequences.png LiberationSans-Regular.ttf < din91379-sequences.txt

[vishalvvr]

I did try the same text in libreoffice (and also gedit) on my fedora-34 machine.

The initial MS Word incorrect rendering image seems to be an word processor issue not the font one(This is my guess, i need to try it by myself and then jump to conclusion).

Still i see few glyph are not rendered correctly in liberation-sans. Thanks for reporting i will be fixing it in next release.

[vk-github18]

There seem to be two issues: 1) The position of some accents is wrong as shown by hb-view/openoffice/gedit. Thanks for looking at this and fixing this. 2) A general Windows-Rendering problem - also shown by e.g. Notepad. It would be great if that problem could be fixed, too. Why is e.g. "A̋" rendered correct under Windows for Noto Sans but not for Liberation Sans - what can be changed in the font for a correct rendering under Windows?

[vishalvvr]

I did fix the rendering issue on windows, attaching sample LiberationSans-fonts.zip

Current rendering with above fix files

Would kindly request you to review this fix and confirm if it works for you.

Thank you :)

[vk-github18]

Thanks for your quick fix, it seems to work. I will check with MS Word tomorrow at work. The results at home: Libreoffice/Windows10: looks similar to your screenshot Notepad/Windows10: Some letters are cut off at the bottom.

[kimmus]

Are not some of the diacritics too high above the letters, e.g. above m, n, r compared with c, g, p, s, z? There are some collisions, too, e.g. with f, h and k.

[vishalvvr]

Are not some of the diacritics too high above the letters, e.g. above m, n, r compared with c, g, p, s, z? There are some collisions, too, e.g. with f, h and k.

Indeed and i am currently working on those :)

[vk-github18]

Word 2016 (32Bit)/Windows10

The windows problem seems to be solved, great that you started finetuning.

[vk-github18]

For reference I am adding a complete list of all letters, symbols, diacritics and combinations defined in DIN SPEC 91379. latin_list_all.txt latin_letters_all.txt

[m-m-moeller]

@vishalvvr , @vk-github18 , @kimmus, @msg-tbrtz, @kenmcd and all other contributers: Many thanks for working on a solution for all DIN SPEC 91379 content - including this and other related issues here. It's hard to find open source fonts which support all the DIN SPEC 91379 letters, symbols, diacritics and combinations AND offer same width as Arial, Times New Roman and Courier New. Same width is important, as many form sheets or official documents require this. Your work can help to avoid a lot of "refactoring" for form sheets and official documents in public services organizations - which will allow to spend "tax money" for better purposes and will be benificial for the public.