Metrics of NotoSansThai is not matched with Roboto, leads to Thai language cut off

GoogleCodeExporter commented 9 years ago

- Steps to reproduce the problem (including sample code if appropriate).

1. start a browser, ex, Chrome, Opera
2. go to "www.google.com"
3. type some Thai character with upper case tone marks in the search input box, 
ex นั้

- What happened.
you will find the upper case tone mark is cut off. Please check attachment 
"Nexus4_Thai_cut.png".

I used Nexus4 with 442, both Chrome and Opera can reproduce this

- What you think the correct behavior should be.

The text should not be cut off.

It seems this is a generic issue and is due to font metrics mismatch between 
Thai font and Roboto.

I also wrote a simple test app to reproduce this issue.
please check attachment "ThaiCut.zip"

Original issue reported on code.google.com by awu...@gmail.com on 11 Aug 2014 at 6:51

Attachments:

GoogleCodeExporter commented 9 years ago

Thanks for the report.

Cibu, I think you were investigating the Thai vertical metrics issues. Can you 
see if you can dig up your findings?

Original comment by roozbeh@google.com on 28 Aug 2014 at 5:08

Changed state: Accepted

GoogleCodeExporter commented 9 years ago

I looked a bit into this. The bug is not in the fonts, but the web app on 
Android.

The character sequence itself (นั้, <0E19, 0E31, 0E49>) goes vertically 
up to 2068 units in the Noto Sans Thai UI font, and to 2215 units in Noto Sans 
Thai font (the "Roboto limit" is 2189 unites). It's not clear to me what Google 
Search on mobile is using, but my guess is that the non-UI font is used, with 
*some* expectations of UI metrics.

Original comment by roozbeh@google.com on 3 Sep 2014 at 8:30

GoogleCodeExporter commented 9 years ago

Original comment by roozbeh@google.com on 16 Jan 2015 at 6:08

gongshoudao / noto

Metrics of NotoSansThai is not matched with Roboto, leads to Thai language cut off #129