Noto Sans CJK specimens are not made using Noto Sans CJK fonts

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Open each of the four "Noto Sans CJK" specimens—JP, KR, SC, and TC—by 
clicking the full name on this page: http://www.google.com/get/noto/#/
2. Examine the results, which show only two actual weights. Thin (incorrectly 
labeled as undefined) through Medium use glyphs of the same weight, and are 
glyphs from fonts other than Noto Sans CJK), and Bold and Heavy use heavier 
glyphs of the same weight, and are again from other fonts.
3. Please see the four attached screenshots.

What is the expected output? What do you see instead?

The expected output is sample text set using the appropriate Noto Sans CJK font 
and weight.

Please use labels and text to provide additional information.

It seems that font fallback was involved in making these specimens. The Noto 
Sans CJK TC is the most severe in that a mixture of different fonts are used 
(different apparent weights and sizes, which is more apparent in the two 
heavier specimens), and some of the characters are specific to Simplified 
Chinese, and thus should not appear in a Traditional Chinese specimen.

Original issue reported on code.google.com by ken.lu...@gmail.com on 21 Apr 2015 at 12:51

Attachments:

• NotoSansCJKjp.png
• NotoSansCJKkr.png
• NotoSansCJKsc.png
• NotoSanaCJKtc.png

[GoogleCodeExporter]

Original comment by roozbeh@google.com on 21 Apr 2015 at 9:59

• Changed state: New
• Added labels: Priority-Critical
• Removed labels: Priority-Medium

[GoogleCodeExporter]

Fix was pushed 45 minutes ago, site isn't live with the new data yet, 
apparently.

Original comment by dougf...@google.com on 21 Apr 2015 at 10:05

• Changed state: Started

[GoogleCodeExporter]

Looks like the change is live now.

Original comment by dougf...@google.com on 21 Apr 2015 at 10:37

• Changed state: Fixed

[GoogleCodeExporter]

The correct fonts and weights are now being used, but two issues persist: 1) 
the "Thin" weight is still displaying as "undefined" as its name; and 2) Noto 
Sans CJK TC sample still includes characters that are specific to Simplified 
Chinese (this is because the sample is simply U+4E00 through U+4E40, which 
includes characters that are out-of-scope of Traditional Chinese).

Original comment by ken.lu...@gmail.com on 21 Apr 2015 at 10:52

[GoogleCodeExporter]

Thanks Ken.

The 'undefined' issue is a problem with the web site, which I've requested to 
be fixed (it's not under our direct control). That should happen in a day or so.

I'll file a separate bug for the sample text fix, thanks for pointing this out.

Original comment by dougf...@google.com on 21 Apr 2015 at 10:56