NFSL2001

Results 202 comments of NFSL2001

Additional: U+2EC1 ⻁ radical supplement should probably mirror CN of U+864E 虎 and map to uni864EuE0101-JP instead, such that uni864E-CN can be freed up. Serif is mapped to uni864EuE0101-JP. ![image](https://github.com/user-attachments/assets/d1a1f947-cefd-48bf-a397-1db22107f1b8)

Verified all fixed in v2.005. ![Image](https://github.com/user-attachments/assets/2cc501be-c029-47cb-83e2-7bb548a2cacb) ![Image](https://github.com/user-attachments/assets/30880912-cb70-402c-9d67-7adc338d9642)

It seems that the timing of `paged.polyfill.min.js` loaded into browser might affect if the `auto: false` is set or not.

(seem like `getRow()` is not correct.... need more fixes)

From what I saw the fonts are loaded multiple times to build a new GlyphOrder. Tried to sidestep it by using `deepcopy` (although I believe it's not required at all).

Pretty sure the reload happens *before* setting the new GlyphOrder. No other files call for `_openFonts()` in merge module. https://github.com/fonttools/fonttools/blob/770917d89e9151023cce011c7f7223f4ce84169e/Lib/fontTools/merge/__init__.py#L79-L91

Encountered this issue when trying to parse/write similar otb font files. As a very hacky way, I temporarily stuffed empty glyphs in `glyf` to match the required lengths. In https://github.com/fonttools/fonttools/blob/main/Lib/fontTools/ttLib/tables/_g_l_y_f.py:...

Seems to be similar to https://github.com/adobe-fonts/source-han-sans/issues/335.

Here is a quick filter where CN ≠ HK with problematics one listed. Action to take as follow. CN remap to HK: `𠂊 𠘑 𠹭 𢹂 𣳾 𥧌 𦐂 𦥑...

CID+16799 uni5E54-TW; CID+62253 uni5E54uE0101-JP. They are almost similar and leaving uni5E54uE0101-JP is ok.