Identify variants which aren't used in any decompositions / which people are more likely to use

[Transfusion]

The premise of this tool is to leverage knowledge of character variants, which includes the source characters of various radicals, whether dictionary-indexed, simplified, or otherwise transformed in a consistent matter.

The right part of 価, the top part of 要, 栗, the top-right part of 湮, etc is semantically a 西, however, there are two identical (to the eye) "覀"s that are shown.

The approach taken to handle issue #2 is to convert all IDS decompositions to Unihan characters outside of the CJK Radicals Supplement in the ETL process, hence these characters are retrievable by U+8980 and not by U+2EC3.

[Transfusion]

Larger issue with U+7F52 罒, which can be reached either through 網 (semantic meaning) or 四 (orthographic variant of)

隸楷中也有很多「四」寫作「罒」形，所以後世通常認為「罒」是「四」的異體字

— Wiktionary

There are decompositions using a mix of U+7F52 罒 and U+2EAB ⺫, and U+2EB2 ⺲ is present although fortunately not used in any decomps