Looking at the "collation settings" table in the v16 docs, I think some
readers may have a little difficulty understanding what each row means.
https://www.postgresql.org/docs/devel/collation.html#ICU-COLLATION-SETTINGS
The "Key" column isn't meaningful and it's a bit arduous to read the
whole description column for every row in the table, just to understand
which keys I might be interested in.
I like how Peter's recent blog used the alias to organize the keys.
http://peter.eisentraut.org/blog/2023/05/16/overview-of-icu-collation-settings
I'd suggest that we add a column to this table with the alias, and also
have a consistent first sentence of each description, generally aligned
with the description from the upstream XML that Peter also referenced in
his blog.
I have an example below (and patch attached); I think this would make
the table a bit more understandable.
-Jeremy
===
Key: ka
Alias: colAlternate **<-added**
Values: noignore, shifted
Default: noignore
Description: **Collation parameter key for alternate handling.** If set
to shifted, causes some characters (e.g. punctuation or space) to be
ignored in comparison. Key ks must be set to level3 or lower to take
effect. Set key kv to control which character classes are ignored.
===
Key: kb
Alias: colBackwards **<-added**
Values: true, false
Default: false
Description: **Collation parameter key for backwards comparison of** the
level 2 differences. For example, locale und-u-kb sorts 'àe' before 'aé'.
--
http://about.me/jeremy_schneider