Add new plural operand to CLDR / ICU
In order to handle French and related languages properly, we should add a new plural operand 'e' to the plural rule spec and code. That will allow French rules (and rules of related languages) to know about the exponent used for the number, and be properly used for formatting compact numbers of a million or more.
See https://docs.google.com/document/d/1Wx9Drhpl9p2ZqVZMGQ7KUF4pUfPtuJupv8oQ_Gf6sEE/edit# for a working draft of the proposal, which might be tuned in discussion.
At this time, it is too late to change ICU to be able to deal with ‘1c3’ format, so the spec is modified to handle both of them the same (with the provision to separate later).
Also fixed the test code to verify that DigitalQuantity works properly if called with the right setup options. More background in the original design spec, in the PRs and in .
Found problem with 'e' when doing the spec.
Note that we need to document
(a) the new ‘e' operand, and
(b) there are possible migration issues. The translations will normally be missing, and just fall back to 'other’. The distinction made in the rules is for the special purpose in French and other languages, as documented in the linked proposal.
Note: this doesn’t need to be part of submission, since we’ll just be touching French (for this release).
Is there a reason this doesn’t have a fix version? I’ll add 38; feel free to adjust.