Closed tahonermann closed 4 years ago
As in, _$
and _¥
are seen as an identifier (_
) and then a separate non-identifier?
Do we have places where _$
used together are already pre-existing? If not we'll have to consider a TR31-style solution to making an underscore followed by certain symbols a single identifier, or hand-crafting special grammar rules for UDLs.
The (abstract) request is that "$" and "€" should be valid suffixes for user-defined literals. It is believed that currency symbols are not valid identifier characters at this time and in a TR31-enabled future. Suggesting to use ISO currency codes (e.g. _USD and _EUR) instead and saying "no" to the request is a reasonable response, in my view.
The following messages on the CWG mailing list are relevant:
(Oh, and the progressive encoding corruption through those messages is entertaining! The first message contains _=C2=A2
, but that soon becomes _��
, and then progresses to ����
)
SG16 discussed this issue at its April 22nd, 2020 telecon. The following poll was performed:
Poll: Is there any objection to unanimous consent for recommending rejection of this proposal?
The SG16 recommendation has been communicated to the EWG chair.
Closing this issue.
JF requested that SG16 consider this issue and provide a recommendation to EWG.
CWG issue 1871 tracks a request to allow the use of non-identifier characters, specifically currency symbols, as UDL suffixes.