Intentional or not but it is completely wrong because the output is e"\u{301}" not e\u{301} (outer quotes omitted for clarity). Implementations don't need to do anything special here. Same with the new example you provided.- VictorOn Fri, Nov 25, 2022 at 6:59 AM Tom Honermann <tom@honermann.net> wrote:On 11/24/22 12:16 PM, Victor Zverovich wrote:
As promised here's a paper with the wording for the approved escaping changes: https://isocpp.org/files/papers/P2713R0.html.
Note that one example in https://github.com/cplusplus/nbballot/issues/515 (the one that uses s8) is incorrect so I didn't include it.That example was intentional. The example is:
string s8 = format("e{:?}", "\u0301"); // s8 has value "e\u{301}"
// (the combining character is escaped since it does
// not follow a non-escaped non-combining character
// produced from the same field)The formatted field argument begins with a combining character. I wrote this example with the assumption that we don't want to require implementations to have to be aware of characters from outside of the field content to be formatted as that seems like it would introduce additional complexity that is not required today; please correct me if I'm mistaken.
A similar example follows. We certainly wouldn't want the combining character from the field to combine with the escaped tab character sequence in this case.
string s9 = format("\t{:?}", "\u0301"); // s9 has value "\t\u{301}"
// (the combining character is escaped since it does
// not follow a non-escaped non-combining character
// produced from the same field)Tom.
Cheers,Victor
On Fri, Nov 11, 2022 at 6:20 PM Bryce Adelstein Lelbach aka wash <brycelelbach@gmail.com> wrote:
Please email me when you have it.
On Fri, Nov 11, 2022 at 1:58 PM Victor Zverovich <victor.zverovich@gmail.com> wrote:
> Can you have it available for our November 30th meeting (that I still need to create a calendar entry for)?
Sure.
- Victor
On Fri, Nov 11, 2022 at 1:56 PM Tom Honermann <tom@honermann.net> wrote:
Can you have it available for our November 30th meeting (that I still need to create a calendar entry for)?
Tom.
On 11/11/22 6:54 PM, Victor Zverovich via Lib-Ext wrote:
I am putting together a paper but it won't be ready today and probably needs to be looked at by SG16 to check the wording.
- Victor
On Fri, Nov 11, 2022 at 1:48 PM Corentin <corentin.jabot@gmail.com> wrote:
I think Victor is working on it, I'm happy to co author.
On Fri, Nov 11, 2022, 13:44 Bryce Adelstein Lelbach aka wash <brycelelbach@gmail.com> wrote:
We voted to resolve format escaping NB comments US-38-098 and FR-005-134 at the 2022-11 Kona meeting in both SG16 and Library Evolution.
However, the resolution we voted on is not as simple as "adopt the proposed resolution in the comments". It took me 10 minutes of reading minutes and poll results to determine what exactly we had approved.
Therefore, I'm going to need a short paper that resolves these NB comments for the Library Evolution electronic poll. It should have a brief overview and motivation (I'm not looking for much there) and the specific wording that resolves these NB comments.
Ideally, I'd get the paper today, or at least by 2022-11-14 so that it can be included in the post meeting Library Evolution polls. At the latest, I'd need it by 2023-01-01.--
Bryce Adelstein Lelbach aka wash (he/him/his)
US Programming Language Standards Chair
ISO C++ Library Evolution ChairPrincipal Architect @ NVIDIA
--
_______________________________________________ Lib-Ext mailing list Lib-Ext@lists.isocpp.org Subscription: https://lists.isocpp.org/mailman/listinfo.cgi/lib-ext Link to this post: http://lists.isocpp.org/lib-ext/2022/11/24215.php
--
Bryce Adelstein Lelbach aka wash (he/him/his)
US Programming Language Standards Chair
ISO C++ Library Evolution ChairPrincipal Architect @ NVIDIA
--