Message382108
| Author | hauntsaninja |
|---|---|
| Recipients | BTaskaya, Zac Hatfield-Dodds, corona10, gvanrossum, hauntsaninja, kj, levkivskyi, serhiy.storchaka |
| Date | 2020-11-30.02:25:13 |
| SpamBayes Score | -1.0 |
| Marked as misclassified | Yes |
| Message-id | <1606703113.66.0.149283824457.issue42195@roundup.psfhosted.org> |
| In-reply-to |
| Content | |
|---|---|
You're right that Callable[[P], int] is invalid, so a flat representation isn't currently ambiguous. But a flat representation would foist the responsibility for checking that to all users of `__args__`. If your code isn't ParamSpec aware, your failures will probably be less obvious or even silent. The flat representation trades off a) implementation complexity with b) usability / aesthetics, c) future proof-ness. I'm not morally opposed to such a tradeoff / am happy to defer to people with more experience in making such tradeoffs. |
|
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2020-11-30 02:25:13 | hauntsaninja | set | recipients: + hauntsaninja, gvanrossum, serhiy.storchaka, levkivskyi, corona10, Zac Hatfield-Dodds, BTaskaya, kj |
| 2020-11-30 02:25:13 | hauntsaninja | set | messageid: <1606703113.66.0.149283824457.issue42195@roundup.psfhosted.org> |
| 2020-11-30 02:25:13 | hauntsaninja | link | issue42195 messages |
| 2020-11-30 02:25:13 | hauntsaninja | create | |