| 26 Sep 2025 |
Taeer Bar-Yam | is there a reason that in value.hh, we pass ptrSize around as a template variable, rather than using sizeof(void *) at the point of use, or defining static constexpr detail::ptrSize = sizeof(void *)? Do we ever want to instantiate things at multiple ptrSizes during the same compilation? | 13:10:28 |
Taeer Bar-Yam | Never mind. I see there's some games being played to get it to specialize correctly for 32 and 64 bit systems. | 13:37:37 |
John Ericson | who was going to work on the "deprecated features" thing? | 16:21:46 |
John Ericson | * who was going to work on the "deprecated features" / "warnings" thing? | 16:21:52 |
John Ericson | I want to disable absolute paths haha | 16:22:06 |
John Ericson | * I want to disable absolute paths in the language, haha | 16:22:21 |
John Ericson | Taeer Bar-Yam: I am curious whether you can remove std::string earlier in your ExprString PR commit history? | 16:55:13 |
John Ericson | what is just stopping us from having the Value own it, and a std::string_view pointing within the value to avoid downcasting? | 16:55:42 |
John Ericson | (perhaps this is something that Sergei Zimmerman (xokdvium) already talked to you about, in which case my bad) | 16:55:57 |
John Ericson | is the issue that the value might be garbage collected too earlier / root set problems? | 16:56:33 |
Taeer Bar-Yam | Yes. It will immediately get garbage collected | 16:56:59 |
John Ericson | OK | 16:57:23 |
Taeer Bar-Yam | probably I could rearrange the commits. I wasn't sure how much effort to put into cleaning up the commit history because i wasn't sure if I should squash them all before merging anyways | 16:57:47 |
Sergei Zimmerman (xokdvium) | In reply to @shine:proqqul.net probably I could rearrange the commits. I wasn't sure how much effort to put into cleaning up the commit history because i wasn't sure if I should squash them all before merging anyways In case the in-between commits are broken then there isn’t much point to keeping those separate. Also the diff is small enough | 16:58:45 |
John Ericson | I don't know how much things can/should be squashed, I'll leave that to sergei :), but yeah anything that will be squashed away of course don't worry about the history | 16:59:06 |