Switch build to C++20 #4378
Switch build to C++20 #4378
Conversation
|
Thanks for this Jim! Not a review, but I know we have a few places with TODOs for when we make the move. Do you think they could be handled as part of the move? Separate PRs would be OK too, especially for any big ones. 
|
jimporter
force-pushed
the
c++20
branch
7 times, most recently
from
76774f0 to
e943fc2
Compare
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #4378 +/- ##
==========================================
- Coverage 89.35% 89.34% -0.02%
==========================================
Files 260 260
Lines 17549 17310 -239
==========================================
- Hits 15681 15465 -216
+ Misses 1868 1845 -23 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
jimporter
force-pushed
the
c++20
branch
13 times, most recently
from
3fe9219 to
6b49987
Compare
|
Hey Jim, I haven't been able to follow this very closely, but I just wanted to let you know that I am really happy to see it progressing. It will be great to have this 😃 |
jimporter
force-pushed
the
c++20
branch
2 times, most recently
from
76cf46f to
a2b9923
Compare
Instead of creating temporary `std::string`s and adding them to a buffer, make these functions take the destination buffer and add to it directly.
This also means that we now require MSVC 2022.
There was a problem hiding this comment.
LGTM, very solid @jimporter. The build is tested in Windows 11 Pro and it works with the Qt mismatch.
It's the same here as well. MSVC 2019 (latest from Chocolatey) appears to crash while attempting to compile sftp_utils.cpp (and bunch of other sftp-related files, too): It seems like it's going to be a hard requirement to switch to VS2022. I don't see any downsides to that since VS2026 is also just around the corner; it's a good time to upgrade. |
There was a problem hiding this comment.
LGTM, also ran the cli-tests on Win11 Pro, no issues.
Thanks for the initiative, @jimporter! This will enable us to modernize the codebase further.
The coverage regressions are all due to already uncovered code being touched by refactoring.
Wow, haven't seen an ICE from MSVC in a while. That takes me back! Glad to see it's all working under MSVC 2022 though. |
4 participants
I've split this into several commits for fixing the various compilation failures when switching to C++20. The main issues were:
std::u8stringuses the newchar8_t, which is incompatible withchar, so I mostly switched over to usingstd::string/char(semantically, this should mean effectively "a string using who-knows-what encoding, but probably UTF-8"). Longer term, it might make sense to try and usestd::filesystem::pathas much as possible.fmt's format strings are now type-checked at compile time, which also revealed a few bugs, likeformatcalls missing arguments and some nested calls toformat.