| 26 Oct 2025 |
 samasaur | so there's no need to do a real patch, we can just set CMAKE_FIND_FRAMEWORK=FIRST in qtbase itself which will override the setup hook | 06:59:41 |
| samasaur | and that is much easier to do conditionally on darwin as well | 07:00:34 |
| samasaur | actually im gonna force-push to that PR you linked with this better solution | 07:00:57 |
| samasaur | because it's literally just
diff --git a/pkgs/development/libraries/qt-6/modules/qtbase/default.nix b/pkgs/development/libraries/qt-6/modules/qtbase/default.nix
index ab35d1906f..5789292e47 100644
--- a/pkgs/development/libraries/qt-6/modules/qtbase/default.nix
+++ b/pkgs/development/libraries/qt-6/modules/qtbase/default.nix
@@ -292,6 +292,7 @@
# When this variable is not set, cmake tries to execute xcodebuild
# to query the version.
"-DQT_INTERNAL_XCODE_VERSION=0.1"
+ "-DCMAKE_FIND_FRAMEWORK=FIRST"
]
++ lib.optionals isCrossBuild [
"-DQT_HOST_PATH=${pkgsBuildBuild.qt6.qtbase}"
| 07:01:25 |
| samasaur | not sure whether we want to do a PR to master so that qt is usable on darwin while we wait for staging, but this way we'll at least have the right fix available | 07:02:12 |
| samasaur | personally i just pinned my one offending package (prismlauncher) and am just gonna wait | 07:02:31 |
 prince213 | Nice to see a shorter patch to this | 07:05:12 |
| prince213 | I'd suggest that we target this to master, and add a comment to remove this line when 455592 reaches master | 07:05:53 |
| samasaur | yeah the patch way was overkill | 07:05:53 |
| samasaur | i had found something similar upstream when i was trying to solve the issue and i got in my own head about it | 07:06:14 |
| samasaur | this also has the benefit of being a) the right solution; b) essentially exactly the same thing as the CMake setup hook PR that went to stagign | 07:06:46 |
| samasaur | * this also has the benefit of being a) the right solution; b) essentially exactly the same thing as the CMake setup hook PR that went to staging | 07:06:47 |
| samasaur | huh okay apparently if you force-push to a closed PR on github you can no longer reopen it | 07:07:41 |
| prince213 | Time to start a new one I guess | 07:08:10 |
| samasaur | i guess so | 07:08:15 |
| samasaur | wait a second, maybe if i force-push back to what it was i can reopen it and then force-push again | 07:08:29 |
| prince213 | Ah try it | 07:08:39 |
| prince213 | Maybe you can only reopen if its HEAD is at the original place | 07:08:57 |
| samasaur | yeah that's my theory | 07:09:10 |
| samasaur | we'll see | 07:09:11 |
| samasaur | you can! | 07:10:37 |
| samasaur | wild | 07:10:43 |
| prince213 | Nice to know | 07:10:54 |
| prince213 | Technically it's "to fix Hydra builds" so it should go to staging-next | 07:13:16 |
| prince213 | But I'm not sure | 07:13:18 |
| samasaur | i can retarget | 07:13:35 |
| samasaur | okay done | 07:18:21 |
| samasaur | k i'm going to sleep. if the decision is that we should merge this into staging-next as a stopgap, i think that PR is good to go. if the decision is to target master, I can retarget again in the morning. if the decision is to wait for a staging cycle, then we have nothing left to do :) | 07:21:06 |
| prince213 | Nice work! See you in another day | 07:26:02 |
 EsperLily [she/her] | emily: i managed to get MacVim to build with the nixpkgs SDK. I didn't even have to do anything to get ibtool to work, i guess Xcode just automatically handled the fact that it wasn't in the SDK. thanks for suggesting this! https://github.com/NixOS/nixpkgs/pull/455760 | 09:01:31 |
