Skip to content

Conversation

@github-actions
Copy link
Contributor

This PR contains auto-generated changes after a merge in the external gir-files repository.

@github-actions github-actions bot force-pushed the auto-pr-branch branch 3 times, most recently from c501c72 to 1770de7 Compare October 12, 2025 07:29
@github-actions github-actions bot force-pushed the auto-pr-branch branch 2 times, most recently from abc9b30 to 9b240a2 Compare October 26, 2025 02:32
@github-actions github-actions bot force-pushed the auto-pr-branch branch 2 times, most recently from c9868a6 to 7e62964 Compare November 16, 2025 01:40
@bilelmoussaoui
Copy link
Member

Sorry for the big commit but i don't have the time to cleanly do it.
Here is a diff of what has to be fixed in the GIR file

user@toolbx ~/P/g/gir-files (main)> git diff
diff --git a/GLibWin32-2.0.gir b/GLibWin32-2.0.gir
index 668a6a5..2ef5188 100644
--- a/GLibWin32-2.0.gir
+++ b/GLibWin32-2.0.gir
@@ -40,7 +40,7 @@ and/or use gtk-doc annotations.  -->
           <type name="gint" c:type="const gint"/>
         </parameter>
         <parameter name="os_type" transfer-ownership="none">
-          <type name="GLib.Win32OSType" c:type="const GWin32OSType"/>
+          <type name="OSType" c:type="const GWin32OSType"/>
         </parameter>
       </parameters>
     </function>
diff --git a/GioWin32-2.0.gir b/GioWin32-2.0.gir
index 63781b4..48d914e 100644
--- a/GioWin32-2.0.gir
+++ b/GioWin32-2.0.gir
@@ -218,7 +218,7 @@ when the stream is closed.</doc>
     </function-macro>
     <record name="NetworkMonitor" c:type="GWin32NetworkMonitor">
       <field name="parent_instance" introspectable="0" writable="1">
-        <type c:type="GNetworkMonitorBase"/>
+        <type name="gpointer" c:type="GNetworkMonitorBase"/>
       </field>
       <field name="priv" writable="1">
         <type name="NetworkMonitorPrivate" c:type="GWin32NetworkMonitorPrivate*"/>
@@ -226,7 +226,7 @@ when the stream is closed.</doc>
     </record>
     <record name="NetworkMonitorClass" c:type="GWin32NetworkMonitorClass">
       <field name="parent_class" introspectable="0" writable="1">
-        <type c:type="GNetworkMonitorBaseClass"/>
+        <type name="gpointer" c:type="GNetworkMonitorBaseClass"/>
       </field>
     </record>
     <record name="NetworkMonitorPrivate" c:type="GWin32NetworkMonitorPrivate" disguised="1" opaque="1"/>
user@toolbx ~/P/g/gir-files (main)> 

Other than that, we still miss README.md in those new crates, a license symlink like the others and probably a review of what I did.

@bilelmoussaoui bilelmoussaoui force-pushed the auto-pr-branch branch 7 times, most recently from 6a5f921 to f41f3c5 Compare November 21, 2025 13:18
@github-actions github-actions bot force-pushed the auto-pr-branch branch 2 times, most recently from 80d9c57 to dffad11 Compare November 27, 2025 08:50
That is part of glib-unix
@bilelmoussaoui bilelmoussaoui force-pushed the auto-pr-branch branch 2 times, most recently from 74981ff to c2aacc5 Compare November 27, 2025 09:58
@bilelmoussaoui
Copy link
Member

We need #1853 to fix the windows builds

pub fn g_unix_pipe_close(
self_: *mut glib::GUnixPipe,
end: glib::GUnixPipeEnd,
self_: *mut GUnixPipe,
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The commit glib/unix: HACK: Manually fix GIR file probably needs to be dropped and solved differently or otherwise the CI will find changes when running the generator

env:
# git revision of gvsbuild we use for to build GLib and the other dependencies
gvsbuildref: 230e1074047b9654b76217c5d48b10c726fbe0f6
gvsbuildref: 352e6772c61763af1178b1bb0c2cb0adb8f81d4b
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Needs rebasing for your other PR that was merged :)

#[cfg(windows)]
mod win32_input_stream;
#[cfg(windows)]
pub use self::win32_input_stream::Win32InputStream;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For backwards compatibility reasons to backport this, we probably need some kind of pub use gio_unix::* (and the same for win32 and glib) here. I wouldn't do it for main though.

Is that somehow possible or should we just stop backporting new API until next summer?

(
Socket::from_fd(OwnedFd::from_raw_fd(fds[0])).unwrap(),
Socket::from_fd(OwnedFd::from_raw_fd(fds[1])).unwrap(),
/*
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Some commented out code is leftover here

pub use ffi;

#[allow(unused_imports)]
#[allow(clippy::derived_hash_with_manual_eq)]
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Which type is that for? That's potentially a bug


#[allow(clippy::too_many_arguments)]
#[allow(unused_imports)]
#[allow(clippy::empty_line_after_outer_attr)]
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is fixed now :)

- { name: "gdk-pixbuf", test: true, args: "--features v2_42" }
- { name: "gio", test: true, args: "--features v2_80" }
- { name: "gio-win32", test: true, args: "--features v2_84" }
- { name: "gio-win32", test: true, args: "--features v2_66" }
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Happened

@sdroege
Copy link
Member

sdroege commented Dec 1, 2025

This generally looks good to me apart from the above. I guess my main question left here is how to handle backports, but we don't have to answer that as part of this PR.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants