[PATCH 0/1] MR1852: winebus.sys: Declare a variable only when used

[Gerald Pfeifer (＠gerald)]

udev_add_device declares a variable (product) even when it is not used
on all platforms. Use the same condition for the declaration that already
guards the code.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/1852

[PATCH 1/1] winebus.sys: Declare a variable only when used

[Gerald Pfeifer]

From: Gerald Pfeifer <gerald(a)pfeifer.com>

udev_add_device declares a variable (product) even when it is not used
on all platforms. Use the same condition for the declaration that already
guards the code.
---
 dlls/winebus.sys/bus_udev.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/dlls/winebus.sys/bus_udev.c b/dlls/winebus.sys/bus_udev.c
index 3cd8bfd1ceb..34b3305ed88 100644
--- a/dlls/winebus.sys/bus_udev.c
+++ b/dlls/winebus.sys/bus_udev.c
@@ -1329,7 +1329,9 @@ static void udev_add_device(struct udev_device *dev, int fd)
     if (!strcmp(subsystem, "hidraw"))
     {
         static const WCHAR hidraw[] = {'h','i','d','r','a','w',0};
+#ifdef HAVE_LINUX_HIDRAW_H
         char product[MAX_PATH];
+#endif
 
         if (!desc.manufacturer[0]) memcpy(desc.manufacturer, hidraw, sizeof(hidraw));
 
-- 
GitLab

https://gitlab.winehq.org/wine/wine/-/merge_requests/1852

[Rémi Bernon]

Are there systems with hidraw backends which are not Linux? I'll approve this as the hidraw code isn't guarded consistently already but if it's not the case I think maybe we should probably guard all hidraw code and not just some parts.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/1852#note_19916

[PATCH 0/1] MR1852: winebus.sys: Declare a variable only when used - approved

[Rémi Bernon]

This merge request was approved by Rémi Bernon.
-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/1852

[Gerald Pfeifer (＠gerald)]

On Tue Dec 20 15:03:53 2022 +0000, Rémi Bernon wrote:
> Are there systems with hidraw backends which are not Linux? I'll approve
> this as the hidraw code isn't guarded consistently already but if it's
> not the case I think maybe we should probably guard all hidraw code and
> not just some parts.
> I'm not sure. (One benefit of not simply guarding everything at once is broader compile time evaluation.)

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/1852#note_19927