AddItem can fail the id assertion under certain conditions (Apple Silicon?)

[juliusl]

Describe the bug
A clear and concise description of what the bug is.

If the window title is of a certain length, followed by an input widget (for the sake of this example InputText) w/ a certain length of label and hint, followed by a button w/ a label of a certain length, it can fail the assertion at line 9931 of Imgui.cpp.

To Reproduce
Steps to reproduce the behavior, or example code that triggers the bug.

This was tough to narrow down a repro for. It seems the characters used as well as lengths changes the overall result.

Will fail w/

Assertion failed: (id != window->ID && "Cannot have an empty ID at the root of a window. If you need an empty label, use ## and read the FAQ about how the ID Stack works!"), function ItemAdd, file imgui.cpp, line 9183.

imgui
            .window("aaaaaaa: aaa-aaaaaaaa-0.0.0") 
            .build(|| {
                let mut sc = String::new();
                imgui.input_text("aaaaaa aaaaaaaa##aaaaaa", &mut sc)
                    .hint("Type '*' to list all resources")
                    .enter_returns_true(true)
                    .build();
                imgui.button("aaaaaa##aaaaaaaa");
});

Will work (only difference is the hint is commented out)

imgui
            .window("aaaaaaa: aaa-aaaaaaaa-0.0.0") 
            .build(|| {
                let mut sc = String::new();
                imgui.input_text("aaaaaa aaaaaaaa##aaaaaa", &mut sc)
                    // .hint("Type '*' to list all resources")
                    .enter_returns_true(true)
                    .build();
                imgui.button("aaaaaa##aaaaaaaa");
});

Expected behavior
A clear and concise description of what you expected to happen.

Screenshots
If applicable, add screenshots to help explain your problem.

Please describe your environment

• imgui-rs version [e.g. 0.4.0]: 0.11.0
• Operating system version: Mac OS 14.2.1
• Apple M1 Max
• winit, wgpu, native

Other information
Add related issues, suggestions about how to fix, any other context here.

This doesn't seem to repro on Ubuntu 22.04 on x86-64 which is why I think it might be hardware related. It seems related to the scratch_txt buffer.

[salvatorebenedetto]

I believe this is due to #762 which has been fixed in cef617b but it is not included in 0.11.0, so until they make a new release you can try and check out manually cef617b or later.

[juliusl]

@salvatorebenedetto thanks, your explanation clears that up. Hoping to see 0.12.0 soon.