Replies: 8 comments 7 replies
-
|
What are you typing into the input box? Does the problem happen with the vsgimgui example? I have just tried the demo window in the vsgimgui example enabling the text input and have tried all the combinations and nothing crashes. Some of the behaviour seems a bit laggy, but I presume this is an ImGui issue. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, the issue occurs with vsgimgui_example as well. To replicate, click the "Demo Window".
|
Beta Was this translation helpful? Give feedback.
-
|
|
Beta Was this translation helpful? Give feedback.
-
|
Hi Kannode,
Are you typing on the number pad by any chance? I've got a fix - I need to
clean it up and submit it as a PR.
Regards,
Roland
…On Tue, 6 Dec 2022 at 06:28, kannode ***@***.***> wrote:
Typing into any input box in ImGui consistently results in the below
assertion failure on Linux and mac and likely on widows as well.
Assertion failed: (special_key > 256 && "ASCII stop at 127, but we use the
range [257, 511]"), function _mapToSpecialKey, file SendEventsToImGui.cpp,
line 107.
Am I missing something?
I searched issues and discussions but it seems like no one else is having
this issue.
—
Reply to this email directly, view it on GitHub
<#630>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAPOEQ4CLSQUFT4CDF7DHR3WLY67LANCNFSM6AAAAAASUUTIDA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Roland, No, I am not typing on the keypad. |
Beta Was this translation helpful? Give feedback.
-
|
Hi Robert and Kannode,
I just sent PR#30 that should fix this.
vsg-dev/vsgImGui#30
Regards,
Roland
…On Tue, 6 Dec 2022 at 07:19, Robert Osfield ***@***.***> wrote:
What are you typing into the input box?
Does the problem happen with the vsgimgui example?
I have just tried the demo window in the vsgimgui example enabling the
text input and have tried all the combinations and nothing crashes. Some of
the behaviour seems a bit laggy, but I presume this is an ImGui issue.
—
Reply to this email directly, view it on GitHub
<#630 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAPOEQ24G5364Y3K4UTNGVDWLZE5LANCNFSM6AAAAAASUUTIDA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***
.com>
|
Beta Was this translation helpful? Give feedback.
-
|
I may have found the issue, but do not know where it originates. it seems after I pressed they key "a", _vsgToIntermediateMap cannot find the keyAndModifier in line 103, itr is thus "end" as seen at the bottom of the snapshot, resulting in special_key becoming 0 due to line 104... and subsequently the assertion fails. Happy to continue looking at this but I feel if someone else has a better understanding of the mapToSpecialKey function, they would have a better shot a fixing it. |
Beta Was this translation helpful? Give feedback.
-
|
I have the demo window accessible in my application and don't have any
trouble typing into that Hello World box or using any of the other widgets.
I'm using a version of ImGui with the modifications from PR#30 on Kubuntu
22.04.
I'll build the stock vsgImGui and see if there is a problem.
One idea - are you letting vsgImgui pull its own version of ImGui via
CMake? I noticed that they have changed keyboard handling in the latest
version.
Roland
[image: image.png]
…On Tue, 6 Dec 2022 at 09:54, kannode ***@***.***> wrote:
[image: Capture]
[image: Capture2]
<http://url>
—
Reply to this email directly, view it on GitHub
<#630 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAPOEQZO6TMW6OU4WZVQ4EDWLZXD3ANCNFSM6AAAAAASUUTIDA>
.
You are receiving this because you commented.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
You have a point. I believe it may be picking up the imgui from my vcpkg toolchain. Let me figure this out some more before going on a hunt for a nonexistent problem. |
Beta Was this translation helpful? Give feedback.
-
|
My ImGui has suddenly become 1.89.2. The submodule is pulling master - I
thought it was a specific version. I was using version 1.88 before and it
was working fine.
…On Tue, 6 Dec 2022 at 12:49, kannode ***@***.***> wrote:
You have a point. I believe it may be picking up the imgui from my vcpkg
toolchain. Let me figure this out some more before going on a hunt for a
nonexistent problem.
—
Reply to this email directly, view it on GitHub
<#630 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAPOEQ4K7IM6A4SEN4ATTGTWL2LSNANCNFSM6AAAAAASUUTIDA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
If I build using ImGui 1.89.2 (master) then I get an assert failure too.
…On Tue, 6 Dec 2022 at 12:49, kannode ***@***.***> wrote:
You have a point. I believe it may be picking up the imgui from my vcpkg
toolchain. Let me figure this out some more before going on a hunt for a
nonexistent problem.
—
Reply to this email directly, view it on GitHub
<#630 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAPOEQ4K7IM6A4SEN4ATTGTWL2LSNANCNFSM6AAAAAASUUTIDA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Curious findings. vsgImGui pulls in a specific commit for ImGui 1.88 from the ImGui repo. Are you guys modifying your local checkout of vsgImGui to change this, or is vsgImGui somehow pulling in another version of ImGui even though it has a local checkout and build of it:? On Thursday afternoon I'll have wrapped up my client work for the week and can look into these issue further then. |
Beta Was this translation helpful? Give feedback.
-
|
I don't recall modifying it, but must have. I just cloned a fresh
vsgImGui and it cloned ImGui 1.88.
…On Tue, 6 Dec 2022 at 21:12, Robert Osfield ***@***.***> wrote:
Curious findings. vsgImGui pulls in a specific commit for ImGui 1.88 from
the ImGui repo. Are you guys modifying your local checkout of vsgImGui to
change this, or is vsgImGui somehow pulling in another version of ImGui
even though it has a local checkout and build of it:?
On Thursday afternoon I'll have wrapped up my client work for the week and
can look into these issue further then.
—
Reply to this email directly, view it on GitHub
<#630 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAPOEQ5DDQTAMFVRGHQLSZDWL4GP3ANCNFSM6AAAAAASUUTIDA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Some more info.
|
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the details. The mystery continues as to why different platforms are behaving differently. The management of the imgui.h header shouldn't affect things, as long as there's no bug in the update process. The vsgimgui_example should only ever see the installed include/vsgImGui/imgui.h and never references the vsgImGui source code tree. If one manually adds the headers to the vsgImGui source code when building an example/3rd party application then things could get confused, but this would be rather perverse given CMake's find_pacakage() should handle this all automatically for you. |
Beta Was this translation helpful? Give feedback.
-
Typing into any input box in ImGui consistently results in the below assertion failure on Linux and mac and likely on widows as well.
Assertion failed: (special_key > 256 && "ASCII stop at 127, but we use the range [257, 511]"), function _mapToSpecialKey, file SendEventsToImGui.cpp, line 107.
Am I missing something?
I searched issues and discussions but it seems like no one else is having this issue.
Beta Was this translation helpful? Give feedback.
All reactions