gallery
gallery copied to clipboard
Prompt lab crashes if clipboard has large amount of text
Describe the bug If the clipboard has a large amount of text prompt lab will crash upon bringing up paste menu
To Reproduce
- Copy a whole article like AMD Radeon RX 9060 XT launches on June 5, starting at $299 for example
- Open prompt lab
- Hold to bring up paste menu
- Observe
Expected behavior It doesn't crash and processes and much of the prompt as possible within restrained context length
Screenshots
https://github.com/user-attachments/assets/420165d3-767d-4f36-98e8-50070e5a60c1
Desktop (please complete the following information): This isn't relevant for mobile only app?
Smartphone (please complete the following information):
- Device: Galaxy S25 Plus
- OS: Andorid 15, One UI 7.0
- Browser: irrelevant? Consider removing this from issue template
- Version 1.0.2
Additional context None
Hi @nonetrix,
Thank you for raising this issue and for your feedback.
I was able to reproduce the crash on a Samsung Galaxy S24 Ultra (App Version 1.0.3) when attempting to paste large amount of text (Select+All) from the linked article you provided. During my testing, I noticed that getting a clean, text-only copy from that specific article was challenging due to its mix of images and formatting, sometimes resulting in clipboard issues. This observation seems relevant because when I tested an alternative scenario – copying a similar large amount of text (~4096 tokens) from a PDF document containing only plain text – the app did not crash.
We've logged this as a bug to investigate further and we'll update if we have any further questions or when we have an update on a potential fix.
Regarding the "Desktop" and "Browser" sections of the issue template, those are indeed irrelevant for a mobile-only app. We'll look into streamlining the issue template to avoid confusion.
Thanks again for your help in improving the app!
Hi @nonetrix ,
As we continue our investigation into this issue, we are requesting a full Bug Report. This will be a great help to our engineers in diagnosing the root cause.
We've put together a step-by-step guide on how to capture and share one here: https://github.com/google-ai-edge/gallery/blob/main/Bug_Reporting_Guide.md
Thank you for your ongoing help and patience.