✨ Idea: Clipboard-Aware ICU4X Integration for Locale-Sensitive Formatting

[AbigaleD]

Hi all! I am Abigale,currenly studying in ShanghaiTech University and expect to further my study in University of Amsterdam and Vrije Universiteit Amsterdam in the Netherlands.It would be my pleasure to make contribution to this project.

I’d like to propose an idea to extend ICU4X’s usage to the clipboard level on client platforms.

📋 Idea: Build a tool or API that detects structured content (dates, numbers, currency, lists, etc.) copied to the clipboard, and formats it using ICU4X according to the user’s locale.

Example use case:
• A user copies "Sep 12, 2020" from a web page.
• When pasting into an app or form, it automatically becomes "12 de septiembre de 2020" (for Spanish locale) or "2020年9月12日" (for Chinese). And maybe it could be directly transfered into local timezone.

Possible implementations:
• A WASM demo, desktop app (Tauri), or browser extension
• Auto-locale detection and formatting preview
• Optionally, combine with LLM to interpret mixed or ambiguous input

This could be useful for:
• Multilingual workflows
• Cross-locale clipboard managers
• Developers working with internationalized content and different location

Would love to hear your thoughts — does this fit into the ICU4X ecosystem? Is this something the community would be interested in exploring or collaborating on?

Thanks! 🙌

Feel free to contact ~ abigale.dong@outlook.com

[Manishearth]

From ICU4X's point of view I suspect the part of this relevant to us is the parsing part: where you detect and parse dates. If we provided such an API, such an app could be built onto this (and I'd encourage doing so!), but the ICU4X repo itself probably would not contain it.

We've historically been rather reluctant to support i18n-aware "parsing", where you get a random string and decide to parse it as a date regardless of the format. @sffc can say more, I wonder if there's a narrow enough implementation that would work for us.

[sffc]

ICU4X already does everything it needs to do in order to support this use case, I think. Parsing should be done with an LLM, ICU4X does the formatting, and the clipboard is a system thing that is out of scope. This is not a bad project, but it's not something in scope for this team.

[AbigaleD]

Hi @sffc and @Manishearth,
Thanks so much for your kind and insightful feedback!

I’ve been reading through the GoSC project list, and I noticed that it’s labeled as a 2024 list — but since it’s already 2025, I was wondering: are these projects still considered up-to-date and open for contributions? I see discussions and contributions from both 2024 and 2025, so I’m a bit confused about the current status.

As for my background: I’ve worked as an intern on Large Language Models (LLMs), studied NLP in university courses, and recently joined a camp focused on building LLMs using Rust. These experiences have made me fit for the need in these project, and I’d love to get involved.My personal website:abigaled.github.io is still under construction but I could say more about what I can do if mentors are intereseted.

That said, I’m still new to open-source and haven’t made my first contribution yet. I understand that it’s ideal to start with a PR, but I’m also aware that the proposal deadline is approaching.

If you have any suggestions on how to approach project selection or write a strong proposal, I’d really appreciate the guidance.

[AbigaleD]

I now figure out why:

the contributer guideline is 2024 and it leads to 2024 project list and the actual project list is 2025.
It could be better if you could fix it in case it confuses people,but it does no big harm.

My first question is solved.

[sffc]

Yeah, the Contributor Guidance just hadn't been updated. Thank you for raising this. I have updated it.