perf: implement debouncing and request cancellation for selection events

[Suke2004]

Description:
Rapid word selection triggers multiple overlapping API calls, wasting resources and degrading performance. Add debouncing and cancellation logic to optimize user experience.

✅ Acceptance Criteria:

• Implement debouncing (e.g., 300ms delay).
• Cancel in-flight requests when a new selection occurs.
• Optimize API rate limiting.
• Avoid duplicate lookups with local caching.

[tanish-7869]

can you assign this to me?

[BennyPerumalla]

@tanish-7869 Definately, Before Assigning this can you please describe how you are going to resolve this issue

[tanish-7869]

@Suke2004 Sure! Here’s how I plan to approach resolving this issue:

I’ll introduce a short debounce (around 300ms for example) for user selection or text input events so that API calls are triggered only after the user stops interacting briefly. I’ll use something like an AbortController (or equivalent mechanism depending on where the request is handled) to cancel any previous ongoing API calls when a new selection occurs. This will prevent outdated responses from overwriting newer results. I’ll store results for previously fetched words in a simple in-memory cache to avoid redundant lookups for the same term. With the above changes, API calls will be minimized, improving responsiveness and reducing rate-limit risk.

I’ll first check where the lookup calls are triggered in the extension and then integrate these optimizations at that layer.

[BennyPerumalla]

Assigning this to you, @tanish-7869 . This is a high-priority enhancement.

As you work on this, please ensure the solution is well-thought-out and adheres to our project's coding standards. We'll be looking for a carefully crafted implementation that shows a clear understanding of the requirements.

Restrain yourself from relying much on AI

[tanish-7869]

@BennyPerumalla
Do you have demo video if you could share on working of this extension?? It would be helpful.

[tanish-7869]

@Suke2004 @BennyPerumalla update on the issue I am done with the local caching and avoiding multiple calls for same word. Will work on rest of the tasks.

[BennyPerumalla]

@tanish-7869 Good to hear that, do you till need demo video of working model?

[tanish-7869]

@BennyPerumalla Yes it would be better to validate I don't miss any use-case.