Added language detection and KV-based usage analytics

[hemanth5055]

Description:

This PR introduces a robust analytics and language detection system for the code translation feature. It provides real-time tracking of translation usage and enables detailed insights into which language pairs are most frequently used. The main updates include:

1. KV-based Analytics Tracking:

   • Every translation request now records the source and target language pair in Cloudflare Workers KV (LANG_TRANSLATION_ANALYTICS).
   • Each KV entry maintains a usage count that increments with every translation request.
   • Keys are normalized (lowercase and trimmed) to avoid duplicates caused by inconsistent casing or whitespace.

2. Source Language Detection:

   • Automatically detects the programming language of submitted code using Google Gemini AI before performing translation.
   • Guarantees accurate analytics tracking for multiple programming languages.

3. /v1/analytics Endpoint:

   • Introduces a new GET endpoint to fetch all translation analytics stored in KV.
   • Returns a JSON object where keys represent source-target language pairs and values indicate their usage counts.
   • Enables monitoring of translation usage patterns and identification of popular language combinations.

---

Testing:

• Verified that translation requests increment analytics counts correctly in KV.

• Tested the /v1/analytics endpoint to ensure all keys and counts are returned accurately.

• Confirmed that source language detection reliably identifies programming languages for various code snippets.

---

Impact:

• Enables comprehensive tracking of translation usage and helps identify the most popular language pairs.
• Provides actionable insights into feature usage for monitoring, optimization, and future improvements.
• Improves reliability by ensuring KV writes are awaited and keys are normalized.

---

Next Steps / Suggestions for Reviewers:

• Please review the KV write logic and analytics counting mechanism.
• Confirm that language detection works accurately for all supported languages.
• Suggest any improvements for performance or code readability.

Before testing or running the Worker locally, please add your KV namespace ID for LANG_TRANSLATION_ANALYTICS in your wrangler.json under the KV binding section.

closes #20

[Copilot]

Pull Request Overview

This PR introduces a comprehensive analytics system for the code translation feature, tracking usage patterns through Cloudflare Workers KV storage. It implements automatic source language detection using Google Gemini AI and provides an analytics endpoint for monitoring translation usage.

• Added KV-based analytics tracking that records source-target language pairs with usage counts
• Implemented automatic source language detection for submitted code using AI
• Created a new /v1/analytics endpoint to retrieve usage statistics

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
backend/wrangler.jsonc	Added KV namespace binding for analytics storage
backend/src/index.ts	Implemented analytics tracking, language detection, and analytics endpoint

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

The analytics key should be normalized (lowercase and trimmed) to avoid duplicates, similar to how the rate limit key is handled on line 19.

Suggested change

	const key = `${source}-${dest}`;
	const normalizedSource = source.trim().toLowerCase();
	const normalizedDest = dest.trim().toLowerCase();
	const key = `${normalizedSource}-${normalizedDest}`;

[Copilot]

The detectLanguage function should normalize the returned language name to lowercase to ensure consistent analytics keys, preventing duplicate entries like 'Python' vs 'python'.

Suggested change

	return result.response.text().trim();
	return result.response.text().trim().toLowerCase();

[Copilot]

JSON.parse should be wrapped in a try-catch block to handle potential parsing errors, similar to the error handling in updateAnalytics function.

Suggested change

	stats[key.name] = JSON.parse(val || '{}');
	try {
	stats[key.name] = JSON.parse(val || '{}');
	} catch (e) {
	console.error(`Failed to parse analytics value for key "${key.name}":`, e);
	stats[key.name] = {};
	}

[hemanth5055]

Hey @dineshsutihar , I have made changes according to co-pilot review , you can merge it now.

[dineshsutihar]

Nice work, @hemanth5055! I'm going to merge it now.
In the future, I think we can optimize this by sending the source language from the extension, so we won't need to call Gemini to detect the language.

[hemanth5055]

Nice work, @hemanth5055! I'm going to merge it now. In the future, I think we can optimize this by sending the source language from the extension, so we won't need to call Gemini to detect the language.

Hey @dineshsutihar,
I was thinking along the same lines. We can modify Gemini to return a JSON response like this:

{
  "translation": "",
  "source_language": ""
}

This way, we can eliminate the extra API call that’s currently used just for language detection.
If you like this approach, let me know — I can raise a new issue for it.

[dineshsutihar]

Yes, that sounds like a solid approach, @hemanth5055. Please go ahead and raise a new issue for it.