fix: improve handling of net::ERR_ABORTED errors in URL fetching

[roomote]

Summary

This PR fixes issue #6632 where users were experiencing net::ERR_ABORTED errors when trying to fetch URL content using the @https:// mention syntax.

Problem

When users tried to fetch content from URLs like https://daisyui.com/llms.txt, they would get an error:

Error fetching content for https://daisyui.com/llms.txt: Failed to fetch URL content: net::ERR_ABORTED at https://daisyui.com/llms.txt

The error was not being handled properly, and the retry logic was not optimized for this specific error type.

Solution

1. Added specific handling for net::ERR_ABORTED errors in UrlContentFetcher.ts:

   • When this error is detected, the code now attempts a more aggressive retry with just domcontentloaded event and a shorter timeout (10 seconds)
   • If the retry also fails, a more descriptive error message is provided

2. Improved error messages for users:

   • Added a new translation key url_request_aborted with a helpful message explaining possible causes
   • The error message now suggests checking if the URL is accessible in a regular browser

3. Added comprehensive tests to ensure the new error handling works correctly

Changes

• src/services/browser/UrlContentFetcher.ts: Added specific handling for ERR_ABORTED errors
• src/core/mentions/index.ts: Added error message mapping for ERR_ABORTED
• src/i18n/locales/en/common.json: Added new translation for the error message
• src/services/browser/__tests__/UrlContentFetcher.spec.ts: Added tests for the new error handling

Testing

• All existing tests pass ✅
• Added new tests specifically for ERR_ABORTED error handling ✅
• Manually tested with URLs that might trigger this error

Fixes #6632

---

Important

Improves handling of net::ERR_ABORTED errors in URL fetching by adding retries and enhancing error messages.

• Behavior:
  • Adds handling for net::ERR_ABORTED in UrlContentFetcher.ts, retrying with domcontentloaded event and shorter timeout.
  • Provides descriptive error message if retry fails.
• Error Messages:
  • Adds url_request_aborted translation key in common.json for multiple locales.
• Testing:
  • Adds tests in UrlContentFetcher.spec.ts for ERR_ABORTED handling, ensuring retries and error message behavior.

^{This description was created by for ecc9286. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Reviewed my own code. Found it surprisingly coherent for something I wrote while the browser was actively aborting my attempts.

[roomote]

The error handling here catches all errors in the retry block and converts them to a generic message. Would it be helpful to preserve more specific error types (like ERR_CONNECTION_REFUSED vs ERR_TIMEOUT) in the final error message? This could help users understand why the retry failed.

[roomote]

I notice the retry logic for ERR_ABORTED is similar to the existing retry logic but with different parameters. Could we extract this into a reusable helper method to reduce code duplication? Something like retryNavigation(url, timeout, waitUntil) perhaps?

[roomote]

Minor: Using console.error here while line 133 uses console.warn for other retries. Should we use consistent logging levels for similar retry operations?

[roomote]

Consider extracting this timeout value to a constant like URL_FETCH_ABORTED_RETRY_TIMEOUT for better maintainability.

[roomote]

Great test coverage! One edge case to consider: what happens when the initial request fails with ERR_ABORTED but the retry succeeds after a delay? This could simulate a temporarily blocked request that becomes available.

[daniel-lxs]

@roomote-agent use translate mode to create the other translations for the rest of the languages

[roomote]

Hi @daniel-lxs! I see your request to translate the new error message for all other languages. I'll switch to translate mode and handle this for you. Let me check out the PR branch and get started.

[roomote]

✅ @daniel-lxs I've successfully added the url_request_aborted translations for all 18 supported languages!

Summary of changes:

• Added translations for the new error message in all language files (ca, de, es, fr, hi, id, it, ja, ko, nl, pl, pt-BR, ru, tr, vi, zh-CN, zh-TW)
• Each translation provides helpful context about possible causes (blocked access, authentication requirements, network issues)
• All translations have been validated using the find-missing-translations.js script
• All CI checks have passed successfully, including the check-translations test

The translations maintain consistency with the existing tone and style for each language while accurately conveying the error message and suggested troubleshooting steps.

[daniel-lxs]

@roomote-agent address the review, push to this PR, do not create a new one

[roomote]

Hi @daniel-lxs! I see your request to address the review comments. I found 5 inline review suggestions that I'll implement:

1. Preserve specific error types in retry error messages
2. Extract retry logic into a reusable helper method
3. Use consistent logging levels
4. Extract timeout value to a constant
5. Add test case for successful retry after ERR_ABORTED

I'll work on these improvements and push the changes to this PR branch. Starting now!

[roomote]

✅ @daniel-lxs All review comments have been successfully addressed and pushed to this PR!

Changes made:

1. ✅ Extracted retry logic into a reusable retryNavigation() helper method
2. ✅ Added constant URL_FETCH_ABORTED_RETRY_TIMEOUT (10 seconds) for better maintainability
3. ✅ Fixed logging consistency - changed console.error to console.warn for retry operations
4. ✅ Preserved specific error types in retry error messages (e.g., "ERR_CONNECTION_REFUSED - net::ERR_CONNECTION_REFUSED")
5. ✅ Added test case for successful retry after ERR_ABORTED error

Test Results:

• All 18 tests passing ✅
• CI checks all green ✅

The code is now cleaner, more maintainable, and provides better error information to users. Thanks for the great review suggestions!

[daniel-lxs]

LGTM