fix resolver parsing

[dogancanbakir]

closes #2350

Summary by CodeRabbit

• Improvements
  • Resolver configuration parsing now trims whitespace and accepts comma-separated entries, making multi-value resolver lines easier to write and more forgiving.
  • When a resolver argument appears to be a file path but the file is missing, a warning is logged while still retaining the argument, helping surface potential configuration issues without discarding input.

[coderabbitai]

Walkthrough

Updates resolver handling in runner/options.go: lines read from resolver files are trimmed; comma-separated values on a line are split into individual trimmed resolvers; empty items are ignored. When a resolver argument references a non-existent file but looks like a path (contains / or \), a warning is logged and the argument is still added.

Changes

Cohort / File(s)	Summary
Resolver parsing & argument handling runner/options.go	Trim lines when reading resolver files, split comma-separated values into individual trimmed resolver entries, drop empty items, and log a warning (but still add) when a resolver argument appears to be a file path and does not exist.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Poem

🐰 I hopped through lines both short and long,
I trimmed and split what once was wrong,
Commas scattered, now each one stays,
A warning whispers through my maze,
Resolvers found — I twitch with song! 🎶

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'fix resolver parsing' directly matches the main objective of the PR, which is to fix resolver file parsing to be consistent regardless of path type.
Linked Issues check	✅ Passed	The changes implement comma-separated resolver handling and file path validation with warnings, directly addressing issue #2350's requirement for consistent resolver parsing.
Out of Scope Changes check	✅ Passed	All changes in runner/options.go are scoped to resolver processing and validation, directly supporting the linked issue objective.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch 2350_fix_resolver_parsing

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

📜 Review details

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 0cbc12b and 8222571.

📒 Files selected for processing (1)

• runner/options.go

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (8)

• GitHub Check: Test Builds (macOS-latest)
• GitHub Check: Test Builds (windows-latest)
• GitHub Check: Test Builds (ubuntu-latest)
• GitHub Check: Functional Test (ubuntu-latest)
• GitHub Check: Functional Test (windows-latest)
• GitHub Check: Functional Test (macOS-latest)
• GitHub Check: Analyze (go)
• GitHub Check: release-test

🔇 Additional comments (2)

runner/options.go (2)

737-748: Good enhancement to resolver file parsing flexibility.

The addition of whitespace trimming and comma-separated value splitting makes the resolver file parser more robust and user-friendly. It now handles multiple resolver formats gracefully.

---

740-740: No action required. The project's go.mod specifies Go 1.24.1, which fully supports strings.SplitSeq (introduced in Go 1.23). The code is compatible with the project's minimum Go version.

[Mzack9999]

The issue seems unrelated to the file parsing, but rather to the file not existing on the file system (I guess maybe a typo from the user), as in fact it gets appended to the slice directly (previous line 741). The resolver file format (one resolver per line) is common to all tools, if we want to support comma separated resolvers on the same line within file, maybe we should move this to utils. What do you think?

[dogancanbakir]

It makes sense. I'll open an issue in utils to track this.

[dogancanbakir]

Also, just asked user to check the path and it seem this isn't a path issue #2350 (comment)

[Mzack9999]

looks like the issue persists even with this change. For some reason the path lead to a non-existing file check. Maybe we can try to enforce additional checks on the resolver arguments (ex. presence of \ or /) which are prohibited as qualified domain name, and warning out the user and using default resolvers in case or erroring out since a resolver file was requested (ex. curl has hard fail when -dns-server is used and fails, without automatic fallbacks, but maybe we should be more fault tolerant and ease automation). What do you think?

[Mzack9999]

suggested approch