fix(provider): filter empty text blocks for base_url providers

[zongtingwei]

Issue for this PR

Closes #13811

Type of change

• Bug fix
• New feature
• Refactor / code improvement
• Documentation

What does this PR do?

This fixes an empty-content filtering gap in normalizeMessages().

Right now empty text/reasoning blocks are only filtered for specific providers. When base_url is used with an OpenAI-compatible provider that routes to Bedrock, that path can still send empty text blocks and Bedrock rejects the request with messages: text content blocks must be non-empty.

This change moves the empty-content filtering into the shared normalization path so it also applies to OpenAI-compatible base_url setups. It also adds regression tests for openai-compatible, anthropic, bedrock, and openai variants.

How did you verify your code works?

I verified it with:

bun test test/provider/transform.test.ts

Result: 120 pass

Screenshots / recordings

N/A, no UI changes.

Checklist

• I have tested my changes locally
• I have not included unrelated changes in this PR

[github-actions]

Thanks for your contribution!

This PR doesn't have a linked issue. All PRs must reference an existing issue.

Please:

1. Open an issue describing the bug/feature (if one doesn't exist)
2. Add Fixes #<number> or Closes #<number> to this PR description

See CONTRIBUTING.md for details.

[github-actions]

The following comment was made by an LLM, it may be inaccurate:

Based on my search, I found several related PRs that address similar empty content filtering issues:

Potentially Related PRs:

1. fix(opencode): filter empty text content blocks for all providers #17742 - fix(opencode): filter empty text content blocks for all providers

   • Directly addresses filtering empty text blocks across providers

2. fix(provider): handle empty content for bedrock/openai-compatible APIs #17396 - fix(provider): handle empty content for bedrock/openai-compatible APIs

   • Specifically handles empty content for Bedrock and OpenAI-compatible APIs

3. fix(provider): skip empty-text filtering for assistant messages in normalizeMessages (#16748) #16750 - fix(provider): skip empty-text filtering for assistant messages in normalizeMessages

   • Related to normalization logic but adds exceptions for assistant messages

4. fix: expand Anthropic detection and strip whitespace-only text blocks #12634 - fix: expand Anthropic detection and strip whitespace-only text blocks

   • Addresses whitespace-only text filtering (relevant given the PR's whitespace preservation decision)

5. fix: empty tool-result content and cache control level for custom @ai… #17363 - fix: empty tool-result content and cache control level for custom @ai…

   • Handles empty content in tool results

6. fix(opencode): filter empty text/reasoning parts when loading from database #19311 - fix(opencode): filter empty text/reasoning parts when loading from database

   • Filters empty text/reasoning parts

Why these are related: PR #21190 addresses empty text block filtering in the shared normalization path for base_url providers routing to Bedrock. The PRs above have tackled similar issues with empty content filtering across different provider types and contexts. #17742 and #17396 appear most directly related as they address the same core issue of empty text block handling for similar provider scenarios.

[github-actions]

Thanks for updating your PR! It now meets our contributing guidelines. 👍

[rekram1-node]

Automated PR Cleanup

Thank you for contributing to opencode.

Due to the high volume of PRs from users and AI agents, we periodically close older PRs using automated criteria so maintainers can focus review time on the most active and community-supported contributions.

This PR was closed because it matched the following cleanup criteria:

• The PR was created more than 1 month ago
• The PR had fewer than 2 positive reactions
• Positive reactions are counted as thumbs-up, heart, celebration, or rocket reactions on the PR

PRs created within the last month are not affected by this cleanup.

If you believe this PR was closed incorrectly, or if you are still actively working on it, please leave a comment explaining why it should be reopened. A maintainer can review and reopen it if appropriate.

Thanks again for taking the time to contribute.