Varchar 1 fix on parallel ingestion.

[khustup2]

🚀 🚀 Pull Request

Impact

• Bug fix (non-breaking change which fixes expected existing functionality)
• Enhancement/New feature (adds functionality without impacting existing logic)
• Breaking change (fix or feature that would cause existing functionality to change)

Description

Things to be aware of

Things to worry about

Additional Context

[claude]

Claude finished @khustup2's task in 1m 21s —— View job

---

PR Review Complete ✓

Summary

The fix correctly addresses the bug where format_type_be() was dropping type modifiers (like the length in VARCHAR(1)), causing catalog sync issues in parallel ingestion scenarios. The change to format_type_with_typemod() properly preserves type modifiers when storing column metadata in the shared catalog.

What I verified:

• ✓ Core C++ fix in table_storage.cpp:261 is correct and minimal
• ✓ Other usages of format_type_be() are in error messages only (appropriate usage)
• ✓ Comprehensive regression test added with clear documentation of the bug
• ✓ Test validates the critical path: Instance B writing to a VARCHAR(1) column discovered from catalog

Found 1 minor issue - see inline comment about bidirectional sync validation.

---

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[claude]

Test doesn't verify bidirectional sync: Instance A can't see Instance B's writes. Add a verification that primary_conn can SELECT the 3 rows after Instance B inserts them (before cleanup), confirming the catalog sync works both ways.