Fix empty %b/%s argument dropped in redis command formatting (#2275)

[two-headBoy]

What problem does this PR solve?

Issue Number: resolve #2275

Problem Summary:
An empty value passed as a %b/%s argument was silently dropped, producing a Redis command with the wrong number of arguments and causing "wrong number of arguments" errors.

What is changed and the side effects?

Changed:

• Empty %b/%s arguments now emit an empty bulk string instead of being dropped.
• %s with NULL, and %b with NULL and non-zero size, return an error to avoid undefined behavior.
• Added unittests.

Side effects:

• Performance effects: none.

• Breaking backward compatibility: none.

---

Check List:

• Please make sure your changes are compilable.
• When providing us with a new feature, it is best to add related tests.
• Please follow Contributor Covenant Code of Conduct.

[Copilot]

Copilot review overview

🟡 Changes recommended

A critical variadic pointer-type issue and a moderate binary-data test gap remain unresolved.

Get a fresh assessment by requesting another Copilot review.

Review effort: Lite
Findings: 1 · 1

Open (2)

• Fix variadic pointer type mismatch for const char* arguments · New
• Compare binary output bytes instead of using ASSERT_STREQ · New

What changed in this PR

Fixes Redis %s/%b formatting so empty arguments produce empty bulk strings and invalid null arguments return errors.

Changes:

• Preserves empty format arguments.
• Adds null-pointer validation.
• Adds regression tests for empty and binary arguments.

File	Summary
test/​brpc_redis_unittest.cpp	Adds regression coverage; binary payload assertion should compare serialized bytes. Moderate issue (4 votes).
src/​brpc/​redis_command.cpp	Updates argument handling and validation; variadic pointer typing causes undefined behavior. Critical issue (1 vote).

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

Copilot review overview

🟡 Changes recommended

Address the variadic pointer compatibility issue and strengthen binary payload test coverage.

Get a fresh assessment by requesting another Copilot review.

Review effort: Lite
Findings: 1

Open (1)

• Avoid variadic char* and const char* type mismatch · New

Resolved since last review (2)

• Fix variadic pointer type mismatch for const char* arguments
• Compare binary output bytes instead of using ASSERT_STREQ

[Copilot]

Copilot review overview

🟡 Changes recommended

Resolve the variadic pointer type mismatches for mutable %s and %b buffers.

Get a fresh assessment by requesting another Copilot review.

Review effort: Lite
Findings: 2

Open (2)

• Undefined variadic type mismatch for mutable %b buffers · New
• Avoid variadic char* and const char* type mismatch

[wwbmmm]

[replied by brpc-oncall robot] The empty %s/%b fix is correct and the binary test now compares full bytes; the variadic char*/const char* concern is now documented as a pre-existing, unfixable-without-refactor limitation of this API, which I agree is out of scope here, and the only remaining observation is that the new EINVAL paths still abort when reached through RedisRequest::AddCommand.