[FR] Automated Aggregation of Multiple Test Failures in GoogleTest

[SuperAdam47]

Does the feature exist in the most recent commit?

No.

The current GoogleTest API does not provide a supported way to automatically capture, aggregate, and programmatically inspect multiple failures within a single test scope. Existing helpers such as EXPECT_NONFATAL_FAILURE and EXPECT_FATAL_FAILURE are limited to validating only one expected failure.

Why do we need this feature?

GoogleTest allows multiple EXPECT_* assertions to fail within one test, but there is no automation-friendly way to:

• Capture all failures in a defined scope
• Validate multiple expected failures in one test
• Access structured failure metadata programmatically

This limitation makes it difficult to:

• Build higher-level testing frameworks on top of GoogleTest
• Perform meta-testing (testing tests)
• Validate error-reporting behavior
• Generate structured CI or tooling outputs

Describe the proposal.

Introduce an opt-in failure aggregation mechanism that:

• Captures all fatal and non-fatal failures within a scoped block
• Exposes structured failure data (message, type, file, line)
• Allows validating multiple failures in a single test
• Does not alter existing test behavior or outputs unless explicitly enabled

GTEST_CAPTURE_FAILURES({
  EXPECT_EQ(a, b);
  EXPECT_TRUE(x);
});

EXPECT_FAILURE_COUNT(2);

Is the feature specific to an operating system, compiler, or build system version?

No. The feature is:

• Platform-independent
• Compiler-agnostic
• Build-system neutral (works with CMake, Bazel, etc.)

It would rely only on GoogleTest’s internal failure reporting mechanisms and be implemented in a portable, cross-platform manner.