[SDK] custom hash and equality for attribute processor

[nikhilbhatia08]

Fixes #3635

nostd::string_view and access the value with key.data() but this may not be null terminated. So we need a custom hash and equality for the view container

Changes

Please provide a brief description of the changes here.

For significant contributions please make sure you have completed the following items:

• CHANGELOG.md updated for non-trivial changes
• Unit tests have been added
• Changes in public API reviewed

[codecov]

Codecov Report

❌ Patch coverage is 93.54839% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 90.10%. Comparing base (74d348a) to head (7320c35).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...ntelemetry/sdk/metrics/view/attributes_processor.h	85.72%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3643      +/-   ##
==========================================
+ Coverage   90.08%   90.10%   +0.03%     
==========================================
  Files         220      221       +1     
  Lines        7110     7129      +19     
==========================================
+ Hits         6404     6423      +19     
  Misses        706      706              

Files with missing lines	Coverage Δ
...de/opentelemetry/sdk/common/custom_hash_equality.h	100.00% <100.00%> (ø)
...ntelemetry/sdk/metrics/view/attributes_processor.h	84.62% <85.72%> (-1.87%)	⬇️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[nikhilbhatia08]

std::unordered_map does not support heterogeneous lookup is_transparent before cpp20, even with custom hash/equality older standard libraries (especially libc++ on macOS and pre cpp20) do not accept string_view keys without materializing a std::string. Any ideas on how to make this work with even before versions of cpp20 ?

[Copilot]

Pull Request Overview

This PR fixes a bug in the attributes processor where nostd::string_view keys were not being properly hashed and compared when used in std::unordered_map. The issue occurred because string_view.data() may not be null-terminated, causing incorrect lookups and comparisons.

• Implements custom StringViewHash and StringViewEqual functors for safe string_view operations
• Adds cross-platform heterogeneous lookup helper functions for compatibility
• Updates all unordered_map declarations to use the new custom hash and equality functors

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
sdk/include/opentelemetry/sdk/metrics/view/attributes_processor.h	Adds custom hash/equality functors and updates FilteringAttributesProcessor classes to use them
sdk/test/metrics/attributes_processor_test.cc	Updates test unordered_map declarations to use new custom functors
sdk/test/metrics/sum_aggregation_test.cc	Updates test unordered_map declarations to use new custom functors

[Copilot]

Creating a new string_view from an existing string_view is redundant and may not address the original issue. The problem is that string_view may not be null-terminated, but this code still uses the same data pointer. Consider using std::string construction instead: return std::hash<std::string>{}(std::string{sv.data(), sv.size()});

Suggested change

	return std::hash<opentelemetry::nostd::string_view>{}(
	opentelemetry::nostd::string_view{sv.data(), sv.size()});
	// Materialize to std::string to ensure safe hashing (string_view may not be null-terminated)
	return std::hash<std::string>{}(std::string{sv.data(), sv.size()});

[owent]

LGTM

[marcalff]

Please document where the new hash function comes from.

[marcalff]

LGTM, see minor changes.

[marcalff]

LGTM, thanks for the fix.