Skip plotting benchmarks without data

[bernhardmgruber]

Fixes: #6729

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 595b0631-a5d4-492e-becc-0d55c728fd84

📥 Commits

Reviewing files that changed from the base of the PR and between 0fd8182 and 45e3daa.

📒 Files selected for processing (1)

• benchmarks/scripts/sol.py

---

📝 Walkthrough

Summary by CodeRabbit

• Bug Fixes
  • Improved robustness of benchmark data processing with stronger validation and conditional filtering to avoid empty results.
  • Added warnings and graceful handling when individual benchmarks or sub-benchmarks produce missing or invalid data.
  • Stops aggregation/plotting and reports an error when all benchmarks are skipped, preventing misleading outputs.

important:

Walkthrough

benchmarks/scripts/sol.py adds defensive filtering and validation: offset-type filtering no longer collapses dfs to empty, alg_dfs skips sub-benchmarks with empty data or missing bw, and sol() returns early if all benchmarks are skipped.

Changes

Data Validation and Skipping

Layer / File(s)	Summary
Offset-type filter and empty-result check benchmarks/scripts/sol.py	filter_by_offset_type filters into a temporary subset and only replaces df when that subset is non-empty (lines 30-34).
alg_dfs validation and skip logic benchmarks/scripts/sol.py	alg_dfs computes fused_algname early, warns and skips sub-benchmarks when the post-filter df is empty or when bw has no non-null values, and defers variant coercion and bw scaling until after checks (lines 63-76).
sol early-return on all skipped benchmarks/scripts/sol.py	sol() uses alg_dfs result and prints an error and returns early if no benchmarks remain (lines 174-178).

Assessment against linked issues

Objective	Addressed	Explanation
Skip benchmarks lacking bandwidth data [#6729]	✅

Out-of-scope changes

Code Change	Explanation
32-bit offset-type filtering refactor (benchmarks/scripts/sol.py, lines 30-34)	This change improves offset-type filtering behavior but does not address missing bw data referenced in #6729.

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: ea645a47-557f-4498-8d12-c8cfb4ef03fe

📥 Commits

Reviewing files that changed from the base of the PR and between 589097d and 0fd8182.

📒 Files selected for processing (1)

• benchmarks/scripts/sol.py

[coderabbitai]

⚠️ Potential issue | 🟠 Major | ⚡ Quick win

important: Guard for missing bw column before indexing at Line 70. Right now df["bw"] can still raise KeyError if the benchmark output omits bw entirely (not just NaN values), which defeats the hardening in this PR. Use a schema check in the skip condition, e.g. treat ("bw" not in df.columns) the same as missing bandwidth data.

[bernhardmgruber]

I think a KeyError is appropriate, since then the input is wrong.

[coderabbitai]

@bernhardmgruber Understood — the five affected benchmarks have bw present but all-NaN, so dropna().empty is the right check for the actual issue. A fully absent bw column is structurally malformed input and a KeyError is a reasonable loud failure there. Fair enough.

[NaderAlAwar]

Important: alg_dfs() can now return an empty dict when all selected benchmarks are skipped, which causes alg_bws() to crash because medians remains None. We should add an early return with an error or warning message