chore: add codecov coverage tracking#493
Merged
xdustinface merged 1 commit intov0.42-devfrom Mar 4, 2026
Merged
Conversation
Contributor
📝 WalkthroughWalkthroughAdds repository-wide coverage: new Codecov config, CI script flag/logic to run per-crate cargo-llvm-cov, GitHub Actions steps to install llvm tools/generate LCOV and upload to Codecov, and README badges plus a per-crate coverage section. Changes
Sequence Diagram(s)sequenceDiagram
participant Dev as Developer
participant GH as GitHub Actions
participant CI as ci_config.py
participant Cargo as cargo / cargo-llvm-cov
participant Codecov as Codecov
Dev->>GH: push or open PR (coverage=true)
GH->>GH: install llvm-tools & cargo-llvm-cov
GH->>CI: run run-group with --coverage
CI->>CI: emit crate_flags
CI->>Cargo: "cargo llvm-cov clean --workspace"
loop per crate
CI->>Cargo: "cargo-llvm-cov --no-report -p <crate> --all-features"
Cargo-->>CI: produce per-crate coverage artifacts
end
CI->>Cargo: "cargo llvm-cov report --lcov"
Cargo-->>GH: produce lcov.info
GH->>Codecov: upload lcov.info (codecov-action)
Codecov-->>GH: update badges & reports
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~22 minutes Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
xdustinface
force-pushed
the
chore/coverage
branch
from
9f39c71 to
f9baf1a
Compare
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
.github/scripts/ci_config.py (1)
176-184: Consolidate duplicate coverage checks.The
if coverage:condition is checked twice (lines 178 and 181). These can be combined into a single block.♻️ Suggested consolidation
coverage = getattr(args, "coverage", False) if coverage: github_output("crate_flags", ",".join(crates)) - - if coverage: github_group_start("Cleaning coverage state") subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"]) github_group_end()🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/ci_config.py around lines 176 - 184, Combine the duplicate coverage checks by converting the two separate "if coverage:" blocks into one: after computing coverage = getattr(args, "coverage", False) perform a single if coverage: that calls github_output("crate_flags", ",".join(crates)) and then runs the cleanup group (github_group_start("Cleaning coverage state"), subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"]), github_group_end()). This keeps the same calls to github_output, github_group_start, subprocess.run and github_group_end but under one unified conditional.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@README.md`:
- Around line 14-15: The Codecov badge URLs currently reference the wrong GitHub
org "dashevo"; update both anchor href and image src URLs in the README badges
(the two <a>...<img> badge entries) to use "dashpay/rust-dashcore" instead of
"dashevo/rust-dashcore" for the master and v0.42-dev branches so both href and
src reflect the correct organization.
---
Nitpick comments:
In @.github/scripts/ci_config.py:
- Around line 176-184: Combine the duplicate coverage checks by converting the
two separate "if coverage:" blocks into one: after computing coverage =
getattr(args, "coverage", False) perform a single if coverage: that calls
github_output("crate_flags", ",".join(crates)) and then runs the cleanup group
(github_group_start("Cleaning coverage state"), subprocess.run(["cargo",
"llvm-cov", "clean", "--workspace"]), github_group_end()). This keeps the same
calls to github_output, github_group_start, subprocess.run and github_group_end
but under one unified conditional.
ℹ️ Review info
Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: e34e1e37-4b82-472f-a0ac-5f5a36cb638a
📒 Files selected for processing (5)
.codecov.yml.github/scripts/ci_config.py.github/workflows/build-and-test.yml.github/workflows/rust.ymlREADME.md
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (1)
README.md (1)
14-15:⚠️ Potential issue | 🟡 MinorUse the correct org (
dashpay) for all Codecov links.Line 14, Line 15, and the per-crate links (Line 27–Line 38) still point to
dashevo/rust-dashcore, which doesn’t match this repo (dashpay/rust-dashcore). Please update all new Codecov URLs consistently.Also applies to: 27-38
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@README.md` around lines 14 - 15, The README contains Codecov links pointing to the wrong GitHub org; update every Codecov URL that contains "dashevo/rust-dashcore" to "dashpay/rust-dashcore" (including the two top badges with alt="Coverage (master)" and alt="Coverage (develop)" and all per-crate badge/link entries referenced later), ensuring both href and src attributes are replaced consistently across the file.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/scripts/ci_config.py:
- Line 179: The script currently calls github_output("crate_flags",
",".join(crates)) before the skip/filter logic, causing skipped crates (e.g.,
dash-fuzz on Windows) to be advertised; move the github_output("crate_flags",
",".join(...)) invocation to after the code that computes the final list of
executed crates (the filtering/skip logic that mutates or produces the filtered
'crates' list) so it emits only actually-run crates, and apply the same change
for the other outputs mentioned around the same area (the calls on/near lines
186-188) so all exported flags reflect the post-filtered executed set. Ensure
you reference and use the final filtered variable (the same 'crates' or its
filtered counterpart) when calling github_output and not the original unfiltered
list.
- Around line 178-182: The cleanup subprocess.run call for "cargo llvm-cov clean
--workspace" currently ignores failures; update the logic in the section guarded
by the coverage variable to fail fast by invoking subprocess.run with check=True
(or capture the CompletedProcess and if result.returncode != 0 call sys.exit or
raise a RuntimeError), and ensure you log or include context before exiting;
reference the existing subprocess.run invocation and the surrounding
github_group_start/github_group_end calls so the cleanup either succeeds or the
CI stops immediately.
---
Duplicate comments:
In `@README.md`:
- Around line 14-15: The README contains Codecov links pointing to the wrong
GitHub org; update every Codecov URL that contains "dashevo/rust-dashcore" to
"dashpay/rust-dashcore" (including the two top badges with alt="Coverage
(master)" and alt="Coverage (develop)" and all per-crate badge/link entries
referenced later), ensuring both href and src attributes are replaced
consistently across the file.
ℹ️ Review info
Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 12cb21c6-5c1c-4b63-9a73-ae497d724edc
📒 Files selected for processing (5)
.codecov.yml.github/scripts/ci_config.py.github/workflows/build-and-test.yml.github/workflows/rust.ymlREADME.md
🚧 Files skipped from review as they are similar to previous changes (3)
- .github/workflows/build-and-test.yml
- .github/workflows/rust.yml
- .codecov.yml
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (1)
.github/scripts/ci_config.py (1)
179-188:⚠️ Potential issue | 🟡 MinorEmit
crate_flagsfrom executed crates, not configured crates.Line 179 exports all group crates before the skip at Lines 186-188, so skipped crates (e.g.,
dash-fuzzon Windows) can still be sent to Codecov flags.Suggested fix
- if coverage: - github_output("crate_flags", ",".join(crates)) + executed_crates = [] + if coverage: github_group_start("Cleaning coverage state") subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"], check=True) github_group_end() @@ if args.os == "windows-latest" and crate == "dash-fuzz": github_notice(f"Skipping {crate} on Windows (honggfuzz not supported)") continue + executed_crates.append(crate) github_group_start(f"Testing {crate}") @@ if result.returncode != 0: failed.append(crate) github_error(f"Test failed for {crate} on {args.os}") + + if coverage: + github_output("crate_flags", ",".join(executed_crates))🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/ci_config.py around lines 179 - 188, The script currently calls github_output("crate_flags", ",".join(crates)) before skipping platform-specific crates, causing skipped crates (like "dash-fuzz" on Windows) to be emitted; change the logic to record only actually executed crates: inside the for crate in crates loop (and after the if args.os == "windows-latest" and crate == "dash-fuzz" skip) append non-skipped crate names to a new executed_crates list, and then call github_output("crate_flags", ",".join(executed_crates)) after the loop (or at the point where you know skips are applied) so Codecov only receives flags for crates that ran.
🧹 Nitpick comments (1)
.github/scripts/ci_config.py (1)
180-182: Ensure coverage cleanup always closes the GitHub log group.With
check=Trueat Line 181, an exception bypasses Line 182 and leaves the group unclosed. Wrap withtry/finallyand return a controlled error.Suggested hardening
if coverage: github_output("crate_flags", ",".join(crates)) github_group_start("Cleaning coverage state") - subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"], check=True) - github_group_end() + try: + subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"], check=True) + except subprocess.CalledProcessError as e: + github_error(f"Failed to clean coverage state: {e}") + return e.returncode or 1 + finally: + github_group_end()🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/ci_config.py around lines 180 - 182, The github group opened by github_group_start("Cleaning coverage state") may remain unclosed if subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"], check=True) raises; wrap the subprocess.run call in a try/finally so github_group_end() always runs, and convert the raised exception into a controlled return/error value (or re-raise after logging) to ensure the group is closed; update the block around github_group_start / subprocess.run / github_group_end accordingly and reference these symbols (github_group_start, subprocess.run call with the "cargo llvm-cov clean" args, and github_group_end) when making the change.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In @.github/scripts/ci_config.py:
- Around line 179-188: The script currently calls github_output("crate_flags",
",".join(crates)) before skipping platform-specific crates, causing skipped
crates (like "dash-fuzz" on Windows) to be emitted; change the logic to record
only actually executed crates: inside the for crate in crates loop (and after
the if args.os == "windows-latest" and crate == "dash-fuzz" skip) append
non-skipped crate names to a new executed_crates list, and then call
github_output("crate_flags", ",".join(executed_crates)) after the loop (or at
the point where you know skips are applied) so Codecov only receives flags for
crates that ran.
---
Nitpick comments:
In @.github/scripts/ci_config.py:
- Around line 180-182: The github group opened by github_group_start("Cleaning
coverage state") may remain unclosed if subprocess.run(["cargo", "llvm-cov",
"clean", "--workspace"], check=True) raises; wrap the subprocess.run call in a
try/finally so github_group_end() always runs, and convert the raised exception
into a controlled return/error value (or re-raise after logging) to ensure the
group is closed; update the block around github_group_start / subprocess.run /
github_group_end accordingly and reference these symbols (github_group_start,
subprocess.run call with the "cargo llvm-cov clean" args, and github_group_end)
when making the change.
ℹ️ Review info
Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 0d425ee6-ecbc-4074-8faf-81d9640e25cc
📒 Files selected for processing (4)
.codecov.yml.github/scripts/ci_config.py.github/workflows/build-and-test.ymlREADME.md
🚧 Files skipped from review as they are similar to previous changes (1)
- .codecov.yml
xdustinface
force-pushed
the
chore/coverage
branch
2 times, most recently
from
12ab005 to
1b4e854
Compare
Inspired by #492 but integrated into the existing workflows.
xdustinface
force-pushed
the
chore/coverage
branch
from
1b4e854 to
f17bfc9
Compare
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (1)
.github/scripts/ci_config.py (1)
176-180:⚠️ Potential issue | 🟠 MajorReset coverage artifacts before emitting and collecting coverage flags.
Line 179 enables coverage reporting without first cleaning prior
llvm-covartifacts. With cached build state, coverage can be stale/inaccurate.Suggested fix
coverage = getattr(args, "coverage", False) if coverage: - github_output("crate_flags", ",".join(crates)) + github_group_start("Cleaning previous coverage state") + clean_result = subprocess.run(["cargo", "llvm-cov", "clean", "--workspace"]) + github_group_end() + if clean_result.returncode != 0: + github_error("Failed to clean coverage state") + return 1 + github_output("crate_flags", ",".join(crates))#!/bin/bash # Verify coverage flow and absence/presence of llvm-cov clean. rg -n -C3 'coverage|crate_flags|llvm-cov' .github/scripts/ci_config.py rg -n -C3 'rust-cache|shared-key|coverage|llvm-cov report' .github/workflows/build-and-test.yml🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.github/scripts/ci_config.py around lines 176 - 180, When coverage is enabled (the coverage variable checked in the block that calls github_output("crate_flags", ",".join(crates))), ensure you first remove stale llvm-cov artifacts before emitting crate_flags: add a cleanup step that deletes or resets previous coverage outputs (e.g., remove coverage directories or llvm-cov generated files/artifacts) and any cached build state that could skew results, then proceed to call github_output with the joined crates; update the logic around the coverage check so the cleanup always runs when coverage is true and before github_output is invoked.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In @.github/scripts/ci_config.py:
- Around line 176-180: When coverage is enabled (the coverage variable checked
in the block that calls github_output("crate_flags", ",".join(crates))), ensure
you first remove stale llvm-cov artifacts before emitting crate_flags: add a
cleanup step that deletes or resets previous coverage outputs (e.g., remove
coverage directories or llvm-cov generated files/artifacts) and any cached build
state that could skew results, then proceed to call github_output with the
joined crates; update the logic around the coverage check so the cleanup always
runs when coverage is true and before github_output is invoked.
ℹ️ Review info
Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: ddba4591-9e37-453a-8a11-7cdd3779e46d
📒 Files selected for processing (5)
.codecov.yml.github/scripts/ci_config.py.github/workflows/build-and-test.yml.github/workflows/rust.ymlREADME.md
🚧 Files skipped from review as they are similar to previous changes (1)
- .github/workflows/rust.yml
QuantumExplorer
approved these changes
Mar 4, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Inspired by #492 but integrated into the existing workflows.
Summary by CodeRabbit
Documentation
Chores