[bot] Fix Style/StringHashKeys by 6[bot] · Pull Request #880 · 6/nitrocop

6 · 2026-03-30T05:50:37Z

Automated fix fix for Style/StringHashKeys corpus conformance.

Details


Cop	`Style/StringHashKeys`
Backend	codex / hard
Model	gpt-5.4 (xhigh)
Mode	fix
Turns	1
Run	#432

Result

Updated [string_hash_keys.rs](/home/runner/work/nitrocop/nitrocop/src/cop/style/string_hash_keys.rs) so `Style/StringHashKeys` now flags Prism `SourceFileNode` keys, which is the missing `__FILE__ => ...` case RuboCop catches. I also fixed [offense.rb](/home/runner/work/nitrocop/nitrocop/tests/fixtures/cops/style/string_hash_keys/offense.rb) to use valid minimal `__FILE__` reproductions with correctly aligned `^` markers.

Validation passed with `cargo test --lib -- cop::style::string_hash_keys`, and the required corpus check `python3 scripts/check_cop.py Style/StringHashKeys --rerun --clone --sample 15` reported `0` new FP, `0` new FN, and `2` resolved FN. Commit: `a35f4921` (`Fix Style/StringHashKeys __FILE__ keys`).

Changes

  2 files changed

Agent conversation excerpt

Reading docs/agent-ci.md first, then I’ll inspect the current cop, RuboCop’s implementation, and the existing fixtures to pin down the two false negatives before editing anything.

exec_command: sed -n '1,220p' docs/agent-ci.md

exec_command: sed -n '1,260p' src/cop/style/string_hash_keys.rs

exec_command: sed -n '1,240p' vendor/rubocop/lib/rubocop/cop/style/string_hash_keys.rb

exec_command: sed -n '1,260p' vendor/rubocop/spec/rubocop/cop/style/string_hash_keys_spec.rb

exec_command: sed -n '1,220p' tests/fixtures/cops/style/string_hash_keys/offense.rb

exec_command: sed -n '1,260p' tests/fixtures/cops/style/string_hash_keys/no_offense.rb

I have the current implementation and the two failing shapes. Next I’m checking how Prism parses __FILE__ => ... in both contexts, because these FNs usually mean the cop is only looking at one node kind or one traversal path.

exec_command: nl -ba src/cop/style/string_hash_keys.rs | sed -n '1,260p'

exec_command: rg -n "__FILE__|and_return\(__FILE__" -n tests vendor src -S

exec_command: rg -n "parse|prism" scripts src tests | head -n 40

The fixtures already contain the two missed cases, which is useful. I’m drilling into the visitor and Prism node shapes now so I can make the smallest possible detection change and keep the existing 688k matches stable.

exec_command: nl -ba src/cop/style/string_hash_keys.rs | sed -n '80,260p'

exec_command: nl -ba src/cop/style/string_hash_keys.rs | sed -n '260,360p'

exec_command: rg -n "SourceFileNode|source_file_node|__FILE__" src/cop -S

exec_command: rg -n "debug.*prism|print.*ast|as_source_file_node|pretty_print" src tests scripts -S

exec_command: cargo test --lib -- cop::style::string_hash_keys

The targeted test run is in progress. In parallel I’m preparing to inspect the AST shape for __FILE__, because the likely gap is that Prism represents it differently from quoted strings and the visitor only handles StringNode.

exec_command: rg -n "fn .*source_file_node|source_file_node\(" ~/.cargo/registry/src -S | head -n 40

exec_command: rg -n "debug: prism parse|debug: AST walk|RUST_LOG|debug" src/linter.rs src/main.rs -S

exec_command: cargo metadata --format-version 1 --no-deps | jq -r '.packages[] | select(.name=="nitrocop") | .manifest_path'

exec_command: cargo metadata --format-version 1 | jq -r '.packages[] | select(.name=="ruby-prism") | .manifest_path'

exec_command: rg -n "struct SourceFileNode|enum Node|fn as_source_file_node|SourceFileNode" /home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/ruby-prism-1.9.0 -S

write_stdin: ``

The failure is confirmed and isolated: the fixture expects 12 offenses, the cop only emits 10, exactly missing the two __FILE__ keys. I’m checking the generated Prism node API next so I can patch this without broadening the matcher beyond RuboCop’s behavior.

exec_command: find target/debug/build -path '*ruby-prism*' -type f | rg '/out/|generated|node' -n

exec_command: rg -n "SourceFileNode" target/debug/build /home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/ruby-prism-1.9.0 -S | head -n 60

exec_command: sed -n '14127,14220p' target/debug/build/ruby-prism-a0d940e303385573/out/bindings.rs

exec_command: sed -n '2670,2690p' target/debug/build/ruby-prism-sys-b415b3d38a1d5133/out/bindings.rs

Plan:

Confirm the failing cases are specifically __FILE__ hash keys and isolate the Prism node type involved.
Patch src/cop/style/string_hash_keys.rs to treat that node like RuboCop does, while preserving the existing narrow exemptions.
Re-run the targeted cop test.
Run the required corpus check with scripts/check_cop.py.
Commit only the cop source and its fixture files if validation stays clean.

Plan:

Confirm the failing cases are specifically __FILE__ hash keys and isolate the Prism node type involved.
Patch src/cop/style/string_hash_keys.rs to treat that node like RuboCop does, while preserving the existing narrow exemptions.
Re-run the targeted cop test.
Run the required corpus check with scripts/check_cop.py.
Commit only the cop source and its fixture files if validation stays clean.

update_plan

I’ve narrowed the gap to SourceFileNode handling. Before editing, I’m doing a quick RuboCop sanity check on the exact __FILE__ patterns so the patch stays aligned with the source of truth.

exec_command: printf '%s ' '__FILE__ => [1]' > /tmp/string_hash_keys_fn1.rb && rubocop --only Style/StringHashKeys /tmp/string_hash_keys_fn1.rb

exec_command: printf '%s ' 'expect(Coverage).to receive(:result).once.and_return(__FILE__ => [0, 1])' > /tmp/string_hash_keys_fn2.rb && rubocop --only Style/StringHashKeys /tmp/string_hash_keys_fn2.rb

write_stdin: ``

One thing surfaced from the RuboCop check: the standalone __FILE__ => [...] snippet isn’t valid Ruby on its own. I’m going to tighten the fixture to a valid minimal reproduction of the real corpus case before patching the cop, so the test exercises the actual AST shape instead of an error-recovery parse.

exec_command: nl -ba tests/fixtures/cops/style/string_hash_keys/offense.rb

exec_command: printf '%s ' 'assert_equal({' ' __FILE__ => [1]' '})' > /tmp/string_hash_keys_fn1_valid.rb && rubocop --only Style/StringHashKeys /tmp/string_hash_keys_fn1_valid.rb

I’m correcting the new fixture lines themselves too. Both were added with placeholder ^ markers at column 0, so they won’t validate the real source locations until those columns are fixed.

exec_command: python3 - <<'PY' line = 'h = { __FILE__ => [1] }' print(line) print(line.index('__FILE__')) line2 = 'expect(Coverage).to receive(:result).once.and_return(__FILE__ => [0, 1])' print(line2) print(line2.

exec_command: printf '%s ' 'h = { __FILE__ => [1] }' > /tmp/string_hash_keys_fn1_simple.rb && rubocop --only Style/StringHashKeys /tmp/string_hash_keys_fn1_simple.rb

I’m editing two places now: the offense fixture to use valid, correctly annotated __FILE__ reproductions, and the cop to treat Prism SourceFileNode keys like RuboCop’s string-key handling.

apply_patch

The patch is in place. I’m running the targeted cop test now to confirm the two __FILE__ cases are detected and that the existing exemptions still hold.

Task prompt (3482 tokens)

Fix Style/StringHashKeys — 0 FP, 2 FN

Instructions

You are fixing ONE cop in nitrocop, a Rust Ruby linter that uses Prism for parsing.

Current state: 688,706 matches, 0 false positives, 2 false negatives.
Focus on: FN (RuboCop flags code nitrocop misses).

⚠ 688,706 existing matches must not regress. Validate with check_cop.py before committing.

Workflow

Read the Pre-diagnostic Results and Corpus FP/FN Examples sections below first
Verify with RuboCop first (for FP fixes): before writing any code, confirm RuboCop's
behavior on BOTH the specific FP case AND the general pattern:
```
echo '<specific FP case>' > /tmp/test.rb && rubocop --only Style/StringHashKeys /tmp/test.rb
echo '<general pattern>' > /tmp/test.rb && rubocop --only Style/StringHashKeys /tmp/test.rb
```
If RuboCop flags the general pattern, your fix must be narrow enough to not suppress it.
Add a test case FIRST:
- FN fix: add the missed pattern to tests/fixtures/cops/style/string_hash_keys/offense.rb with ^ annotation
- FP fix: add the false-positive pattern to tests/fixtures/cops/style/string_hash_keys/no_offense.rb
Verify test fails: cargo test --lib -- cop::style::string_hash_keys
Fix src/cop/style/string_hash_keys.rs
Verify test passes: cargo test --lib -- cop::style::string_hash_keys
Validate against corpus (REQUIRED before committing):
```
python3 scripts/check_cop.py Style/StringHashKeys --rerun --clone --sample 15
```
If this reports FP or FN regression, your fix is too broad — narrow it down.
Add a /// doc comment on the cop struct documenting what you found and fixed
Commit only your cop's files

Fixture Format

Mark offenses with ^ markers on the line AFTER the offending source line.
The ^ characters must align with the offending columns. The message format is Style/StringHashKeys: <message text>.
See the Current Fixture sections below for real examples from this cop.

If your test passes immediately

If you add a test case and it passes without code changes, the corpus mismatch is
caused by config/context differences, not a detection bug.
Do NOT loop trying to make the test fail. Instead:

Investigate config resolution (Include/Exclude, cop enablement, disable comments)
The fix is likely in src/config/ or the cop's config handling, not detection logic
If you cannot determine the root cause within 5 minutes, document your findings as
a /// comment on the cop struct and commit

CRITICAL: Avoid regressions in the opposite direction

When fixing FPs, your change MUST NOT suppress legitimate detections. When fixing FNs,
your change MUST NOT flag code that RuboCop accepts. A fix that eliminates a few issues
in one direction but introduces hundreds in the other is a catastrophic regression.

Before exempting a category of patterns, verify with RuboCop that the general case
is still an offense:

rubocop --only Style/StringHashKeys /tmp/test.rb

If RuboCop flags the general pattern but not your specific case, the difference is in
a narrow context (e.g., enclosing structure, receiver type, argument count) — your fix
must target that specific context, not the broad category.

Rule of thumb: if your fix adds an early return or continue that skips a whole
node type, operator class, or naming pattern, it's probably too broad. Prefer adding a
condition that matches the SPECIFIC differentiating context.

Rules

Only modify src/cop/style/string_hash_keys.rs and tests/fixtures/cops/style/string_hash_keys/
Run cargo test --lib -- cop::style::string_hash_keys to verify your fix (do NOT run the full test suite)
Run python3 scripts/check_cop.py Style/StringHashKeys --rerun --clone --sample 15 before committing to catch regressions
Do NOT touch unrelated files
Do NOT use git stash
Do NOT push — you do not have push permission; the workflow handles pushing after you exit

Prism Notes

hash splits into HashNode (literal {}) and KeywordHashNode (keyword args foo(a: 1)). If you handle one, check if you need the other.
const splits into ConstantReadNode (simple Foo) and ConstantPathNode (qualified Foo::Bar). If you handle one, check if you need the other.

Current Fixture: offense.rb

tests/fixtures/cops/style/string_hash_keys/offense.rb

{ 'foo' => 1 }
  ^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

{ 'bar' => 2, 'baz' => 3 }
  ^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.
              ^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

x = { 'key' => 'value' }
      ^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

# String keys in a hash that is the receiver of gsub (not an argument)
{ 'expiration' => time, 'conditions' => conds }.to_json.gsub("\n", "")
  ^^^^^^^^^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.
                        ^^^^^^^^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

# String keys in a hash nested inside an array argument of IO.popen
IO.popen([{"FOO" => "bar"}, "ruby", "foo.rb"])
           ^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

# String keys in a block inside spawn/system (not direct arg)
system("cmd") do
  x = { 'inner' => 1 }
        ^^^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.
end

# Non-identifier string keys are also flagged (RuboCop autocorrects to :"Content-Type" etc.)
{ "Content-Type" => "text/html" }
  ^^^^^^^^^^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.
{ "foo bar" => 1 }
  ^^^^^^^^^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

Current Fixture: no_offense.rb

tests/fixtures/cops/style/string_hash_keys/no_offense.rb

{ foo: 1 }
{ bar: 2, baz: 3 }
{ :foo => 1 }
x = { key: 'value' }
y = {}
z = { 1 => 'a' }

# IO.popen with env hash (string keys for env vars are exempted)
IO.popen({"RUBYOPT" => '-w'}, 'ruby', 'foo.rb')
IO.popen({"FOO" => "bar"}, 'cmd') do |io|
  io.read
end

# Open3 methods with env hash
Open3.capture3({"RUBYOPT" => '-w'}, 'ruby', 'foo.rb')
Open3.capture2({"PATH" => '/usr/bin'}, 'ls')
Open3.popen3({"HOME" => '/tmp'}, 'bash')

# Open3.pipeline with env hash inside array
Open3.pipeline([{"RUBYOPT" => '-w'}, 'ruby', 'foo.rb'], ['wc', '-l'])

# spawn/system with env hash
spawn({"FOO" => "bar"}, "cmd")
system({"FOO" => "bar"}, "cmd")
Kernel.spawn({"FOO" => "bar"}, "cmd")
Kernel.system({"FOO" => "bar"}, "cmd")

# gsub/gsub! with string replacement hash
"hello".gsub(/pattern/, "old" => "new")
"hello".gsub!(/pattern/, "old" => "new")

# Heredoc used as hash key (Parser gem treats as dstr, not str)
produces(<<-EXAMPLE => 'defined(foo)')
  class bar { }
EXAMPLE

# Another heredoc key style
x = { <<~KEY => 'value' }
  multiline content
KEY

# String keys with invalid encoding (RuboCop skips via valid_encoding? check)
h = { "\x80" => 1 }
h = { "\xC0" => 2 }
h = { "\xFF" => 3 }
h = { "\251" => 4 }

# Multi-line string keys (Parser gem treats as dstr, not str; RuboCop skips)
h = { "first line
second line" => 1 }
h = { %(UPDATE
  dogs
SET
  name = 'fido') => false }
h = { 'multi
line
key' => :value }

Key Source Files

Rust implementation: src/cop/style/string_hash_keys.rs
RuboCop Ruby source (ground truth): vendor/rubocop/lib/rubocop/cop/style/string_hash_keys.rb
RuboCop test excerpts: vendor/rubocop/spec/rubocop/cop/style/string_hash_keys_spec.rb

Read these files before making changes.

Start Here

Use the existing corpus data to focus on the most concentrated regressions first.

Helpful local commands:

python3 scripts/investigate_cop.py Style/StringHashKeys --repos-only
python3 scripts/investigate_cop.py Style/StringHashKeys --context
python3 scripts/verify_cop_locations.py Style/StringHashKeys

Top FN repos:

archan937__motion-bundler__9cbfc90 (1 FN) — example test/unit/require/test_tracer.rb:81
simplecov-ruby__simplecov__522dc7d (1 FN) — example spec/simplecov_spec.rb:35

Representative FN examples:

archan937__motion-bundler__9cbfc90: test/unit/require/test_tracer.rb:81 — Prefer symbols instead of strings as hash keys.
simplecov-ruby__simplecov__522dc7d: spec/simplecov_spec.rb:35 — Prefer symbols instead of strings as hash keys.

Pre-diagnostic Results

Diagnosis Summary

Each example was tested by running nitrocop on the extracted source in isolation
with --force-default-config to determine if the issue is a code bug or config issue.
Note: source context is truncated and may not parse perfectly. If a diagnosis
seems wrong (e.g., your test passes immediately for a 'CODE BUG'), treat it as
a config/context issue instead.

FN: 2 code bug(s)

FN #1: `archan937motion-bundler9cbfc90: test/unit/require/test_tracer.rb:81`

NOT DETECTED — CODE BUG
The cop fails to detect this pattern. Fix the detection logic.

Full-file enclosing chain: block ({..}) (line 80: assert_equal({) > block (do..end) (line 71: it "should log dependencies as expected" do) > block (do..end) (line 70: describe "tracing require statements" do) > block (do..end) (line 7: describe MotionBundler::Require::Tracer do) > class body (line 5: class TestTracer < MiniTest::Unit::TestCase) > module body (line 4: module Require) > module body (line 3: module Unit)

Message: Prefer symbols instead of strings as hash keys.

Ready-made test snippet (add to offense.rb, adjust ^ count):

              __FILE__ => [
^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

Full file context (30 lines before offense):

       51:               MotionBundler::Require::Tracer.send :start
       52:               MotionBundler::Require::Tracer.log.expects(:clear)
       53:               MotionBundler::Require::Tracer.send :start
       54:             end
       55:           end
       56: 
       57:           it "should hook into require relatied core methods" do
       58:             MotionBundler::Require::Tracer.expects :hook
       59:             MotionBundler::Require::Tracer.send :start
       60:           end
       61:         end
       62: 
       63:         describe "calling `stop`" do
       64:           it "should unhook form require relatied core methods" do
       65:             MotionBundler::Require::Tracer.expects :unhook
       66:             MotionBundler::Require::Tracer.send :stop
       67:           end
       68:         end
       69: 
       70:         describe "tracing require statements" do
       71:           it "should log dependencies as expected" do
       72:             assert_equal({}, MotionBundler::Require::Tracer.log.instance_variable_get(:@files_dependencies))
       73: 
       74:             MotionBundler::Require::Tracer.yield do
       75:               require "d/a"
       76:               require "d/b"
       77:               require "d/c"
       78:             end
       79: 
       80:             assert_equal({
>>>    81:               __FILE__ => [
       82:                 lib_file("d/a.rb"),
       83:                 lib_file("d/b.rb"),
       84:                 lib_file("d/c.rb")
       85:               ],
       86:               lib_file("d/b.rb") => [
       87:                 lib_file("d/b/a.rb"),
       88:                 lib_file("d/b/b.rb")

FN #2: `simplecov-rubysimplecov522dc7d: spec/simplecov_spec.rb:35`

NOT DETECTED — CODE BUG
The cop fails to detect this pattern. Fix the detection logic.

Full-file enclosing chain: block (do..end) (line 34: it "uses the result from Coverage" do) > block (do..end) (line 29: context "when running" do) > block (do..end) (line 13: context "with merging disabled" do) > block (do..end) (line 7: describe ".result" do) > block (do..end) (line 6: describe SimpleCov do)

Message: Prefer symbols instead of strings as hash keys.

Ready-made test snippet (add to offense.rb, adjust ^ count):

          expect(Coverage).to receive(:result).once.and_return(__FILE__ => [0, 1])
^ Style/StringHashKeys: Prefer symbols instead of strings as hash keys.

Full file context (30 lines before offense):

        5: 
        6: describe SimpleCov do
        7:   describe ".result" do
        8:     before do
        9:       SimpleCov.clear_result
       10:       allow(Coverage).to receive(:result).once.and_return({})
       11:     end
       12: 
       13:     context "with merging disabled" do
       14:       before do
       15:         allow(SimpleCov).to receive(:use_merging).once.and_return(false)
       16:         expect(SimpleCov).not_to receive(:wait_for_other_processes)
       17:       end
       18: 
       19:       context "when not running" do
       20:         before do
       21:           allow(SimpleCov).to receive(:running).and_return(false)
       22:         end
       23: 
       24:         it "returns nil" do
       25:           expect(SimpleCov.result).to be_nil
       26:         end
       27:       end
       28: 
       29:       context "when running" do
       30:         before do
       31:           allow(SimpleCov).to receive(:running).and_return(true, false)
       32:         end
       33: 
       34:         it "uses the result from Coverage" do
>>>    35:           expect(Coverage).to receive(:result).once.and_return(__FILE__ => [0, 1])
       36:           expect(SimpleCov.result.filenames).to eq [__FILE__]
       37:         end
       38: 
       39:         it "adds not-loaded-files" do
       40:           expect(SimpleCov).to receive(:add_not_loaded_files).once.and_return({})
       41:           SimpleCov.result
       42:         end

github-actions · 2026-03-30T06:12:28Z

✅ Cop-check results (8 shards)

Cop	Baseline FN	FN Δ	Result
`Style/StringHashKeys`	0	0	✅
`Style/StringHashKeys`	0	0	✅
`Style/StringHashKeys`	2	-2	✅
`Style/StringHashKeys`	0	0	✅
`Style/StringHashKeys`	0	0	✅
`Style/StringHashKeys`	0	0	✅
`Style/StringHashKeys`	0	0	✅
`Style/StringHashKeys`	0	0	✅

All shards passed — ready to merge.

[bot] Fix Style/StringHashKeys: in progress

6a26216

6 bot added model:codex-hard type:cop-fix labels Mar 30, 2026

Fix Style/StringHashKeys: agent-generated fix (codex-hard)

0c01ce7

6 bot force-pushed the fix/style-string_hash_keys-23729853159 branch from a35f492 to 0c01ce7 Compare March 30, 2026 06:00

6 bot marked this pull request as ready for review March 30, 2026 06:00

6 bot enabled auto-merge (squash) March 30, 2026 06:00

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[bot] Fix Style/StringHashKeys#880

[bot] Fix Style/StringHashKeys#880
6[bot] wants to merge 2 commits intomainfrom
fix/style-string_hash_keys-23729853159

6 bot commented Mar 30, 2026 •

edited

Loading

Uh oh!

github-actions bot commented Mar 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

0 participants

Conversation

6 bot commented Mar 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Details

Result

Changes

Fix Style/StringHashKeys — 0 FP, 2 FN

Instructions

Workflow

Fixture Format

If your test passes immediately

CRITICAL: Avoid regressions in the opposite direction

Rules

Prism Notes

Current Fixture: offense.rb

Current Fixture: no_offense.rb

Key Source Files

Start Here

Pre-diagnostic Results

Diagnosis Summary

FN #1: archan937__motion-bundler__9cbfc90: test/unit/require/test_tracer.rb:81

FN #2: simplecov-ruby__simplecov__522dc7d: spec/simplecov_spec.rb:35

Uh oh!

github-actions bot commented Mar 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

0 participants

6 bot commented Mar 30, 2026 •

edited

Loading

FN #1: `archan937motion-bundler9cbfc90: test/unit/require/test_tracer.rb:81`

FN #2: `simplecov-rubysimplecov522dc7d: spec/simplecov_spec.rb:35`