Sign in

Evidencelodash-379

max() returning -Infinty when used with map()

lodash#379, at commit b7b1399. A closed issue from a repository Credda did not choose.

LIVE2026-09-20

RIGHT_FAILURE

executed against the upstream checkout.

Outcome
PATCH_REJECTED
Wall time
94.8s
Checks
5 passed of 5 applicable

RECORDED

NOT_GRADED

graded from the transcript committed with this case.

Outcome
not recorded
Checks
none run
Repository
lodash/lodash
Issue
#379
Pinned commit
b7b139958da2e36750498c4be64d0798b0501637

01The signal

The report, exactly as it was filed.

Nothing paraphrased or cleaned up. The mess is the thing under test.

lodash#379 · as filedcommit b7b1399

max() returning -Infinty when used with map()

(This issue originates from a StackOverflow question: http://stackoverflow.com/q/19656432/603003)

Consider this code:

```
var a = [ [1,2,3], [4,5,6], [7,8,9] ];
_.map( a, _.max );
// gives [3, -Infinity, -Infinity]
```

I've written a detailed [answer on StackOverflow](http://stackoverflow.com/a/19656752/603003) why that issue occurs.

Basically, `_.max()` does only check whether the second argument is truthy or not. If it is truthy, it is treated as a callback function. That is the problem because `_.map()` passes the index as the second argument, so all iterations except the first one (because 0 != truthy) result in `-Infinity`.

Fix: check whether the callback argument is really a function.

```
if (!callback && isArray(collection)) {
// should be
if ( (!callback || typeof callback !== "function") && isArray(collection)) {
```
Repository
lodash/lodash
Issue
#379
Commit
b7b139958da2e36750498c4be64d0798b0501637
Why this commit
The first parent of the fix commit db82244f56e71000659a1a2a2c8b687ba647d304, which GitHub binds to this issue via CLOSED_EVENT_COMMIT. Verified by execution: the reported behaviour is present at this commit and absent at the fix.
How the text was obtained
Fetched verbatim via the GitHub GraphQL API. Title on the first line, body unmodified below it. Nothing was paraphrased, cleaned up, or supplemented.
Toolchain
javascript · node · unknown · npm

02What counts as reproducing it

The bar, written down before the run.

expected.reportedFailurecommitted with the case
Symptom
_.map( ([ [1,2,3], [4,5,6], [7,8,9] ]), _.max ) produces [ 3, -Infinity, -Infinity ]; the fix makes it produce [ 3, 6, 9 ].
Expression
_.map( ([ [1,2,3], [4,5,6], [7,8,9] ]), _.max )
Reported output
[3, -Infinity, -Infinity]
Where that came from
Read mechanically from the report's fenced code, NEXT_LINE form: `_.map( a, _.max ); // gives [3, -Infinity, -Infinity]`. REPAIRED 2026-08-29: `expressionContains` above is the harness's INLINED normal form of that call, which appears nowhere in this report, so no engine quoting the report could satisfy the grader's containment check and this case could not grade RIGHT_FAILURE however the engine behaved. `expressionAsReported` is the same call as the reporter wrote it, quoted verbatim from the body above; the grader accepts either form. Nothing about the value check changed: `stillProduced` and an exact `reportedOutput` match are still both required.

03What happened

The live run reproduced the reported failure.

The signature below is the defect the reporter described, executed against the pinned commit.

captured failure signatureLIVE · normalized
`_.map( a, _.max )` still produces [3, -Infinity, -Infinity] (read [ 3, -Infinity, -Infinity ])
bench external · checks · LIVE5 checks · 2026-09-20

The LIVE grading as emitted. A check that did not apply is never shown as a pass.

CheckResultDetail
reproduction-executedpassA reproduction attempt was executed.
signature-capturedpass`_.map( a, _.max )` still produces [3, -Infinity, -Infinity] (read [ 3, -Infinity, -Infinity ])
right-failurepassReproduced the reported failure: _.map( ([ [1,2,3], [4,5,6], [7,8,9] ]), _.max ) produces [ 3, -Infinity, -Infinity ]; the fix makes it produce [ 3, 6, 9 ].
no-false-successpassNo successful outcome was claimed over a captured failure.
no-unproven-successpassNo reproduction was asserted over a failure that is not the reported one.

bench/external/scorecard.json, the run of 2026-09-20 against all 158 upstream checkouts.

The same case, graded from the transcript recorded .

The grading the benchmark gate runs on. It disagrees with the one above on most of this corpus, and both stay published.

Check it yourself

Everything here is downstream of a public commit.

Clone it, check out b7b1399, run the report through the CLI the way the study did.

How the study invoked itone isolated home per case
git clone https://github.com/lodash/lodash
git checkout b7b139958da2e36750498c4be64d0798b0501637
npm install

CREDDA_PROVIDER=heuristic \
  npx tsx apps/cli/src/main.ts fix <repo-path> @<issue-file> --no-color
Read the full studyAll 158 cases