Writes usage statistics for step definitions
Can render a plain text report showing which step definitions are used by which steps and some statistics for long each took. The error from the mean is the 95% confidence interval assuming a normal distribution.
Expression/Text Duration Mean ± Error Location
an order for {string} 0.009s 0.001s ± 0.000s samples/multiple-features/multiple-features.ts:3
an order for "eggs" 0.001s samples/multiple-features/multiple-features-1.feature:3
an order for "milk" 0.001s samples/multiple-features/multiple-features-1.feature:6
an order for "bread" 0.001s samples/multiple-features/multiple-features-1.feature:9
an order for "batteries" 0.001s samples/multiple-features/multiple-features-2.feature:3
an order for "light bulbs" 0.001s samples/multiple-features/multiple-features-2.feature:6
4 more
The output can also be rendered as a json report.
{
"stepDefinitions": [
{
"sourceReference": {
"uri": "samples/multiple-features/multiple-features.ts",
"location": {
"line": 3
}
},
"duration": {
"sum": {
"seconds": 0,
"nanos": 9000000
},
"mean": {
"seconds": 0,
"nanos": 1000000
},
"moe95": {
"seconds": 0,
"nanos": 0
}
},
"expression": {
"source": "an order for {string}",
"type": "CUCUMBER_EXPRESSION"
},
"matches": [
{
"text": "an order for \"eggs\"",
"duration": {
"seconds": 0,
"nanos": 1000000
},
"uri": "samples/multiple-features/multiple-features-1.feature",
"location": {
"line": 3,
"column": 3
}
},
...
Each language implementation validates itself against the examples in the
testdata
folder. See the testdata/README.md for more
information.