NET-Web-API-w-Angular/my-app/node_modules/karma-coverage/docs/configuration.md

7.3 KiB
Executable file

Configuration

type

Type: String

Description: Specify a reporter type.

Possible Values:

  • html (default)
  • lcov (lcov and html)
  • lcovonly
  • text
  • text-summary
  • cobertura (xml format supported by Jenkins)
  • teamcity (code coverage System Messages for TeamCity)
  • json (json format supported by grunt-istanbul-coverage)
  • json-summary
  • in-memory (supported since v0.5.4)
  • none (Does nothing. Use to specify that no reporting is needed)

dir

Type: String

Description: This will be used to output coverage reports. When you set a relative path, the directory is resolved against the basePath.

subdir

Type: String

Description: This will be used in complement of the coverageReporter.dir option to generate the full output directory path. By default, the output directory is set to ./config.dir/BROWSER_NAME/, this option allows you to custom the second part. You can either pass a string or a function which will be called with the browser name passed as the only argument.

coverageReporter: {
  dir: 'coverage',
  subdir: '.'
  // Would output the results into: .'/coverage/'
}
coverageReporter: {
  dir: 'coverage',
  subdir: 'report'
  // Would output the results into: .'/coverage/report/'
}
coverageReporter: {
  dir: 'coverage',
  subdir: function(browser) {
    // normalization process to keep a consistent browser name across different
    // OS
    return browser.toLowerCase().split(/[ /-]/)[0];
  }
  // Would output the results into: './coverage/firefox/'
}

file

Type: String

Description: If you use one of these reporters, cobertura, lcovonly, teamcity, text or text-summary,you may set the file option to specify the output file.

coverageReporter: {
  type : 'text',
  dir : 'coverage/',
  file : 'coverage.txt'
}

check

Type: Object

Description: This will be used to configure minimum threshold enforcement for coverage results. If the thresholds are not met, karma will return failure. Thresholds, when specified as a positive number are taken to be the minimum percentage required. When a threshold is specified as a negative number it represents the maximum number of uncovered entities allowed.

For example, statements: 90 implies minimum statement coverage is 90%. statements: -10 implies that no more than 10 uncovered statements are allowed.

global applies to all files together and each on a per-file basis. A list of files or patterns can be excluded from enforcement via the excludes property. On a per-file or pattern basis, per-file thresholds can be overridden via the overrides property.

emitWarning allows to log exceeding coverage threshold into console as warning and not fail tests.

coverageReporter: {
  check: {
    emitWarning: false,
    global: {
      statements: 50,
      branches: 50,
      functions: 50,
      lines: 50,
      excludes: [
        'foo/bar/**/*.js'
      ]
    },
    each: {
      statements: 50,
      branches: 50,
      functions: 50,
      lines: 50,
      excludes: [
        'other/directory/**/*.js'
      ],
      overrides: {
        'baz/component/**/*.js': {
          statements: 98
        }
      }
    }
  }
}

watermarks

Type: Object

Description: This will be used to set the coverage threshold colors. The first number is the threshold between Red and Yellow. The second number is the threshold between Yellow and Green.

coverageReporter: {
  watermarks: {
    statements: [ 50, 75 ],
    functions: [ 50, 75 ],
    branches: [ 50, 75 ],
    lines: [ 50, 75 ]
  }
}

includeAllSources

Type: Boolean

Description: You can opt to include all sources files, as indicated by the coverage preprocessor, in your code coverage data, even if there are no tests covering them. (Default false)

coverageReporter: {
  type : 'text',
  dir : 'coverage/',
  file : 'coverage.txt',
  includeAllSources: true
}

sourceStore

Type: istanbul.Store

Description: You can opt to specify a source store allowing for external coverage collectors access to the instrumented code.

coverageReporter: {
  type : 'text',
  dir : 'coverage/',
  file : 'coverage.txt',
  sourceStore : require('istanbul').Store.create('fslookup')
}

reporters

Type: Array of Objects

Description: You can use multiple reporters, by providing array of options.

coverageReporter: {
  reporters:[
    {type: 'html', dir:'coverage/'},
    {type: 'teamcity'},
    {type: 'text-summary'}
  ],
}

instrumenter

Type: Object

Description: Karma-coverage can infers the instrumenter regarding of the file extension. It is possible to override this behavior and point out an instrumenter for the files matching a specific pattern. To do so, you need to declare an object under with the keys represents the pattern to match, and the instrumenter to apply. The matching will be done using minimatch. If two patterns match, the last one will take the precedence.

For example you can use Ibrik (an Istanbul analog for CoffeeScript files) with:

coverageReporter: {
  instrumenters: { ibrik : require('ibrik') },
  instrumenter: {
    '**/*.coffee': 'ibrik'
  },
  // ...
}

You can pass options additional options to specific instrumenter with:

var to5Options = { experimental: true };

// [...]

coverageReporter: {
  instrumenters: { isparta : require('isparta') },
  instrumenter: {
    '**/*.js': 'isparta'
  },
  instrumenterOptions: {
    isparta: { to5 : to5Options }
  }
}

useJSExtensionForCoffeeScript

Type: boolean

Description: If set to true, then CoffeeScript files instrumented with Ibrik will use the .js extension for the transpiled source (without this option, the JavaScript files will keep the original .coffee extension). This option is required if you use a module loader such as RequireJS that expects files to use a .js extension.

Example of using RequireJS with CoffeeScript:

coverageReporter:
  useJSExtensionForCoffeeScript: true
  instrumenters:
    ibrik : require('ibrik')
  instrumenter:
    '**/*.coffee': 'ibrik'
# ...

reporter[type='in-memory']

This is a different kind of reporter. Instead of writing a report physicaly to disk, it raises an event coverage_complete. This event can only be caught when using karma via the public api

var Server = require('karma').Server
var server = new Server({files: [/*...*/], port: 9876, coverageReporter: { type: 'in-memory' }, preprocessors: { '**/*.js': 'coverage' }, reporters: ['coverage'] }, function(exitCode) {
  console.log('Karma has exited with ' + exitCode)
  process.exit(exitCode)
})

server.on('coverage_complete', function (browser, coverageReport) {
  console.log('Covrage report: ', coverageReport)
})

server.start();

karma.runner.run({port: 9876});

The coverage report will be a merged result in json format.