@grafana/data#MutableDataFrame TypeScript Examples

function getData() {
  const data = new MutableDataFrame({
    fields: [
      { name: 'Time', type: FieldType.time, values: [] },
      {
        name: 'Value',
        type: FieldType.number,
        values: [],
        config: {
          custom: {
            width: 100,
          },
        },
      },
      {
        name: 'Message',
        type: FieldType.string,
        values: [],
        config: {
          custom: {
            align: 'center',
          },
        },
      },
    ],
  });
  return data;
}

makeLog = (overides: Partial<LogRowModel>): LogRowModel => {
  const uid = overides.uid || '1';
  const entry = `log message ${uid}`;
  return {
    uid,
    entryFieldIndex: 0,
    rowIndex: 0,
    dataFrame: new MutableDataFrame(),
    logLevel: LogLevel.debug,
    entry,
    hasAnsi: false,
    labels: {},
    raw: entry,
    timeFromNow: '',
    timeEpochMs: 1,
    timeLocal: '',
    timeUtc: '',
    ...overides,
  };
}

export function appendResponseToBufferedData(response: LokiTailResponse, data: MutableDataFrame) {
  // Should we do anything with: response.dropped_entries?

  const streams: LokiStreamResult[] = response.streams;
  if (!streams || !streams.length) {
    return;
  }

  let baseLabels: Labels = {};
  for (const f of data.fields) {
    if (f.type === FieldType.string) {
      if (f.labels) {
        baseLabels = f.labels;
      }
      break;
    }
  }

  for (const stream of streams) {
    // Find unique labels
    const unique = findUniqueLabels(stream.stream, baseLabels);
    const allLabelsString = Object.entries(stream.stream)
      .map(([key, val]) => `${key}="${val}"`)
      .sort()
      .join('');

    // Add each line
    for (const [ts, line] of stream.values) {
      data.values.ts.add(new Date(parseInt(ts.substr(0, ts.length - 6), 10)).toISOString());
      data.values.tsNs.add(ts);
      data.values.line.add(line);
      data.values.labels.add(unique);
      data.values.id.add(createUid(ts, allLabelsString, line));
    }
  }
}

/**
 * Constructs dataFrame with supplied fields and other data. Also makes sure it is properly reversed if needed.
 */
function constructDataFrame(
  times: ArrayVector<string>,
  timesNs: ArrayVector<string>,
  lines: ArrayVector<string>,
  uids: ArrayVector<string>,
  labels: Labels,
  reverse?: boolean,
  refId?: string
) {
  const dataFrame = {
    refId,
    fields: [
      { name: 'ts', type: FieldType.time, config: { title: 'Time' }, values: times }, // Time
      { name: 'line', type: FieldType.string, config: {}, values: lines, labels }, // Line
      { name: 'id', type: FieldType.string, config: {}, values: uids },
      { name: 'tsNs', type: FieldType.time, config: { title: 'Time ns' }, values: timesNs }, // Time
    ],
    length: times.length,
  };

  if (reverse) {
    const mutableDataFrame = new MutableDataFrame(dataFrame);
    mutableDataFrame.reverse();
    return mutableDataFrame;
  }

  return dataFrame;
}

onSeriesParsed = (data: DataFrame[], text: string) => {
    const { query, onChange, onRunQuery } = this.props;
    this.setState({ text });
    if (!data) {
      data = [new MutableDataFrame()];
    }
    onChange({ ...query, data });
    onRunQuery();
  };

onSourceChange = (item: SelectableValue<string>) => {
    const { datasource, query, onChange, onRunQuery } = this.props;
    let data: DataFrame[] | undefined = undefined;
    if (item.value === 'panel') {
      if (query.data) {
        return;
      }
      data = [...datasource.data];
      if (!data) {
        data = [new MutableDataFrame()];
      }
      this.setState({ text: toCSV(data) });
    }
    onChange({ ...query, data });
    onRunQuery();
  };

onSeriesParsed = (data: DataFrame[], text: string) => {
    const { options, onOptionsChange } = this.props;
    if (!data) {
      data = [new MutableDataFrame()];
    }
    // data is a property on 'jsonData'
    const jsonData = {
      ...options.jsonData,
      data,
    };

    onOptionsChange({
      ...options,
      jsonData,
    });
    this.setState({ text });
  };

makeLog = (overrides: Partial<LogRowModel>): LogRowModel => {
  const uid = overrides.uid || '1';
  const entry = `log message ${uid}`;
  return {
    entryFieldIndex: 0,
    rowIndex: 0,
    // Does not need to be filled with current tests
    dataFrame: new MutableDataFrame(),
    uid,
    logLevel: LogLevel.debug,
    entry,
    hasAnsi: false,
    labels: {},
    raw: entry,
    timeFromNow: '',
    timeEpochMs: 1,
    timeLocal: '',
    timeUtc: '',
    searchWords: [],
    ...overrides,
  };
}

row: LogRowModel = {
  entryFieldIndex: 0,
  rowIndex: 0,
  dataFrame: new MutableDataFrame(),
  entry: '4',
  labels: (null as any) as Labels,
  hasAnsi: false,
  raw: '4',
  logLevel: LogLevel.info,
  timeEpochMs: 4,
  timeFromNow: '',
  timeLocal: '',
  timeUtc: '',
  uid: '1',
}

setup = (propOverrides?: Partial<Props>, rowOverrides?: Partial<LogRowModel>) => {
  const props: Props = {
    theme: {} as GrafanaTheme,
    showDuplicates: false,
    row: {
      dataFrame: new MutableDataFrame(),
      entryFieldIndex: 0,
      rowIndex: 0,
      logLevel: 'error' as LogLevel,
      timeFromNow: '',
      timeEpochMs: 1546297200000,
      timeLocal: '',
      timeUtc: '',
      hasAnsi: false,
      entry: '',
      raw: '',
      uid: '0',
      labels: {},
      ...(rowOverrides || {}),
    },
    getRows: () => [],
    onClickFilterLabel: () => {},
    onClickFilterOutLabel: () => {},
    ...(propOverrides || {}),
  };

  return mount(<LogDetails {...props} />);
}

function buildData(theme: GrafanaTheme, overrides: ConfigOverrideRule[]): DataFrame {
  const data = new MutableDataFrame({
    fields: [
      { name: 'Time', type: FieldType.time, values: [] }, // The time field
      {
        name: 'Quantity',
        type: FieldType.number,
        values: [],
        config: {
          decimals: 0,
          custom: {
            align: 'center',
          },
        },
      },
      { name: 'Status', type: FieldType.string, values: [] }, // The time field
      {
        name: 'Value',
        type: FieldType.number,
        values: [],
        config: {
          decimals: 2,
        },
      },
      {
        name: 'Progress',
        type: FieldType.number,
        values: [],
        config: {
          unit: 'percent',
          custom: {
            width: 100,
          },
        },
      },
    ],
  });

  for (let i = 0; i < 1000; i++) {
    data.appendRow([
      new Date().getTime(),
      Math.random() * 2,
      Math.random() > 0.7 ? 'Active' : 'Cancelled',
      Math.random() * 100,
      Math.random() * 100,
    ]);
  }

  return applyFieldOverrides({
    data: [data],
    fieldOptions: {
      overrides,
      defaults: {},
    },
    theme,
    replaceVariables: (value: string) => value,
  })[0];
}

// Build the DataFrames to return to Grafana
  buildDataFrames(series) {
    let data: MutableDataFrame[] = [];
    let error = {} as DataQueryError;

    series.data.forEach((target) => {
      if (target.meta.errorMessage) {
        error.message = target.meta.errorMessage;
        throw new Error(target.meta.errorMessage);
      }
      if (target.columns) {
        var fields = [];
        target.columns.forEach((column) => {
          fields.push({ name: column.text });
        });
        const frame = new MutableDataFrame({
          refId: target.refId,
          fields: fields,
        });
        target.rows.forEach((element) => {
          var row = [];
          element.forEach((entry) => {
            row.push(entry);
          });
          frame.appendRow(row);
        });
        data.push(frame);
      } else {
        // time series

        let datapoints = target.datapoints;
        const timeValues = datapoints.map((datapoint) => datapoint[1]);
        const values = datapoints.map((datapoint) => datapoint[0]);

        const fields = [
          { name: 'Time', values: timeValues, type: FieldType.time },
          {
            name: target.target,
            values: values,
          },
        ];

        data.push(
          new MutableDataFrame({
            refId: target.refId,
            fields: fields,
          })
        );
      }
    });
    return { data, state: 'done' };
  }

createEmptyDataFrame = (
  propNames: string[],
  timeField: string,
  logMessageField?: string,
  logLevelField?: string
): MutableDataFrame => {
  const series = new MutableDataFrame({ fields: [] });

  series.addField({
    name: timeField,
    type: FieldType.time,
  });

  if (logMessageField) {
    series.addField({
      name: logMessageField,
      type: FieldType.string,
    }).parse = (v: any) => {
      return v || '';
    };
  } else {
    series.addField({
      name: '_source',
      type: FieldType.string,
    }).parse = (v: any) => {
      return JSON.stringify(v, null, 2);
    };
  }

  if (logLevelField) {
    series.addField({
      name: 'level',
      type: FieldType.string,
    }).parse = (v: any) => {
      return v || '';
    };
  }

  const fieldNames = series.fields.map(field => field.name);

  for (const propName of propNames) {
    // Do not duplicate fields. This can mean that we will shadow some fields.
    if (fieldNames.includes(propName)) {
      continue;
    }

    series.addField({
      name: propName,
      type: FieldType.string,
    }).parse = (v: any) => {
      return v || '';
    };
  }

  return series;
}

describe('InputDatasource', () => {
  const data = readCSV('a,b,c\n1,2,3\n4,5,6');
  const instanceSettings: DataSourceInstanceSettings<InputOptions> = {
    id: 1,
    type: 'x',
    name: 'xxx',
    meta: {} as PluginMeta,
    jsonData: {
      data,
    },
  };

  describe('when querying', () => {
    test('should return the saved data with a query', () => {
      const ds = new InputDatasource(instanceSettings);
      const options = getQueryOptions<InputQuery>({
        targets: [{ refId: 'Z' }],
      });

      return ds.query(options).then(rsp => {
        expect(rsp.data.length).toBe(1);

        const series: DataFrame = rsp.data[0];
        expect(series.refId).toBe('Z');
        expect(series.fields[0].values).toEqual(data[0].fields[0].values);
      });
    });
  });

  test('DataFrame descriptions', () => {
    expect(describeDataFrame([])).toEqual('');
    expect(describeDataFrame(null)).toEqual('');
    expect(
      describeDataFrame([
        new MutableDataFrame({
          name: 'x',
          fields: [{ name: 'a' }],
        }),
      ])
    ).toEqual('1 Fields, 0 Rows');
  });
});

/**
 * Transform LokiResponse data and appends it to MutableDataFrame. Used for streaming where the dataFrame can be
 * a CircularDataFrame creating a fixed size rolling buffer.
 * TODO: Probably could be unified with the logStreamToDataFrame function.
 * @param response
 * @param data Needs to have ts, line, labels, id as fields
 */
export function appendLegacyResponseToBufferedData(response: LokiLegacyStreamResponse, data: MutableDataFrame) {
  // Should we do anything with: response.dropped_entries?

  const streams: LokiLegacyStreamResult[] = response.streams;
  if (!streams || !streams.length) {
    return;
  }

  let baseLabels: Labels = {};
  for (const f of data.fields) {
    if (f.type === FieldType.string) {
      if (f.labels) {
        baseLabels = f.labels;
      }
      break;
    }
  }

  for (const stream of streams) {
    // Find unique labels
    const labels = parseLabels(stream.labels);
    const unique = findUniqueLabels(labels, baseLabels);

    // Add each line
    for (const entry of stream.entries) {
      const ts = entry.ts || entry.timestamp;
      data.values.ts.add(ts);
      data.values.line.add(entry.line);
      data.values.labels.add(unique);
      data.values.id.add(createUid(ts, stream.labels, entry.line));
    }
  }
}

describe('dataFrameToLogsModel', () => {
  it('given empty series should return empty logs model', () => {
    expect(dataFrameToLogsModel([] as DataFrame[], 0, 'utc')).toMatchObject(emptyLogsModel);
  });

  it('given series without correct series name should return empty logs model', () => {
    const series: DataFrame[] = [
      toDataFrame({
        fields: [],
      }),
    ];
    expect(dataFrameToLogsModel(series, 0, 'utc')).toMatchObject(emptyLogsModel);
  });

  it('given series without a time field should return empty logs model', () => {
    const series: DataFrame[] = [
      new MutableDataFrame({
        fields: [
          {
            name: 'message',
            type: FieldType.string,
            values: [],
          },
        ],
      }),
    ];
    expect(dataFrameToLogsModel(series, 0, 'utc')).toMatchObject(emptyLogsModel);
  });

  it('given series without a string field should return empty logs model', () => {
    const series: DataFrame[] = [
      new MutableDataFrame({
        fields: [
          {
            name: 'time',
            type: FieldType.time,
            values: [],
          },
        ],
      }),
    ];
    expect(dataFrameToLogsModel(series, 0, 'utc')).toMatchObject(emptyLogsModel);
  });

  it('given one series should return expected logs model', () => {
    const series: DataFrame[] = [
      new MutableDataFrame({
        fields: [
          {
            name: 'time',
            type: FieldType.time,
            values: ['2019-04-26T09:28:11.352440161Z', '2019-04-26T14:42:50.991981292Z'],
          },
          {
            name: 'message',
            type: FieldType.string,
            values: [
              't=2019-04-26T11:05:28+0200 lvl=info msg="Initializing DatasourceCacheService" logger=server',
              't=2019-04-26T16:42:50+0200 lvl=eror msg="new token…t unhashed token=56d9fdc5c8b7400bd51b060eea8ca9d7',
            ],
            labels: {
              filename: '/var/log/grafana/grafana.log',
              job: 'grafana',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['foo', 'bar'],
          },
        ],
        meta: {
          limit: 1000,
        },
      }),
    ];
    const logsModel = dataFrameToLogsModel(series, 0, 'utc');
    expect(logsModel.hasUniqueLabels).toBeFalsy();
    expect(logsModel.rows).toHaveLength(2);
    expect(logsModel.rows).toMatchObject([
      {
        entry: 't=2019-04-26T11:05:28+0200 lvl=info msg="Initializing DatasourceCacheService" logger=server',
        labels: { filename: '/var/log/grafana/grafana.log', job: 'grafana' },
        logLevel: 'info',
        uniqueLabels: {},
        uid: 'foo',
      },
      {
        entry: 't=2019-04-26T16:42:50+0200 lvl=eror msg="new token…t unhashed token=56d9fdc5c8b7400bd51b060eea8ca9d7',
        labels: { filename: '/var/log/grafana/grafana.log', job: 'grafana' },
        logLevel: 'error',
        uniqueLabels: {},
        uid: 'bar',
      },
    ]);

    expect(logsModel.series).toHaveLength(2);
    expect(logsModel.meta).toHaveLength(2);
    expect(logsModel.meta[0]).toMatchObject({
      label: 'Common labels',
      value: series[0].fields[1].labels,
      kind: LogsMetaKind.LabelsMap,
    });
    expect(logsModel.meta[1]).toMatchObject({
      label: 'Limit',
      value: `1000 (2 returned)`,
      kind: LogsMetaKind.String,
    });
  });

  it('given one series without labels should return expected logs model', () => {
    const series: DataFrame[] = [
      new MutableDataFrame({
        fields: [
          {
            name: 'time',
            type: FieldType.time,
            values: ['1970-01-01T00:00:01Z'],
          },
          {
            name: 'message',
            type: FieldType.string,
            values: ['WARN boooo'],
          },
          {
            name: 'level',
            type: FieldType.string,
            values: ['dbug'],
          },
        ],
      }),
    ];
    const logsModel = dataFrameToLogsModel(series, 0, 'utc');
    expect(logsModel.rows).toHaveLength(1);
    expect(logsModel.rows).toMatchObject([
      {
        entry: 'WARN boooo',
        labels: undefined,
        logLevel: LogLevel.debug,
        uniqueLabels: {},
      },
    ]);
  });

  it('given multiple series with unique times should return expected logs model', () => {
    const series: DataFrame[] = [
      toDataFrame({
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:01Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo'],
            labels: {
              foo: 'bar',
              baz: '1',
              level: 'dbug',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['0'],
          },
        ],
      }),
      toDataFrame({
        name: 'logs',
        fields: [
          {
            name: 'time',
            type: FieldType.time,
            values: ['1970-01-01T00:00:00Z', '1970-01-01T00:00:02Z'],
          },
          {
            name: 'message',
            type: FieldType.string,
            values: ['INFO 1', 'INFO 2'],
            labels: {
              foo: 'bar',
              baz: '2',
              level: 'err',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['1', '2'],
          },
        ],
      }),
    ];
    const logsModel = dataFrameToLogsModel(series, 0, 'utc');
    expect(logsModel.hasUniqueLabels).toBeTruthy();
    expect(logsModel.rows).toHaveLength(3);
    expect(logsModel.rows).toMatchObject([
      {
        entry: 'INFO 1',
        labels: { foo: 'bar', baz: '2' },
        logLevel: LogLevel.error,
        uniqueLabels: { baz: '2' },
      },
      {
        entry: 'WARN boooo',
        labels: { foo: 'bar', baz: '1' },
        logLevel: LogLevel.debug,
        uniqueLabels: { baz: '1' },
      },
      {
        entry: 'INFO 2',
        labels: { foo: 'bar', baz: '2' },
        logLevel: LogLevel.error,
        uniqueLabels: { baz: '2' },
      },
    ]);

    expect(logsModel.series).toHaveLength(2);
    expect(logsModel.meta).toHaveLength(1);
    expect(logsModel.meta[0]).toMatchObject({
      label: 'Common labels',
      value: {
        foo: 'bar',
      },
      kind: LogsMetaKind.LabelsMap,
    });
  });
  it('given multiple series with equal times should return expected logs model', () => {
    const series: DataFrame[] = [
      toDataFrame({
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:00Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo 1'],
            labels: {
              foo: 'bar',
              baz: '1',
              level: 'dbug',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['0'],
          },
        ],
      }),
      toDataFrame({
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:01Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo 2'],
            labels: {
              foo: 'bar',
              baz: '2',
              level: 'dbug',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['1'],
          },
        ],
      }),
      toDataFrame({
        name: 'logs',
        fields: [
          {
            name: 'time',
            type: FieldType.time,
            values: ['1970-01-01T00:00:00Z', '1970-01-01T00:00:01Z'],
          },
          {
            name: 'message',
            type: FieldType.string,
            values: ['INFO 1', 'INFO 2'],
            labels: {
              foo: 'bar',
              baz: '2',
              level: 'err',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['2', '3'],
          },
        ],
      }),
    ];
    const logsModel = dataFrameToLogsModel(series, 0, 'utc');
    expect(logsModel.hasUniqueLabels).toBeTruthy();
    expect(logsModel.rows).toHaveLength(4);
    expect(logsModel.rows).toMatchObject([
      {
        entry: 'WARN boooo 1',
        labels: { foo: 'bar', baz: '1' },
        logLevel: LogLevel.debug,
        uniqueLabels: { baz: '1' },
      },
      {
        entry: 'INFO 1',
        labels: { foo: 'bar', baz: '2' },
        logLevel: LogLevel.error,
        uniqueLabels: { baz: '2' },
      },
      {
        entry: 'WARN boooo 2',
        labels: { foo: 'bar', baz: '2' },
        logLevel: LogLevel.debug,
        uniqueLabels: { baz: '2' },
      },
      {
        entry: 'INFO 2',
        labels: { foo: 'bar', baz: '2' },
        logLevel: LogLevel.error,
        uniqueLabels: { baz: '2' },
      },
    ]);
  });

  it('should fallback to row index if no id', () => {
    const series: DataFrame[] = [
      toDataFrame({
        labels: { foo: 'bar' },
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:00Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo 1'],
          },
        ],
      }),
    ];
    const logsModel = dataFrameToLogsModel(series, 0, 'utc');
    expect(logsModel.rows[0].uid).toBe('0');
  });

  it('given multiple series with equal ids should return expected logs model', () => {
    const series: DataFrame[] = [
      toDataFrame({
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:00Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo 1'],
            labels: {
              foo: 'bar',
              baz: '1',
              level: 'dbug',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['0'],
          },
        ],
      }),
      toDataFrame({
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:01Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo 2'],
            labels: {
              foo: 'bar',
              baz: '2',
              level: 'dbug',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['1'],
          },
        ],
      }),
      toDataFrame({
        fields: [
          {
            name: 'ts',
            type: FieldType.time,
            values: ['1970-01-01T00:00:01Z'],
          },
          {
            name: 'line',
            type: FieldType.string,
            values: ['WARN boooo 2'],
            labels: {
              foo: 'bar',
              baz: '2',
              level: 'dbug',
            },
          },
          {
            name: 'id',
            type: FieldType.string,
            values: ['1'],
          },
        ],
      }),
    ];
    const logsModel = dataFrameToLogsModel(series, 0, 'utc');
    expect(logsModel.hasUniqueLabels).toBeTruthy();
    expect(logsModel.rows).toHaveLength(2);
    expect(logsModel.rows).toMatchObject([
      {
        entry: 'WARN boooo 1',
        labels: { foo: 'bar' },
        logLevel: LogLevel.debug,
        uniqueLabels: { baz: '1' },
      },
      {
        entry: 'WARN boooo 2',
        labels: { foo: 'bar' },
        logLevel: LogLevel.debug,
        uniqueLabels: { baz: '2' },
      },
    ]);
  });
});

describe('sortLogsResult', () => {
  const firstRow: LogRowModel = {
    rowIndex: 0,
    entryFieldIndex: 0,
    dataFrame: new MutableDataFrame(),
    entry: '',
    hasAnsi: false,
    labels: {},
    logLevel: LogLevel.info,
    raw: '',
    timeEpochMs: 0,
    timeFromNow: '',
    timeLocal: '',
    timeUtc: '',
    uid: '1',
  };
  const sameAsFirstRow = firstRow;
  const secondRow: LogRowModel = {
    rowIndex: 1,
    entryFieldIndex: 0,
    dataFrame: new MutableDataFrame(),
    entry: '',
    hasAnsi: false,
    labels: {},
    logLevel: LogLevel.info,
    raw: '',
    timeEpochMs: 10,
    timeFromNow: '',
    timeLocal: '',
    timeUtc: '',
    uid: '2',
  };

  describe('when called with SortOrder.Descending', () => {
    it('then it should sort descending', () => {
      const logsResult: LogsModel = {
        rows: [firstRow, sameAsFirstRow, secondRow],
        hasUniqueLabels: false,
      };
      const result = sortLogsResult(logsResult, SortOrder.Descending);

      expect(result).toEqual({
        rows: [secondRow, firstRow, sameAsFirstRow],
        hasUniqueLabels: false,
      });
    });
  });

  describe('when called with SortOrder.Ascending', () => {
    it('then it should sort ascending', () => {
      const logsResult: LogsModel = {
        rows: [secondRow, firstRow, sameAsFirstRow],
        hasUniqueLabels: false,
      };
      const result = sortLogsResult(logsResult, SortOrder.Ascending);

      expect(result).toEqual({
        rows: [firstRow, sameAsFirstRow, secondRow],
        hasUniqueLabels: false,
      });
    });
  });

  describe('when buildQueryTransaction', () => {
    it('it should calculate interval based on time range', () => {
      const queries = [{ refId: 'A' }];
      const queryOptions = { maxDataPoints: 1000, minInterval: '15s' };
      const range = { from: dateTime().subtract(1, 'd'), to: dateTime(), raw: { from: '1h', to: '1h' } };
      const transaction = buildQueryTransaction(queries, queryOptions, range, false);

      expect(transaction.request.intervalMs).toEqual(60000);
    });

    it('it should calculate interval taking minInterval into account', () => {
      const queries = [{ refId: 'A' }];
      const queryOptions = { maxDataPoints: 1000, minInterval: '15s' };
      const range = { from: dateTime().subtract(1, 'm'), to: dateTime(), raw: { from: '1h', to: '1h' } };
      const transaction = buildQueryTransaction(queries, queryOptions, range, false);

      expect(transaction.request.intervalMs).toEqual(15000);
    });

    it('it should calculate interval taking maxDataPoints into account', () => {
      const queries = [{ refId: 'A' }];
      const queryOptions = { maxDataPoints: 10, minInterval: '15s' };
      const range = { from: dateTime().subtract(1, 'd'), to: dateTime(), raw: { from: '1h', to: '1h' } };
      const transaction = buildQueryTransaction(queries, queryOptions, range, false);

      expect(transaction.request.interval).toEqual('2h');
    });
  });
});

describe('ElasticResponse', () => {
  let targets;
  let response: any;
  let result: any;

  describe('simple query and count', () => {
    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'count', id: '1' }],
          bucketAggs: [{ type: 'date_histogram', field: '@timestamp', id: '2' }],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    doc_count: 10,
                    key: 1000,
                  },
                  {
                    doc_count: 15,
                    key: 2000,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 1 series', () => {
      expect(result.data.length).toBe(1);
      expect(result.data[0].target).toBe('Count');
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].datapoints[0][0]).toBe(10);
      expect(result.data[0].datapoints[0][1]).toBe(1000);
    });
  });

  describe('simple query count & avg aggregation', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [
            { type: 'count', id: '1' },
            { type: 'avg', field: 'value', id: '2' },
          ],
          bucketAggs: [{ type: 'date_histogram', field: '@timestamp', id: '3' }],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '3': {
                buckets: [
                  {
                    '2': { value: 88 },
                    doc_count: 10,
                    key: 1000,
                  },
                  {
                    '2': { value: 99 },
                    doc_count: 15,
                    key: 2000,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 2 series', () => {
      expect(result.data.length).toBe(2);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].datapoints[0][0]).toBe(10);
      expect(result.data[0].datapoints[0][1]).toBe(1000);

      expect(result.data[1].target).toBe('Average value');
      expect(result.data[1].datapoints[0][0]).toBe(88);
      expect(result.data[1].datapoints[1][0]).toBe(99);
    });
  });

  describe('single group by query one metric', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'count', id: '1' }],
          bucketAggs: [
            { type: 'terms', field: 'host', id: '2' },
            { type: 'date_histogram', field: '@timestamp', id: '3' },
          ],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    '3': {
                      buckets: [
                        { doc_count: 1, key: 1000 },
                        { doc_count: 3, key: 2000 },
                      ],
                    },
                    doc_count: 4,
                    key: 'server1',
                  },
                  {
                    '3': {
                      buckets: [
                        { doc_count: 2, key: 1000 },
                        { doc_count: 8, key: 2000 },
                      ],
                    },
                    doc_count: 10,
                    key: 'server2',
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 2 series', () => {
      expect(result.data.length).toBe(2);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].target).toBe('server1');
      expect(result.data[1].target).toBe('server2');
    });
  });

  describe('single group by query two metrics', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [
            { type: 'count', id: '1' },
            { type: 'avg', field: '@value', id: '4' },
          ],
          bucketAggs: [
            { type: 'terms', field: 'host', id: '2' },
            { type: 'date_histogram', field: '@timestamp', id: '3' },
          ],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    '3': {
                      buckets: [
                        { '4': { value: 10 }, doc_count: 1, key: 1000 },
                        { '4': { value: 12 }, doc_count: 3, key: 2000 },
                      ],
                    },
                    doc_count: 4,
                    key: 'server1',
                  },
                  {
                    '3': {
                      buckets: [
                        { '4': { value: 20 }, doc_count: 1, key: 1000 },
                        { '4': { value: 32 }, doc_count: 3, key: 2000 },
                      ],
                    },
                    doc_count: 10,
                    key: 'server2',
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 2 series', () => {
      expect(result.data.length).toBe(4);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].target).toBe('server1 Count');
      expect(result.data[1].target).toBe('server1 Average @value');
      expect(result.data[2].target).toBe('server2 Count');
      expect(result.data[3].target).toBe('server2 Average @value');
    });
  });

  describe('with percentiles ', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'percentiles', settings: { percents: [75, 90] }, id: '1' }],
          bucketAggs: [{ type: 'date_histogram', field: '@timestamp', id: '3' }],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '3': {
                buckets: [
                  {
                    '1': { values: { '75': 3.3, '90': 5.5 } },
                    doc_count: 10,
                    key: 1000,
                  },
                  {
                    '1': { values: { '75': 2.3, '90': 4.5 } },
                    doc_count: 15,
                    key: 2000,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 2 series', () => {
      expect(result.data.length).toBe(2);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].target).toBe('p75');
      expect(result.data[1].target).toBe('p90');
      expect(result.data[0].datapoints[0][0]).toBe(3.3);
      expect(result.data[0].datapoints[0][1]).toBe(1000);
      expect(result.data[1].datapoints[1][0]).toBe(4.5);
    });
  });

  describe('with extended_stats', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [
            {
              type: 'extended_stats',
              meta: { max: true, std_deviation_bounds_upper: true },
              id: '1',
            },
          ],
          bucketAggs: [
            { type: 'terms', field: 'host', id: '3' },
            { type: 'date_histogram', id: '4' },
          ],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '3': {
                buckets: [
                  {
                    key: 'server1',
                    '4': {
                      buckets: [
                        {
                          '1': {
                            max: 10.2,
                            min: 5.5,
                            std_deviation_bounds: { upper: 3, lower: -2 },
                          },
                          doc_count: 10,
                          key: 1000,
                        },
                      ],
                    },
                  },
                  {
                    key: 'server2',
                    '4': {
                      buckets: [
                        {
                          '1': {
                            max: 10.2,
                            min: 5.5,
                            std_deviation_bounds: { upper: 3, lower: -2 },
                          },
                          doc_count: 10,
                          key: 1000,
                        },
                      ],
                    },
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 4 series', () => {
      expect(result.data.length).toBe(4);
      expect(result.data[0].datapoints.length).toBe(1);
      expect(result.data[0].target).toBe('server1 Max');
      expect(result.data[1].target).toBe('server1 Std Dev Upper');

      expect(result.data[0].datapoints[0][0]).toBe(10.2);
      expect(result.data[1].datapoints[0][0]).toBe(3);
    });
  });

  describe('single group by with alias pattern', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'count', id: '1' }],
          alias: '{{term @host}} {{metric}} and {{not_exist}} {{@host}}',
          bucketAggs: [
            { type: 'terms', field: '@host', id: '2' },
            { type: 'date_histogram', field: '@timestamp', id: '3' },
          ],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    '3': {
                      buckets: [
                        { doc_count: 1, key: 1000 },
                        { doc_count: 3, key: 2000 },
                      ],
                    },
                    doc_count: 4,
                    key: 'server1',
                  },
                  {
                    '3': {
                      buckets: [
                        { doc_count: 2, key: 1000 },
                        { doc_count: 8, key: 2000 },
                      ],
                    },
                    doc_count: 10,
                    key: 'server2',
                  },
                  {
                    '3': {
                      buckets: [
                        { doc_count: 2, key: 1000 },
                        { doc_count: 8, key: 2000 },
                      ],
                    },
                    doc_count: 10,
                    key: 0,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 2 series', () => {
      expect(result.data.length).toBe(3);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].target).toBe('server1 Count and {{not_exist}} server1');
      expect(result.data[1].target).toBe('server2 Count and {{not_exist}} server2');
      expect(result.data[2].target).toBe('0 Count and {{not_exist}} 0');
    });
  });

  describe('histogram response', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'count', id: '1' }],
          bucketAggs: [{ type: 'histogram', field: 'bytes', id: '3' }],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '3': {
                buckets: [
                  { doc_count: 1, key: 1000 },
                  { doc_count: 3, key: 2000 },
                  { doc_count: 2, key: 1000 },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return table with byte and count', () => {
      expect(result.data[0].rows.length).toBe(3);
      expect(result.data[0].columns).toEqual([{ text: 'bytes', filterable: true }, { text: 'Count' }]);
    });
  });

  describe('with two filters agg', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'count', id: '1' }],
          bucketAggs: [
            {
              id: '2',
              type: 'filters',
              settings: {
                filters: [{ query: '@metric:cpu' }, { query: '@metric:logins.count' }],
              },
            },
            { type: 'date_histogram', field: '@timestamp', id: '3' },
          ],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: {
                  '@metric:cpu': {
                    '3': {
                      buckets: [
                        { doc_count: 1, key: 1000 },
                        { doc_count: 3, key: 2000 },
                      ],
                    },
                  },
                  '@metric:logins.count': {
                    '3': {
                      buckets: [
                        { doc_count: 2, key: 1000 },
                        { doc_count: 8, key: 2000 },
                      ],
                    },
                  },
                },
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 2 series', () => {
      expect(result.data.length).toBe(2);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].target).toBe('@metric:cpu');
      expect(result.data[1].target).toBe('@metric:logins.count');
    });
  });

  describe('with dropfirst and last aggregation', () => {
    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'avg', id: '1' }, { type: 'count' }],
          bucketAggs: [
            {
              id: '2',
              type: 'date_histogram',
              field: 'host',
              settings: { trimEdges: 1 },
            },
          ],
        },
      ];

      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    '1': { value: 1000 },
                    key: 1,
                    doc_count: 369,
                  },
                  {
                    '1': { value: 2000 },
                    key: 2,
                    doc_count: 200,
                  },
                  {
                    '1': { value: 2000 },
                    key: 3,
                    doc_count: 200,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should remove first and last value', () => {
      expect(result.data.length).toBe(2);
      expect(result.data[0].datapoints.length).toBe(1);
    });
  });

  describe('No group by time', () => {
    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'avg', id: '1' }, { type: 'count' }],
          bucketAggs: [{ id: '2', type: 'terms', field: 'host' }],
        },
      ];

      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    '1': { value: 1000 },
                    key: 'server-1',
                    doc_count: 369,
                  },
                  {
                    '1': { value: 2000 },
                    key: 'server-2',
                    doc_count: 200,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return table', () => {
      expect(result.data.length).toBe(1);
      expect(result.data[0].type).toBe('table');
      expect(result.data[0].rows.length).toBe(2);
      expect(result.data[0].rows[0][0]).toBe('server-1');
      expect(result.data[0].rows[0][1]).toBe(1000);
      expect(result.data[0].rows[0][2]).toBe(369);

      expect(result.data[0].rows[1][0]).toBe('server-2');
      expect(result.data[0].rows[1][1]).toBe(2000);
    });
  });

  describe('No group by time with percentiles ', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'percentiles', field: 'value', settings: { percents: [75, 90] }, id: '1' }],
          bucketAggs: [{ type: 'term', field: 'id', id: '3' }],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '3': {
                buckets: [
                  {
                    '1': { values: { '75': 3.3, '90': 5.5 } },
                    doc_count: 10,
                    key: 'id1',
                  },
                  {
                    '1': { values: { '75': 2.3, '90': 4.5 } },
                    doc_count: 15,
                    key: 'id2',
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return table', () => {
      expect(result.data.length).toBe(1);
      expect(result.data[0].type).toBe('table');
      expect(result.data[0].columns[0].text).toBe('id');
      expect(result.data[0].columns[1].text).toBe('p75 value');
      expect(result.data[0].columns[2].text).toBe('p90 value');
      expect(result.data[0].rows.length).toBe(2);
      expect(result.data[0].rows[0][0]).toBe('id1');
      expect(result.data[0].rows[0][1]).toBe(3.3);
      expect(result.data[0].rows[0][2]).toBe(5.5);
      expect(result.data[0].rows[1][0]).toBe('id2');
      expect(result.data[0].rows[1][1]).toBe(2.3);
      expect(result.data[0].rows[1][2]).toBe(4.5);
    });
  });

  describe('Multiple metrics of same type', () => {
    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [
            { type: 'avg', id: '1', field: 'test' },
            { type: 'avg', id: '2', field: 'test2' },
          ],
          bucketAggs: [{ id: '2', type: 'terms', field: 'host' }],
        },
      ];

      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    '1': { value: 1000 },
                    '2': { value: 3000 },
                    key: 'server-1',
                    doc_count: 369,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should include field in metric name', () => {
      expect(result.data[0].type).toBe('table');
      expect(result.data[0].rows[0][1]).toBe(1000);
      expect(result.data[0].rows[0][2]).toBe(3000);
    });
  });

  describe('Raw documents query', () => {
    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [{ type: 'raw_document', id: '1' }],
          bucketAggs: [],
        },
      ];
      response = {
        responses: [
          {
            hits: {
              total: 100,
              hits: [
                {
                  _id: '1',
                  _type: 'type',
                  _index: 'index',
                  _source: { sourceProp: 'asd' },
                  fields: { fieldProp: 'field' },
                },
                {
                  _source: { sourceProp: 'asd2' },
                  fields: { fieldProp: 'field2' },
                },
              ],
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return docs', () => {
      expect(result.data.length).toBe(1);
      expect(result.data[0].type).toBe('docs');
      expect(result.data[0].total).toBe(100);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].datapoints[0].sourceProp).toBe('asd');
      expect(result.data[0].datapoints[0].fieldProp).toBe('field');
    });
  });

  describe('with bucket_script ', () => {
    let result: any;

    beforeEach(() => {
      targets = [
        {
          refId: 'A',
          metrics: [
            { id: '1', type: 'sum', field: '@value' },
            { id: '3', type: 'max', field: '@value' },
            {
              id: '4',
              field: 'select field',
              pipelineVariables: [
                { name: 'var1', pipelineAgg: '1' },
                { name: 'var2', pipelineAgg: '3' },
              ],
              settings: { script: 'params.var1 * params.var2' },
              type: 'bucket_script',
            },
          ],
          bucketAggs: [{ type: 'date_histogram', field: '@timestamp', id: '2' }],
        },
      ];
      response = {
        responses: [
          {
            aggregations: {
              '2': {
                buckets: [
                  {
                    1: { value: 2 },
                    3: { value: 3 },
                    4: { value: 6 },
                    doc_count: 60,
                    key: 1000,
                  },
                  {
                    1: { value: 3 },
                    3: { value: 4 },
                    4: { value: 12 },
                    doc_count: 60,
                    key: 2000,
                  },
                ],
              },
            },
          },
        ],
      };

      result = new ElasticResponse(targets, response).getTimeSeries();
    });

    it('should return 3 series', () => {
      expect(result.data.length).toBe(3);
      expect(result.data[0].datapoints.length).toBe(2);
      expect(result.data[0].target).toBe('Sum @value');
      expect(result.data[1].target).toBe('Max @value');
      expect(result.data[2].target).toBe('Sum @value * Max @value');
      expect(result.data[0].datapoints[0][0]).toBe(2);
      expect(result.data[1].datapoints[0][0]).toBe(3);
      expect(result.data[2].datapoints[0][0]).toBe(6);
      expect(result.data[0].datapoints[1][0]).toBe(3);
      expect(result.data[1].datapoints[1][0]).toBe(4);
      expect(result.data[2].datapoints[1][0]).toBe(12);
    });
  });

  describe('simple logs query and count', () => {
    const targets: any = [
      {
        refId: 'A',
        metrics: [{ type: 'count', id: '1' }],
        bucketAggs: [{ type: 'date_histogram', settings: { interval: 'auto' }, id: '2' }],
        context: 'explore',
        interval: '10s',
        isLogsQuery: true,
        key: 'Q-1561369883389-0.7611823271062786-0',
        liveStreaming: false,
        maxDataPoints: 1620,
        query: '',
        timeField: '@timestamp',
      },
    ];
    const response = {
      responses: [
        {
          aggregations: {
            '2': {
              buckets: [
                {
                  doc_count: 10,
                  key: 1000,
                },
                {
                  doc_count: 15,
                  key: 2000,
                },
              ],
            },
          },
          hits: {
            hits: [
              {
                _id: 'fdsfs',
                _type: '_doc',
                _index: 'mock-index',
                _source: {
                  '@timestamp': '2019-06-24T09:51:19.765Z',
                  host: 'djisaodjsoad',
                  message: 'hello, i am a message',
                  level: 'debug',
                  fields: {
                    lvl: 'debug',
                  },
                },
              },
              {
                _id: 'kdospaidopa',
                _type: '_doc',
                _index: 'mock-index',
                _source: {
                  '@timestamp': '2019-06-24T09:52:19.765Z',
                  host: 'dsalkdakdop',
                  message: 'hello, i am also message',
                  level: 'error',
                  fields: {
                    lvl: 'info',
                  },
                },
              },
            ],
          },
        },
      ],
    };

    it('should return histogram aggregation and documents', () => {
      const result = new ElasticResponse(targets, response).getLogs();
      expect(result.data.length).toBe(2);
      const logResults = result.data[0] as MutableDataFrame;
      const fields = logResults.fields.map(f => {
        return {
          name: f.name,
          type: f.type,
        };
      });

      expect(fields).toContainEqual({ name: '@timestamp', type: 'time' });
      expect(fields).toContainEqual({ name: 'host', type: 'string' });
      expect(fields).toContainEqual({ name: 'message', type: 'string' });

      let rows = new DataFrameView(logResults);
      for (let i = 0; i < rows.length; i++) {
        const r = rows.get(i);
        expect(r._id).toEqual(response.responses[0].hits.hits[i]._id);
        expect(r._type).toEqual(response.responses[0].hits.hits[i]._type);
        expect(r._index).toEqual(response.responses[0].hits.hits[i]._index);
        expect(r._source).toEqual(flatten(response.responses[0].hits.hits[i]._source, null));
      }

      // Make a map from the histogram results
      const hist: KeyValue<number> = {};
      const histogramResults = new MutableDataFrame(result.data[1]);
      rows = new DataFrameView(histogramResults);
      for (let i = 0; i < rows.length; i++) {
        const row = rows.get(i);
        hist[row.Time] = row.Count;
      }

      response.responses[0].aggregations['2'].buckets.forEach((bucket: any) => {
        expect(hist[bucket.key]).toEqual(bucket.doc_count);
      });
    });

    it('should map levels field', () => {
      const result = new ElasticResponse(targets, response).getLogs(undefined, 'level');
      const fieldCache = new FieldCache(result.data[0]);
      const field = fieldCache.getFieldByName('level');
      expect(field.values.toArray()).toEqual(['debug', 'error']);
    });

    it('should re map levels field to new field', () => {
      const result = new ElasticResponse(targets, response).getLogs(undefined, 'fields.lvl');
      const fieldCache = new FieldCache(result.data[0]);
      const field = fieldCache.getFieldByName('level');
      expect(field.values.toArray()).toEqual(['debug', 'info']);
    });
  });
});

describe('getRowContexts', () => {
  describe('when called with a DataFrame and results are returned', () => {
    it('then the result should be in correct format and filtered', async () => {
      const firstResult = new MutableDataFrame({
        refId: 'B',
        fields: [
          { name: 'ts', type: FieldType.time, values: [3, 2, 1] },
          { name: 'line', type: FieldType.string, values: ['3', '2', '1'], labels: {} },
          { name: 'id', type: FieldType.string, values: ['3', '2', '1'], labels: {} },
        ],
      });
      const secondResult = new MutableDataFrame({
        refId: 'B',
        fields: [
          { name: 'ts', type: FieldType.time, values: [6, 5, 4] },
          { name: 'line', type: FieldType.string, values: ['6', '5', '4'], labels: {} },
          { name: 'id', type: FieldType.string, values: ['6', '5', '4'], labels: {} },
        ],
      });
      let called = false;
      const getRowContextMock = (row: LogRowModel, options?: any): Promise<DataQueryResponse> => {
        if (!called) {
          called = true;
          return Promise.resolve({ data: [firstResult] });
        }
        return Promise.resolve({ data: [secondResult] });
      };

      const result = await getRowContexts(getRowContextMock, row, 10);

      expect(result).toEqual({ data: [[['3', '2']], [['6', '5', '4']]], errors: ['', ''] });
    });

    it('then the result should be in correct format and filtered without uid', async () => {
      const firstResult = new MutableDataFrame({
        refId: 'B',
        fields: [
          { name: 'ts', type: FieldType.time, values: [3, 2, 1] },
          { name: 'line', type: FieldType.string, values: ['3', '2', '1'], labels: {} },
        ],
      });
      const secondResult = new MutableDataFrame({
        refId: 'B',
        fields: [
          { name: 'ts', type: FieldType.time, values: [6, 5, 4] },
          { name: 'line', type: FieldType.string, values: ['6', '5', '4'], labels: {} },
        ],
      });
      let called = false;
      const getRowContextMock = (row: LogRowModel, options?: any): Promise<DataQueryResponse> => {
        if (!called) {
          called = true;
          return Promise.resolve({ data: [firstResult] });
        }
        return Promise.resolve({ data: [secondResult] });
      };

      const result = await getRowContexts(getRowContextMock, row, 10);

      expect(result).toEqual({ data: [[['3', '2', '1']], [['6', '5']]], errors: ['', ''] });
    });
  });

  describe('when called with a DataFrame and errors occur', () => {
    it('then the result should be in correct format', async () => {
      const firstError = new Error('Error 1');
      const secondError = new Error('Error 2');
      let called = false;
      const getRowContextMock = (row: LogRowModel, options?: any): Promise<DataQueryResponse> => {
        if (!called) {
          called = true;
          return Promise.reject(firstError);
        }
        return Promise.reject(secondError);
      };

      const result = await getRowContexts(getRowContextMock, row, 10);

      expect(result).toEqual({ data: [[], []], errors: ['Error 1', 'Error 2'] });
    });
  });
});

describe('LogDetails', () => {
  describe('when labels are present', () => {
    it('should render heading', () => {
      const wrapper = setup(undefined, { labels: { key1: 'label1', key2: 'label2' } });
      expect(wrapper.find({ 'aria-label': 'Log Labels' })).toHaveLength(1);
    });
    it('should render labels', () => {
      const wrapper = setup(undefined, { labels: { key1: 'label1', key2: 'label2' } });
      expect(wrapper.text().includes('key1label1key2label2')).toBe(true);
    });
  });
  describe('when row entry has parsable fields', () => {
    it('should render heading ', () => {
      const wrapper = setup(undefined, { entry: 'test=successful' });
      expect(wrapper.find({ title: 'Ad-hoc statistics' })).toHaveLength(1);
    });
    it('should render parsed fields', () => {
      const wrapper = setup(undefined, { entry: 'test=successful' });
      expect(wrapper.text().includes('testsuccessful')).toBe(true);
    });
  });
  describe('when row entry have parsable fields and labels are present', () => {
    it('should render all headings', () => {
      const wrapper = setup(undefined, { entry: 'test=successful', labels: { key: 'label' } });
      expect(wrapper.find({ 'aria-label': 'Log Labels' })).toHaveLength(1);
      expect(wrapper.find({ 'aria-label': 'Parsed Fields' })).toHaveLength(1);
    });
    it('should render all labels and parsed fields', () => {
      const wrapper = setup(undefined, {
        entry: 'test=successful',
        labels: { key: 'label' },
      });
      expect(wrapper.text().includes('keylabel')).toBe(true);
      expect(wrapper.text().includes('testsuccessful')).toBe(true);
    });
  });
  describe('when row entry and labels are not present', () => {
    it('should render no details available message', () => {
      const wrapper = setup(undefined, { entry: '' });
      expect(wrapper.text().includes('No details available')).toBe(true);
    });
    it('should not render headings', () => {
      const wrapper = setup(undefined, { entry: '' });
      expect(wrapper.find({ 'aria-label': 'Log labels' })).toHaveLength(0);
      expect(wrapper.find({ 'aria-label': 'Parsed fields' })).toHaveLength(0);
    });
  });

  it('should render fields from dataframe with links', () => {
    const entry = 'traceId=1234 msg="some message"';
    const dataFrame = new MutableDataFrame({
      fields: [
        { name: 'entry', values: [entry] },
        // As we have traceId in message already this will shadow it.
        {
          name: 'traceId',
          values: ['1234'],
          config: { links: [{ title: 'link', url: 'localhost:3210/${__value.text}' }] },
        },
        { name: 'userId', values: ['5678'] },
      ],
    });
    const wrapper = setup(
      {
        getFieldLinks: (field: Field, rowIndex: number) => {
          if (field.config && field.config.links) {
            return field.config.links.map(link => {
              return {
                href: link.url.replace('${__value.text}', field.values.get(rowIndex)),
                title: link.title,
                target: '_blank',
                origin: field,
              };
            });
          }
          return [];
        },
      },
      { entry, dataFrame, entryFieldIndex: 0, rowIndex: 0 }
    );
    expect(wrapper.find(LogDetailsRow).length).toBe(3);
    const traceIdRow = wrapper.find(LogDetailsRow).filter({ parsedKey: 'traceId' });
    expect(traceIdRow.length).toBe(1);
    expect(traceIdRow.find('a').length).toBe(1);
    expect((traceIdRow.find('a').getDOMNode() as HTMLAnchorElement).href).toBe('localhost:3210/1234');
  });
});

describe('loki result transformer', () => {
  afterAll(() => {
    jest.restoreAllMocks();
  });

  afterEach(() => {
    jest.clearAllMocks();
  });

  describe('legacyLogStreamToDataFrame', () => {
    it('converts streams to series', () => {
      const data = legacyStreamResult.map(stream => ResultTransformer.legacyLogStreamToDataFrame(stream));

      expect(data.length).toBe(2);
      expect(data[0].fields[1].labels['foo']).toEqual('bar');
      expect(data[0].fields[0].values.get(0)).toEqual(legacyStreamResult[0].entries[0].ts);
      expect(data[0].fields[1].values.get(0)).toEqual(legacyStreamResult[0].entries[0].line);
      expect(data[0].fields[2].values.get(0)).toEqual('2764544e18dbc3fcbeee21a573e8cd1b');
      expect(data[1].fields[0].values.get(0)).toEqual(legacyStreamResult[1].entries[0].ts);
      expect(data[1].fields[1].values.get(0)).toEqual(legacyStreamResult[1].entries[0].line);
      expect(data[1].fields[2].values.get(0)).toEqual('55b7a68547c4c1c88827f13f3cb680ed');
    });
  });

  describe('lokiLegacyStreamsToDataframes', () => {
    it('should enhance data frames', () => {
      jest.spyOn(ResultTransformer, 'enhanceDataFrame');
      const dataFrames = ResultTransformer.lokiLegacyStreamsToDataframes(
        { streams: legacyStreamResult },
        { refId: 'A' },
        500,
        {
          derivedFields: [
            {
              matcherRegex: 'tracer=(w+)',
              name: 'test',
              url: 'example.com',
            },
          ],
        }
      );

      expect(ResultTransformer.enhanceDataFrame).toBeCalled();
      dataFrames.forEach(frame => {
        expect(
          frame.fields.filter(field => field.name === 'test' && field.type === 'string').length
        ).toBeGreaterThanOrEqual(1);
      });
    });
  });

  describe('lokiStreamResultToDataFrame', () => {
    it('converts streams to series', () => {
      const data = streamResult.map(stream => ResultTransformer.lokiStreamResultToDataFrame(stream));

      expect(data.length).toBe(2);
      expect(data[0].fields[1].labels['foo']).toEqual('bar');
      expect(data[0].fields[0].values.get(0)).toEqual('2020-01-24T09:19:22.021Z');
      expect(data[0].fields[1].values.get(0)).toEqual(streamResult[0].values[0][1]);
      expect(data[0].fields[2].values.get(0)).toEqual('2b431b8a98b80b3b2c2f4cd2444ae6cb');
      expect(data[1].fields[0].values.get(0)).toEqual('2020-01-24T09:19:22.031Z');
      expect(data[1].fields[1].values.get(0)).toEqual(streamResult[1].values[0][1]);
      expect(data[1].fields[2].values.get(0)).toEqual('75d73d66cff40f9d1a1f2d5a0bf295d0');
    });
  });

  describe('lokiStreamsToDataframes', () => {
    it('should enhance data frames', () => {
      jest.spyOn(ResultTransformer, 'enhanceDataFrame');
      const dataFrames = ResultTransformer.lokiStreamsToDataframes(streamResult, { refId: 'B' }, 500, {
        derivedFields: [
          {
            matcherRegex: 'trace=(w+)',
            name: 'test',
            url: 'example.com',
          },
        ],
      });

      expect(ResultTransformer.enhanceDataFrame).toBeCalled();
      dataFrames.forEach(frame => {
        expect(
          frame.fields.filter(field => field.name === 'test' && field.type === 'string').length
        ).toBeGreaterThanOrEqual(1);
      });
    });
  });

  describe('appendResponseToBufferedData', () => {
    it('should append response', () => {
      const data = new MutableDataFrame();
      data.addField({ name: 'ts', type: FieldType.time, config: { title: 'Time' } });
      data.addField({ name: 'line', type: FieldType.string });
      data.addField({ name: 'labels', type: FieldType.other });
      data.addField({ name: 'id', type: FieldType.string });

      ResultTransformer.appendLegacyResponseToBufferedData({ streams: legacyStreamResult }, data);
      expect(data.get(0)).toEqual({
        ts: '1970-01-01T00:00:00Z',
        line: "foo: 'bar'",
        labels: { foo: 'bar' },
        id: '2764544e18dbc3fcbeee21a573e8cd1b',
      });
    });

    it('should return a dataframe with ts in iso format', () => {
      const tailResponse: LokiTailResponse = {
        streams: [
          {
            stream: {
              filename: '/var/log/grafana/grafana.log',
              job: 'grafana',
            },
            values: [
              [
                '1581519914265798400',
                't=2020-02-12T15:04:51+0000 lvl=info msg="Starting Grafana" logger=server version=6.7.0-pre commit=6f09bc9fb4 branch=issue-21929 compiled=2020-02-11T20:43:28+0000',
              ],
            ],
          },
        ],
      };

      const data = new CircularDataFrame({ capacity: 1 });
      data.addField({ name: 'ts', type: FieldType.time, config: { title: 'Time' } });
      data.addField({ name: 'tsNs', type: FieldType.time, config: { title: 'Time ns' } });
      data.addField({ name: 'line', type: FieldType.string }).labels = { job: 'grafana' };
      data.addField({ name: 'labels', type: FieldType.other });
      data.addField({ name: 'id', type: FieldType.string });

      ResultTransformer.appendResponseToBufferedData(tailResponse, data);
      expect(data.get(0)).toEqual({
        ts: '2020-02-12T15:05:14.265Z',
        tsNs: '1581519914265798400',
        line:
          't=2020-02-12T15:04:51+0000 lvl=info msg="Starting Grafana" logger=server version=6.7.0-pre commit=6f09bc9fb4 branch=issue-21929 compiled=2020-02-11T20:43:28+0000',
        labels: { filename: '/var/log/grafana/grafana.log' },
        id: '19e8e093d70122b3b53cb6e24efd6e2d',
      });
    });
  });
});

describe('FieldOverrides', () => {
  beforeAll(() => {
    standardFieldConfigEditorRegistry.setInit(getStandardFieldConfigs);
  });

  const f0 = new MutableDataFrame();
  f0.add({ title: 'AAA', value: 100, value2: 1234 }, true);
  f0.add({ title: 'BBB', value: -20 }, true);
  f0.add({ title: 'CCC', value: 200, value2: 1000 }, true);
  expect(f0.length).toEqual(3);

  // Hardcode the max value
  f0.fields[1].config.max = 0;
  f0.fields[1].config.decimals = 6;

  const src: FieldConfigSource = {
    defaults: {
      unit: 'xyz',
      decimals: 2,
    },
    overrides: [
      {
        matcher: { id: FieldMatcherID.numeric },
        properties: [
          { prop: 'decimals', value: 1 }, // Numeric
          { prop: 'title', value: 'Kittens' }, // Text
        ],
      },
    ],
  };

  it('will merge FieldConfig with default values', () => {
    const field: FieldConfig = {
      min: 0,
      max: 100,
    };
    const f1 = {
      unit: 'ms',
      dateFormat: '', // should be ignored
      max: parseFloat('NOPE'), // should be ignored
      min: null, // should alo be ignored!
    };

    const f: DataFrame = toDataFrame({
      fields: [{ type: FieldType.number, name: 'x', config: field, values: [] }],
    });
    const processed = applyFieldOverrides({
      data: [f],
      standard: standardFieldConfigEditorRegistry,
      fieldOptions: {
        defaults: f1 as FieldConfig,
        overrides: [],
      },
      replaceVariables: v => v,
      theme: getTheme(),
    })[0];
    const out = processed.fields[0].config;

    expect(out.min).toEqual(0);
    expect(out.max).toEqual(100);
    expect(out.unit).toEqual('ms');
  });

  it('will apply field overrides', () => {
    const data = applyFieldOverrides({
      data: [f0], // the frame
      fieldOptions: src as FieldDisplayOptions, // defaults + overrides
      replaceVariables: (undefined as any) as InterpolateFunction,
      theme: (undefined as any) as GrafanaTheme,
    })[0];
    const valueColumn = data.fields[1];
    const config = valueColumn.config;

    // Keep max from the original setting
    expect(config.max).toEqual(0);

    // Don't Automatically pick the min value
    expect(config.min).toEqual(undefined);

    // The default value applied
    expect(config.unit).toEqual('xyz');

    // The default value applied
    expect(config.title).toEqual('Kittens');

    // The override applied
    expect(config.decimals).toEqual(1);
  });

  it('will apply set min/max when asked', () => {
    const data = applyFieldOverrides({
      data: [f0], // the frame
      fieldOptions: src as FieldDisplayOptions, // defaults + overrides
      replaceVariables: (undefined as any) as InterpolateFunction,
      theme: (undefined as any) as GrafanaTheme,
      autoMinMax: true,
    })[0];
    const valueColumn = data.fields[1];
    const config = valueColumn.config;

    // Keep max from the original setting
    expect(config.max).toEqual(0);

    // Don't Automatically pick the min value
    expect(config.min).toEqual(-20);
  });
});