TransformPart - Scoped, Global

  • Rversion finale: Australia
  • Mis à jour 12 mars 2026
  • 12 minutes de lecture

    • The TransformPart API provides methods to specify details of the transform to be done.

    You can call this class in scoped and global server scripts. When using the TransformPart class, use the sn_clotho namespace identifier.

    There is no constructor for this class. TransformPart objects are returned by many Transformer and TransformPart methods.

    The methods of this class define the transforms to be done. The actual transformation is done when the execute() method is called on the Transformer object.

    The order the TransformPart methods are called is important.
    • You must call the metric() method before calling a transform method.
    • You cannot use the metric() or groupBy() methods after calling a transform method.
    • Intermediate transforms are not returned in a result unless the collect() method is called for the intermediate result you want.
      // where tp is a TransformPart object
// example 1
tp.avg().add(2);
var tr = tp.execute(); 
// tr contains avg+2, but not avg
 
// example 2
tp.avg().add(2);
tp.avg();
var tr = tp.execute(); 
// tr contains both avg and avg + 2

// example 3
tp.avg().collect().add(2);
var tr = tp.execute(); 
// tr contains both avg and avg + 2

    This class is part of the MetricBase application.

    TransformPart - add(Number constant)

    Add the specified number to the value in each time stamp.

    Tableau 1. Parameters
    Name Type Description
    constant Number The number to add to the value in each time stamp.
    Tableau 2. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - avg()

    Aggregate the selected metric series into one series containing the average value for each time stamp.

    Tableau 3. Parameters
    Name Type Description
    None
    Tableau 4. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - bottom(Number count)

    Create a result set that for each time stamp returns specified number of bottom values. This method results in 'count' number of series. Each value retains the label of its source series.

    Tableau 5. Parameters
    Name Type Description
    count Number The number of series to return. The series are labeled 0 to count - 1.
    Tableau 6. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - ceil(Number ceiling)

    Replace the value in any time stamp that is greater than the specified value with the specified value.

    Tableau 7. Parameters
    Name Type Description
    ceiling Number The maximum allowed value for any time stamp.
    Tableau 8. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - collect()

    Mark this transform for collection.

    Transforms that are part of a chain, but not the last transform, are by default not collected. A collected transform is returned as part of the transform result.

    Tableau 9. Parameters
    Name Type Description
    None
    Tableau 10. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - count()

    Aggregate the selected metric series into one series containing the number of values for each time stamp.

    Tableau 11. Parameters
    Name Type Description
    None
    Tableau 12. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - div(Number constant)

    Divide the value in each time stamp by the specified number.

    Tableau 13. Parameters
    Name Type Description
    constant Number The number by which to divide the value of each time stamp.
    Tableau 14. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - filter(Object aggregator, Object duration)

    Create a series using the specified aggregator for the specified time.

    Tableau 15. Parameters
    Name Type Description
    aggregator Object Can be:
    • AVG
    • CHISQUARE
    • LAST
    • MAX
    • MEDIAN
    • MIN
    • STDDEV
    For definitions of these options, see MetricBase transforms.
    duration Object Time period for the series.
    Tableau 16. Returns
    Type Description
    TransformPart TransformPart object that you can use to specify transform characteristics.

    TransformPart - floor(Number floor)

    Replace the value in any time stamp that is less than the specified value with the specified value.

    Tableau 17. Parameters
    Name Type Description
    floor Number The minimum value for any time stamp.
    Tableau 18. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - fractiles(Array fractions)

    Create series made up of the value that the specified percentage of values is below. Returns a series for each fraction in the specified array.

    The value in a time stamp in a returned series is the value at which the specified fraction of the samples for that time stamp is below. For example, if the fraction is 0.5, then the value in the time stamp is the value where half the values in the input series are below (median).

    Tableau 19. Parameters
    Name Type Description
    fractions Array of numbers The fractions to use on the input series.
    Tableau 20. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics. Contains one series for each fraction specified. 
    // returns a single series containing the median for each time stamp, which
// means that half a time stamp's values are below the returned value 
fractiles([.5]) 
// returns four series, one series for each of the 25%, 50%, 75%, and 100% quartiles 
fractiles([.25, .5, .75, 1])
// returns the median, 95% percentile, the max value
fractiles([.50, .95, 1])

    TransformPart - getResult()

    Return the part of the result relevant to this transform.

    The collect() method must be called before the execute() method, and the execute() method must be called before calling the getResult() method.

    Tableau 21. Parameters
    Name Type Description
    None
    Tableau 22. Returns
    Type Description
    TransformResult Contains the transform results associated with this part of the transform. 
    var t = new sn_clotho.Transformer(drones);
t.metric("mb_demo_mt_altitude");
var avgTform = t.avg();
t.execute();
var avgTformResult = avgTform.getResult();

    TransformPart - groupBy(String field)

    Specify a field by which to group data. To specify multiple fields, run successive calls individually.

    Remarque :
    The groupBy() method cannot be called after a transform has been run.
    Tableau 23. Parameters
    Name Type Description
    field String Name of a field in the table by which to group the transform results.
    Tableau 24. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    The following example shows how to group altitude average metrics by fleet and model fields.

    transformer.groupBy("fleet,model").metric("mb_demo_mt_altitude").avg().label('%g:fleet: - %g:model:');

var transformPart = transformer.groupBy("fleet,model")

transformPart.metric("mb_demo_mt_altitude").avg().label('%g:fleet: - %g:model:');

    TransformPart - interpolate(Object count)

    Create a data value for a NaN data item by interpolating from adjacent data values.

    Tableau 25. Parameters
    Name Type Description
    count Object Specifies the number of data samples in each direction to check for a non NaN value. If if a non NaN value is not found, NaN is used.
    Tableau 26. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - iqr()

    Perform an Interquartile range transform.

    Creates a result set of four series.
    • IQR, the median of all entries
    • IQR range, below Q1-1.5IQR, or above Q3+1.5IQR
    • Q1, the median of the smallest half of entries
    • Q3, the median of the largest half of entries
    Tableau 27. Parameters
    Name Type Description
    None
    Tableau 28. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - label(String label)

    Add a label for the resulting series.

    Tableau 29. Parameters
    Name Type Description
    label String The label for the transform results.
    Tableau 30. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - limit(Object count)

    Returns at most the specified number of values, starting at the most recent non-NaN value.

    Tableau 31. Parameters
    Name Type Description
    count Object A number of time stamps.
    Tableau 32. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - log(Number base)

    Run a logarithm on the value in each time stamp where the result is the log of the specified base for the time stamp value.

    Tableau 33. Parameters
    Name Type Description
    base Number The base for the logarithm calculation.
    Tableau 34. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - max()

    Returns a series with the maximum value for each time stamp.

    Tableau 35. Parameters
    Name Type Description
    None
    Tableau 36. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - median()

    Create a series containing the median of values for each time stamp across a set of series.

    If there are n series in the TranformPart object, then if n is odd, the (n / 2 + 1) value for a time stamp is the median. If n is even, the average of the (n / 2) and (n / 2 + 1) values for a time stamp is the median.

    Tableau 37. Parameters
    Name Type Description
    None
    Tableau 38. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - metric(String metric)

    Specify the metric field to be used in the transform.

    You can specify multiple metrics to be used in the transform. The metric() method cannot be called after the transform has been run.

    Tableau 39. Parameters
    Name Type Description
    metric String Name of the metric field.
    Tableau 40. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - min()

    Returns a series with the minimum value for each time stamp.

    Tableau 41. Parameters
    Name Type Description
    None
    Tableau 42. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - mul(Number constant)

    Multiply the value in each time stamp by the specified number.

    Tableau 43. Parameters
    Name Type Description
    constant Number The number by which to multiply the value of each time stamp.
    Tableau 44. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - partition(String aggregator, GlideDateTime duration, GlideDateTime base)

    Partition the series into intervals of the same duration.

    Tableau 45. Parameters
    Name Type Description
    aggregator String The aggregator to use. Can be min, max, avg, or last.
    duration GlideDateTime or an ISO 8601 formatted string The interval length.
    base GlideDateTime or an ISO 8601 formatted string The zero offset for partitioning. For example, if you partition by day (24h), then set the base to Monday at midnight in your time zone. If you partition by 30 days, then set the base to 1st day of the most recent month.
    Tableau 46. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - resample(Number count)

    Specify the number of data points to include in the result.

    Aligns a series with a fixed number of data points in the given range. If the original series has more data points than specified, multiple values are averaged. If the original series has fewer data points than specified, data points are added by interpolating data points between existing data points.

    You can use the resample() method to reduce the number of samples in the result to more closely match the number of samples you are going to display.

    Tableau 47. Parameters
    Name Type Description
    count Number The number of samples to include in the result.
    Tableau 48. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - resample(Number min, Number max)

    Specify the minimum and maximum number of samples to include in the result.

    This method is useful when you are showing series with different time periods (granularities).

    This method determines the average number of points per series and if that fits between the specified minimum and maximum, each series is resampled to that average number of points. If the calculated average is greater than the maximum specified or smaller than the minimum specified, the specified maximum or minimum is used.

    Tableau 49. Parameters
    Name Type Description
    min Number The minimum number of samples to include in the result. If not enough samples are available, interpolation is used to create samples.
    max Number The maximum number of samples to include in the result.
    Tableau 50. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - resample(String aggregator, Number numValues)

    Specify an aggregator to use to create a result set of the specified size. The aggregator can be LAST, AVG, MIN, or MAX.

    You can use the resample() method to reduce the number of samples in the result to more closely match the number of samples you are going to display.

    Tableau 51. Parameters
    Name Type Description
    aggregator String Can be LAST, AVG, MIN, or MAX.
    numValues Number The number of samples to include in the result set.

    When the number of values requested is greater than the number of values in the data for the requested time period, interpolate() is used to add values between existing points to reach the requested number of values.
    Tableau 52. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - resample(String aggregator, GlideDuration duration)

    Specify an aggregator to use to create a result set over the specified duration. The aggregator can be LAST, AVG, MIN, or MAX.

    Resamples the underlying data to the requested period.
    • When the time series is stored at a 1 minute interval and resample(AVG, new GlideDuration("1:00")) is called, the result will have one data point at every hour reflecting the average of the previous 60 values.
    • When the period requested is smaller than the period of the data, interpolation is used to calculate the intervening data points.

    You can use the resample() method to reduce the number of samples in the result to more closely match the number of samples you are going to display.

    Tableau 53. Parameters
    Name Type Description
    aggregator String Can be LAST, AVG, MIN, or MAX.
    duration GlideDuration The time period for the result set.
    Tableau 54. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - round(Number precision)

    Round the value in each time stamp to the specified precision.

    Performs this calculation on each value.
    (v / precision) * precision
    Tableau 55. Parameters
    Name Type Description
    precision Number The value to be used in the rounding calculation.
    Tableau 56. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - stddev()

    Create a series containing the standard deviation of values for each time stamp across a set of series.

    Tableau 57. Parameters
    Name Type Description
    None
    Tableau 58. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - sub(Object constant)

    Subtract the specified number from the value in each time stamp.

    Tableau 59. Parameters
    Name Type Description
    constant Object The number to subtract from the value in each time stamp.
    Tableau 60. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - sum()

    Aggregate the selected metric series into one series containing the sum of all values for each time stamp.

    Tableau 61. Parameters
    Name Type Description
    None
    Tableau 62. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.

    TransformPart - top(Number count)

    Create a result set that for each time stamp returns the specified number of top values. This method results in 'count' number of series. Each value retains the label of its source series.

    Tableau 63. Parameters
    Name Type Description
    count Number The number of series to return. The series are labeled 0 to count - 1.
    Tableau 64. Returns
    Type Description
    TransformPart A TransformPart object that can be used to specify transform characteristics.