Mapper Functions
The mapper functions transforms the stream one message to another message 1 to 1
.
FieldMapper
The FieldMapper is a function that allows us to add fields to one event.
{
"name": "myFieldMapper",
"className": "io.wizzie.normalizer.funcs.impl.FieldMapper",
"properties": {
"dimensions": [
{
"dimension": "dimension1",
"value": "defaultValue1",
"overwrite": false
},
{
"dimension": "dimension2",
"value": "defaultValue2",
"overwrite": true
},
{
"dimension": "dimension3",
"value": "defaultValue3"
}
]
}
}
The FieldMapper has one property that is called dimensions
on this property you define the fields that you want to add and if you want overwrite them if they exists. If we have this json message:
{
"dimension1":"value1",
"dimension2":"value2",
"dimension3": "value3",
"timestamp": 123456789
}
If we use this message using the FieldMapper that is defined on the above example, we get this output:
{"dimension1":"value1", "dimension2":"defaultValue2", "dimension3": "value3", "timestamp": 123456788}
By default the FieldMapper will not overwrite the values if you don’t specify the overwrite property.
SimpleMapper
The SimpleMapper is a function that allows us to simplify the JSON Object into one level or to delete fields. It also selects different fields from JSON Object and renames it.
{
"name":"myMapper",
"className":"io.wizzie.normalizer.funcs.impl.SimpleMapper",
"properties": {
"maps": [
{"dimPath":["A","B","C"], "as":"X"},
{"dimPath":["Y","W","Z"], "as":"Q"},
{"dimPath":["Y","W","P"]},
{"dimPath":["timestamp"]}
]
}
}
The SimpleMapper has three properties that are called maps
, deleteMode
and deleteEmpty
:
maps
: on this property you define the fields that you want to select and if you want rename it..deleteMode
: a boolean value to switch to delete mode. On delete mode the dimensions atmaps
will be removed and the others will stay. It’sfalse
by default.deleteEmpty
: a boolean value to specify if, after deleting in delete mode, remove the resulting empty maps. By default it’s true.
{
"A": {
"B": {
"C": "MyValue"
},
"D": "A"
},
"Y": {
"W": {
"P": 123456,
"Z": "MyOtherValue"
}
},
"timestamp": 123456788
}
If we use this message using the SimpleMapper that is defined on the above example, we get this output:
{"X":"MyValue", "Q":"MyOtherValue", "P": 123456, "timestamp": 123456788}
If we define the next SimpleMapper:
{
"name":"myMapper",
"className":"io.wizzie.normalizer.funcs.impl.SimpleMapper",
"properties": {
"maps": [
{"dimPath":["timestamp"]},
{"dimPath":["value"]}
],
"deleteMode": true
}
}
And we have the next message:
{"timestamp":"1234", "value":"myvalue", "P": 123456}
The result will be:
{"P": 123456}
And if we define the next SimpleMapper:
{
"name":"myMapper",
"className":"io.wizzie.normalizer.funcs.impl.SimpleMapper",
"properties": {
"maps": [
{"dimPath":["A","B","C"]},
{"dimPath":["value"]}
],
"deleteMode": true,
"deleteEmpty": false
}
}
And we have the next message:
{"timestamp":"1234", "value":"myvalue", "P": 123456, "A":{"B":{"C":{"myKey":"myValue"}}}}
The result will be:
{"timestamp": "1234", "P":123456, "A":{"B":{}}}
ReplaceMapper
The ReplaceMapper is a function that allows us to replace current values of messages by others that define by us. The replace value should exists.
{
"name": "myReplaceMapper",
"className": "io.wizzie.normalizer.funcs.impl.ReplaceMapper",
"properties": {
"replacements":[
{
"dimension": "type",
"replacements": [
{"from": "ver", "to": "version"},
{"from": "v", "to": "version"},
{"from": "vrsn", "to": "version"},
{"from": 9, "to": 10},
{"from": "nine", "to": 9}
]
}
]
}
}
The ReplaceMapper has one property that is called replacements
. It is an array where we define the replace values:
dimension
: The dimension that indicate the current value to replace.replacements
: Key-Value pairs for replace the current value. The value fromfrom
will be replace with the one atto
.
If we have this json message:
{
"type":"ver",
"myValue":"VALUE-2",
"myDimension":"currentValue2"
}
If we use this message using the ReplaceMapper that is defined on the above example, we get this output:
{
"type":"version",
"myValue":"VALUE-2",
"myDimension":"replaceValue2"
}
Note: this function supports strings, numbers or boolean replacements. You can use them at your convenience.
JoinMapper
The JoinMapper is a function that allows us to join as many values as we want and assign them in other dimension.
{
"name":"myJoinMapper",
"className":"io.wizzie.normalizer.funcs.impl.JoinMapper",
"properties": {
"dimensionName":"myNewDimension",
"values": [
{"fromDimension":"dimension1", "orDefault":"defaultValue1", "delete": false},
{"fromDimension":"dimension2", "orDefault":"defaultValue2", "delete": true},
{"fromDimension":"dimension3", "orDefault":"defaultValue3"}
],
"delimiter": "-"
}
}
The JoinMapper has three properties that are called dimensionName
, values
and delimiter
:
dimensionName
: Name of new dimension where all join values will be assign. This parameter can’t be null.values
: Array of values, contains a sequence of items with two parametersfromDimension
andorDefault
.fromDimension
: Dimension where we get the value to join. This parameter can’t be null.orDefault
: Default value if value offromDimension
doesn’t exists. This parameter can’t be null.delete
: Delete dimension defined infromDimension
parameter. Default value isfalse
.delimiter
: Separator for each dimension. By default is-
.
If we have this JSON message:
{
"dimension1":"A",
"dimension2":"B",
"timestamp":123456789
}
If we use this message using the JoinMapper that is defined on the above example, we get this output:
{
"dimension1":"A",
"myNewDimension":"A-B-defaultValue3",
"timestamp":123456789
}
MaxValueMapper
The MaxValueMapper is a function that allow us from an array of numbers which is greater. The max value is detected and store in other dimension
{
"name":"myMaxValueMapper",
"className":"io.wizzie.normalizer.funcs.impl.MaxValueMapper",
"properties": {
"dimension": "measures",
"max_dimension_name": "max_measure"
}
}
The MaxValueMapper have two properties named dimension
and max_dimension_name
:
dimension
: Dimension where the numbers are located.max_dimension_name
: The dimension where to save the max value detected.
I we have next json message:
{
"type": "measures",
"timestamp": 123456789,
"measures": [1.70, 1.65, 1.72, 1.8, 1.8, 1.9, 1.86]
}
If we use this message using the MaxValueMapper that is defined on the above example, we get next output:
{
"type": "measures",
"timestamp": 123456789,
"measures": [1.70, 1.65, 1.72, 1.8, 1.8, 1.9, 1.86],
"max_measure": 1.9
}
MinValueMapper
The MinValueMapper is like MaxValueMapper function, except that this function locate the smaller number in a array number.
{
"name":"myMinValueMapper",
"className":"io.wizzie.normalizer.funcs.impl.MinValueMapper",
"properties": {
"dimension": "measures",
"min_dimension_name": "min_measure"
}
}
The MinValueMapper like MaxValueMapper also has two properties dimension
and min_dimension_name
:
dimension
: Dimension where the numbers are located.min_dimension_name
: The dimension where to save the min value detected.
If we have next json message:
{
"type": "measures",
"timestamp": 123456789,
"measures": [1.70, 1.65, 1.72, 1.8, 1.8, 1.9, 1.86]
}
If we use this message using the MinValueMapper that is defined on the above example, we get next output:
{
"type": "measures",
"timestamp": 123456789,
"measures": [1.70, 1.65, 1.72, 1.8, 1.8, 1.9, 1.86],
"min_measure": 1.65
}
ClassificationMapper
The ClassficationMapper allows us to classify a numeric value.
{
"name":"myClassificationMapper",
"className":"io.wizzie.normalizer.funcs.impl.ClassificationMapper",
"properties": {
"dimension": "mark",
"new_dimension": "classification",
"classification": ["F", "D", "C", "B", "A"],
"intervals": [49, 60, 71, 85],
"unknown_value": -1
}
}
The ClassificationMapper has five properties:
dimension
: The dimension that indicate the numeric value to classify.new_dimension
: Dimension where put classification value.classification
: Array of strings where put the classification names.intervals
: Array of limit for each classification.unknown_value
: Single value forunknown
classification.
In the function definition we are classify exams of Alberta Senior High School. We assume next message:
{
"first_name": "John",
"last_name": "Doe",
"subject": "Chemistry",
"mark": 75
}
If we use this message using the ClassificationMapper that is defined on the example above, we get next output:
{
"first_name": "John",
"last_name": "Doe",
"subject": "Chemistry",
"mark": 75,
"classification": "B"
}
StringSplitterMapper
The StringSplitterMapper allows us to split one dimension into multiple dimension.
{
"name":"myStringSplitterFunction",
"className":"io.wizzie.normalizer.funcs.impl.StringSplitterMapper",
"properties": {
"dimension": "DIM-H",
"delimiter": ">",
"fields": ["country", "province", "city"]
}
}
This mapper has some properties:
dimension
: The dimension field that you want to split.delimiter
: The character that the mapper uses to split.fields
: The new fields to the splitter dimensions. This is a JSON Array.delete_dimension
: This is a boolean to indicate if you want to delete the original dimension. Default: false
Input:
{"timestamp": 1477379967, "DIM-H": "Spain>Andalucia>Sevilla"}
Output:
{"timestamp": 1477379967, "country": "Spain", "province": "Andalucia", "city":"Sevilla", "DIM-H": "Spain>Andalucia>Sevilla"}
StringReplaceMapper
The StringReplaceMapper replaces the dimension string value to another one.
{
"name":"myStringReplacementFunction",
"className":"io.wizzie.normalizer.funcs.impl.StringReplaceMapper",
"properties": {
"dimension": "DIM-C",
"target_string": "-",
"replacement_string": ":"
}
}
This mapper has some properties:
dimension
: The dimension that you want to transform.target_string
: The string sequence that you want to replace.replacement_string
: The string sequence that you want to use on the change.
Input:
{"timestamp": 1477379967, "DIM-C": "00-00-AA-FF-11-33"}
Output:
{"timestamp": 1477379967, "DIM-C": "00:00:AA:FF:11:33"}
TimeMapper
The TimeMapper converts different time formats to a specified format.
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"millis",
"toFormat":"secs",
"forceStringOutput": "false"
}
}
This mapper has some properties:
dimension
: The dimension that you want to transform.fromFormat
: The format that will be received by the function.toFormat
: The format that you want at the function output.forceStringOutput
: This property forces the output to be a string. i.e. if you want to have{"timestamp": "1477379967"}
as time output instead of{"timestamp": 1477379967}
you have to set this property to true. Its default value is false.forceTimestamp
: This property forces to output a timestamp if it doesn’t exists at the event. i.e. if you want to have{"timestamp": 1477379967,"a":"hello-world"}
as output when you have:{"a": "hello-word"}
you have to set this property to true. Its default value is true. When forcing timestamp, if you have a null timestamp value at input, current time will be sended at output.
Both fromFormat
and toFormat
must be: “ISO”, “millis”, “secs” or “pattern: …”.
If you choose “pattern: …” as fromFormat or toFormat you have to specify a valid format. A valid format is a JDK date format (you can read more at: Formats and at JDK docs).
Example 1: millis -> secs
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"millis",
"toFormat":"secs",
"forceStringOutput": "false",
"forceTimestamp": "true"
}
}
Input:
{"timestamp": 1234567890000, "DIM-C": "00:00:AA:FF:11:33"}
Output:
{"timestamp": 1234567890, "DIM-C": "00:00:AA:FF:11:33"}
Example 2: millis -> secs (force string)
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"millis",
"toFormat":"secs",
"forceStringOutput": "true"
}
}
Input:
{"timestamp": 1234567890000, "DIM-C": "00:00:AA:FF:11:33"}
Output:
{"timestamp": "1234567890", "DIM-C": "00:00:AA:FF:11:33"}
Example 3: ISO -> secs
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"ISO",
"toFormat":"secs",
"forceStringOutput": "false"
}
}
Input:
{"timestamp": "2009-02-13T23:31:30.000Z", "DIM-C": "00:00:AA:FF:11:33"}
Output:
{"timestamp": 1234567890, "DIM-C": "00:00:AA:FF:11:33"}
Example 4: secs -> pattern
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"secs",
"toFormat":"pattern: yyyyMMdd",
"forceStringOutput": "false"
}
}
Input:
{"timestamp": 1234567890, "DIM-C": "00:00:AA:FF:11:33"}
Output:
{"timestamp": "20090213", "DIM-C": "00:00:AA:FF:11:33"}
Example 5: pattern -> pattern
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"pattern: yyyy-MM-dd",
"toFormat":"pattern: yyyyMMdd",
"forceStringOutput": "false"
}
}
Input:
{"timestamp": "2009-02-13", "DIM-C": "00:00:AA:FF:11:33"}
Output:
{"timestamp": "20090213", "DIM-C": "00:00:AA:FF:11:33"}
Example 6: millis -> secs (no force timestamp)
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"millis",
"toFormat":"secs",
"forceStringOutput": "false",
"forceTimestamp": "false"
}
}
Input:
{"DIM-C": "00:00:AA:FF:11:33"}
Output:
{"DIM-C": "00:00:AA:FF:11:33"}
Example 7: millis -> secs (forcing timestamp)
{
"name": "myTimeMapper",
"className": "io.wizzie.normalizer.funcs.impl.TimeMapper",
"properties": {
"dimension":"timestamp",
"fromFormat":"millis",
"toFormat":"secs",
"forceStringOutput": "false",
"forceTimestamp": "true"
}
}
Input:
{"DIM-C": "00:00:AA:FF:11:33"}
Output:
{"timestamp": 1234567890, "DIM-C": "00:00:AA:FF:11:33"}
FieldTypeConverterMapper
The FieldTypeConverterMapper allows us to convert value types of several dimensions into other types.
{
"name":"myFieldTypeConverterMapper",
"className":"io.wizzie.normalizer.funcs.impl.FieldTypeConverterMapper",
"properties": {
"conversions": [
{
"dimension": "field-A",
"from": "string",
"to": "boolean"
},
{
"dimension": "timestamp",
"from": "string",
"to": "number"
},
{
"dimension": "field-B",
"from": "number",
"to": "boolean",
"newDimension": "isPositive"
}
]
}
}
Input:
{"timestamp": "1503391561", "field-A": "true", "field-B": 1.35}
Output:
{"timestamp": 1503391561, "field-A": true, "field-B": 1.35, "isPositive": true}
This mapper has some properties:
conversions
: Set of conversions that you want to apply.dimension
: The dimension field which contains the value that you want to convert.from
: The source type:NUMBER
,STRING
orBOOLEAN
.to
: The destiny type:NUMBER
,STRING
orBOOLEAN
.newDimension
: If you set this property then the new value field will be renamed. Elsedimension
will be replace.
You can convert next types:
NUMBER
: This type representslong
,integer
,short
,byte
,double
andfloat
STRING
: This type represents strings.BOOLEAN
: This type represents boolean values:true
orfalse
Next table shows available conversion between types:
FROM\TO | NUMBER | STRING | BOOLEAN |
---|---|---|---|
NUMBER | - | ✓ | ✓ |
STRING | ✓ | - | ✓ |
BOOLEAN | ✓ | ✓ | - |
Below you can found a description about conversion between types:
Conversion | Description |
---|---|
NUMBER → NUMBER | Without effect. |
NUMBER → STRING | Convert any number type into string type. If the number value is 1.35 then it will be converted into "1.35" |
NUMBER → BOOLEAN | Convert any number type into boolean type. If number is positive then the returned value will be true , else if numer is negative then the returned value will be false |
STRING → NUMBER | Convert any number string type into number type value. If the string value is "1.35" then it will be converted into 1.35 . |
STRING → STRING | Without effect. |
STRING → BOOLEAN | Convert literal boolean string type into boolean type. If string is “true” then the returned value will be true , else if string values is “false” then the returned value will be false . |
BOOLEAN → NUMBER | Convert any boolean type into integer number type. If boolean is true then the returned value will be 1 else if boolean values is false then the returned value will be 0 . |
BOOLEAN → STRING | Convert any boolean type into string type. If boolean value is true then the returned value will be "true" , else if boolean value is false then the returned value will be "false" . |
BOOLEAN → BOOLEAN | Without effect. |
ArithmeticMapper
The Arithmetic mapper allows to evaluate math expressions over the json fields.
{
"name": "myArithmeticMapper",
"className": "io.wizzie.normalizer.funcs.impl.ArithmeticMapper",
"properties": {
"equations": [
{
"dimensions": ["field1", "field2"],
"equation": "field1+field2",
"as": "sum"
},
{
"dimensions": [
"field1",
"field3"
],
"equation": "field1-field3",
"as": "subtract"
}
]
}
}
This mapper has one property:
-
equations
: The equations that are going to be evaluated. The equations are maps containing this three values:dimensions
: The dimensions that are going to be used at the equation. They are used to map the incoming json values to the solver.equation
: The equation that is going to be evaluatedas
: The key of the field that will be used to insert the equation result.
For the function defined above, if we have:
Input:
{"timestamp": 1477379967, "field1": 3, "field2": 4}
You will get as output:
Output:
{"timestamp": 1477379967, "field1": 3, "field2": 4, "sum": 7.0}
The second equation field1-field3
is not evaluated because the field field3
is not contained at the input json.
If the input is:
{"timestamp": 1477379967, "field1": 3, "field2": 4, "field3": 1}
The output would be:
{"timestamp": 1477379967, "field1": 3, "field2": 4, "field3": 1, "sum": 7.0, "subtract": 2.0}
This function supports long, integer and double for input formats and it outputs double values.
Supported operations and library used
The library used for doing arithmetic operations is: parsii . You can check the supported operations inspecting the java classes of this project.
Some of the operations supported are:
Operation | Arguments | Example | Explanation |
---|---|---|---|
+ | none | a+b | add values |
- | none | a-b | subtract values |
/ | none | a/b | divide values |
* | none | a*b | multiply values |
sqrt | double a | sqrt(a) | square root value |
floor | double a | floor(a) | select integer part of the number |
sin | double a | sin(a) | sine of a |
cos | double a | cos(a) | cosine of a |
tan | double a | tan(a) | tangent of a |
round | double a | round(a) | round to the closest integer |
ceil | double a | ceil(a) | select next integer |
pow | double a, double b | pow(a,b) | a^b |
exp | double a | exp(a) | 10^a |
log | double a | log(a) | ln(a) |
log10 | double a | log10(a) | log10(a) |
min | double a, double b | min(a,b) | min value of a and b |
max | double a, double b | max(a,b) | max value of a and b |
random | double a | random(a) | a random number between 0 and a |
toDegrees | double a | toDegrees(a) | a to degrees |
toRadians | double a | toRadians(a) | a to radians |
SimpleArrayMapper
The SimpleArrayMapper is a function that allows us to create new dimensions from an ArrayList.
{
"name":"myArrayMapper",
"className":"io.wizzie.normalizer.funcs.impl.SimpleArrayMapper",
"properties": {
"dimension": "dimArray",
"dimensionToIndex":{"a": 0, "b": 1, "c": 2},
"deleteDimension": true
}
}
This mapper has some properties:
dimensionToIndex
: On this property you define the dimensions you want to create based on array index.dimension
: The array list dimension to process.deleteDimension
: If you want to remove the source dimension. Default: true.
Example:
- Input:
{"timestamp": 123456789, "dimArray":["ABC", 123, "CDX"]}
- Output:
{"timestamp": 123456789, "a":"ABC", "b":123, "c":"CDX"}
RenameMapper
The RenameMapper is a function that allows us to rename fields from JSON Object.
{
"name":"myMapper",
"className":"io.wizzie.normalizer.funcs.impl.RenameMapper",
"properties": {
"maps": [
{"dimPath":["C"], "as":"X"},
{"dimPath":["Z"], "as":"Q"}
]
}
}
The RenameMapper has one property that is called maps
on this property you define the fields that you want to rename:
{
"C": "MyValue",
"D": "A",
"Z": "MyOtherValue",
"timestamp": 123456788
}
If we use this message using the RenameMapper that is defined on the above example, we get this output:
{"X":"MyValue", "Q":"MyOtherValue", "D":"A", "timestamp": 123456788}
ArrayDecompositionMapper
The ArrayDecompositionMapper is a function that allows us to map an array to a specific dimensions:
{
"name":"arrayDecompositionMapper",
"className":"io.wizzie.normalizer.funcs.impl.ArrayDecompositionMapper",
"properties": {
"dimensionToBuild": ["dim1", "dim2", "dim3"],
"dimension": "array-dim",
"delete_dimension": true
}
}
The ArrayDecompositionMapper has three properties:
dimension
: The dimension that indicate the array that we want to map.dimensionToBuild
: The output dimension that the mapper generates using the dimension array.delete_dimension
: Option to delete the dimension or not.
If we have this json message:
{
"dimension":"VALUE-1",
"otherDimension":"VALUE-2",
"array-dim": ["A", "B", "C"]
}
If we use this message using the ArrayDecompositionMapper that is defined on the above example, we get this output:
{
"dimension":"VALUE-1",
"otherDimension":"VALUE-2",
"dim1":"A",
"dim2":"B",
"dim3":"C"
}
RegexMapper
The RegexMapper is a function that allows us to extract new dimension from String using regex rules.
{
"name": "myRegexMapper",
"className": "io.wizzie.normalizer.funcs.impl.RegexMapper",
"properties": {
"generateDimensions":["dim1", "dim2", "dim3"],
"regexPattern": "test test (?<dim1>.*) test (?<dim2>.*) test (?<dim3>.*) test",
"parseDimension": "message" }
}
The RegexMapper has three properties:
generateDimensions
: The new dimensions generated from the StringregexPattern
: The regex pattern to extract the dimensions. You must use groups(...)
and group names.?<...>
parseDimension
: The string that is checked with the regexPattern.
If we have this json message:
{
"timestamp": 123456789,
"message": "test test hello1 test hello2 test hello3 test"
}
If we use this message using the RegexMapper that is defined on the above example, we get this output:
{
"timestamp": 123456789,
"message": "test test hello1 test hello2 test hello3 test"
"dim1":"hello1",
"dim2":"hello2",
"dim3":"hello3"
}
UpperCaseStringMapper
The UpperCaseStringMapper is a function that allow us to apply a upperCase over a string dimension.
{
"name":"myUpperCaseStringMapper",
"className":"io.wizzie.normalizer.funcs.impl.UpperCaseStringMapper",
"properties": {
"dimension": "dim1"
}
}
The UpperCaseStringMapper have one property named dimension
:
dimension
: Dimension to upper case.
I we have next json message:
{
"type": "measures",
"timestamp": 123456789,
"dim1": "1213aaBBcc"
}
If we use this message using the UpperCaseStringMapper that is defined on the above example, we get next output:
{
"type": "measures",
"timestamp": 123456789,
"dim1": "1213AABBCC"
}
LowerCaseStringMapper
The UpperCaseStringMapper is a function that allow us to apply a upperCase over a string dimension.
{
"name":"myLowerCaseStringMapper",
"className":"io.wizzie.normalizer.funcs.impl.LowerCaseStringMapper",
"properties": {
"dimension": "dim1"
}
}
The UpperCaseStringMapper have one property named dimension
:
dimension
: Dimension to lower case
I we have next json message:
{
"type": "measures",
"timestamp": 123456789,
"dim1": "1213aaBBcc"
}
If we use this message using the UpperCaseStringMapper that is defined on the above example, we get next output:
{
"type": "measures",
"timestamp": 123456789,
"dim1": "1213aabbcc"
}
MapFlattenMapper
The MapFlattenMapper is a function that allows us to transform Map<String, Map<String, Object>>
into an Array<Map<String, Object>>
{
"name":"myMapFlattenMapper",
"className":"io.wizzie.normalizer.funcs.impl.MapFlattenMapper",
"properties": {
"key_dimension": "key_dim",
"flat_dimension": "data",
"output_dimension": "array_data"
}
}
The MapFlattenMapper have 3 properties:
flat_dimension
: TheMap<String, Map<String, Object>>
to transform.key_dimension
: The new key dimension name to store the key value.output_dimension
: The new output dimension to store theArray<Map<String, Object>>
I we have next json message:
{
"data":{
"A":{"dim":"AA"},
"C":{"dim":"BB"}
}
}
If we use this message using the MapFlattenMapper that is defined on the above example, we get next output:
{
"array_data":[
{"key_dim":"A", "dim": "AAA"},
{"key_dim":"C", "dim": "BB"}
]
}
DecimalPrecisionMapper
The DecimalPrecisionMapper is a function that allows us to adjust the precision of the number. This functions always rounds up the number field.
{
"name":"myDecimalPrecisionMapper",
"className":"io.wizzie.normalizer.funcs.impl.DecimalPrecisionMapper",
"properties": {
"scales": [
{
"dimensions": ["latitude", "longitude"],
"numberOfDecimals": 3
},
{
"dimensions": ["pi-number"]
}
]
}
}
The DecimalPrecisionMapper have 2 properties:
dimensions
: Dimensions that contains the values to adjust.numberOfDecimals
: The precision of decimals. By default 2.
I we have next json message:
{
"latitude": 37.123151,
"longitude": 79.001589,
"pi-number": 3.14159265358979323846
}
If we use this message using the DecimalPrecisionMapper that is defined on the above example, we get next output:
{
"latitude": 37.123,
"longitude": 79.002,
"pi-number": 3.14
}