I have below scenario, where two operations need to be performed. One is parsing the list and create a comma separated string. Then, transform that into the output format json
Input -
{
"list": ["ABC","XYZ"]
}
Output -
{
"additionalAttributes" : {
"userContext" : [ {
"auths" : "ABC,XYZ"
} ]
}
}
Check this spec
[
{
"operation": "modify-overwrite-beta",
"spec": {
"list": "=join(',',#(1,list))"
}
}, {
"operation": "shift",
"spec": {
"list": "additionalAttributes.userContext[].auths"
}
}
]
Related
How do I transform below JSON using JOLT
Input will be array of strings
This string will be binary number , 0 or 1.
I want to perform binary addition and then split resultant array as split and every element should be a seperate property
{
"binaryarray": [
"0100",
"0010"
]
}
Output
{
"field1" : 0,
"field2" : 1,
"field3" : 1,
"field4" : 0
}
I was able to partially achieve it
Convert array of string to array of int
perform Addition of array
Convert above to string
Split above string
[
{
"operation": "modify-overwrite-beta",
"spec": {
"binaryarray": ["=toInteger", 0]
}
},
{
"operation": "modify-overwrite-beta",
"spec": {
//
// Sums
"sumIntData": "=intSum(#(1,binaryarray))"
}
},
{
"operation": "modify-overwrite-beta",
"spec": {
"sumIntData": ["=toString", 0]
}
},
{
"operation": "modify-overwrite-beta",
"spec": {
"PID3": "=split('', #(1,sumIntData))"
}
}
]
[
{
"dst_cnt": "149125"
},
{
"src_cnt": "149136"
},
{
"TABLENAME": "NAME"
}
]
I want to subtract dst_cnt and src_cnt from this data with NIfi jolt. Is data operation possible after type conversion in NIfi?
Yes, it's possible. You can try the following transformatios spec
[
{
// combine individual attributes within the common object
"operation": "shift",
"spec": {
"*": {
"*": "&"
}
}
},
{
// sum up the the integer values after determining the negative form of "dst_cnt"
"operation": "modify-overwrite-beta",
"spec": {
"dst_cnt_": "=divide(#(1,dst_cnt),-1)",
"cnt_dif": "=intSum(#(1,dst_cnt_),#(1,src_cnt))"
}
},
{
// only pick the result derived from the subtraction
"operation": "shift",
"spec": {
"cnt_*": "&"
}
}
]
i have the following input json:
{
"tags": {
"event": "observation",
"source": "hunter"
}
}
The output JSON should look like below:
{
"tags" : [ "event:observation", "source:hunter" ]
}
can anyone provide any guidance on how to build a proper jolt specification for the above?
thank you very much for the help ^_^
You can use this specification
[
{ // combine each key-value pair under within common arrays
"operation": "shift",
"spec": {
"tags": {
"*": {
"$": "&2_&1",
"#": "&2_&1"
}
}
}
},
{ // concatenate key-value pairs by colon characters
"operation": "modify-overwrite-beta",
"spec": {
"*": "=join(':',#(1,&))"
}
},
{
"operation": "shift",
"spec": { // make array key common("tags") for all arrays
// through use of _ seperator and * wildcard
"*_*": "&(0,1)"
}
}
]
the demo on the site http://jolt-demo.appspot.com/ is
I am trying to write a jolt transformation for below input -
{
"restaurantId": "ZZ4ORJDY3E",
"chainId": "a-b"
}
expected output is -
{
"ZZ4ORJDY3E" : {
"key" : "ZZ4ORJDY3E",
"start" : "a",
"end" : "b"
}
}
My spec is -
[
{
"operation": "shift",
"spec": {
"#restaurantId": "#restaurantId.key",
"chainId": {
"*-*": {
"$(0,1)": "#restaurantId.start",
"$(0,2)": "#restaurantId.end"
}
}
}
}
]
The spec is not transforming as expected output. i want learn how to use attributes inside string parser.
Spec
[
{
"operation": "shift",
"spec": {
"restaurantId": {
// match any value of restaurantId
"*": {
// write the value to of the key $ to the output
// where the output is the "value of the key".key
// kinda hokey
"$": "&.key"
}
},
"chainId": {
"*-*": {
// write each part of the chainId to the output
// at the value of restaurantId from back up the tree
"$(0,1)": "#(3,restaurantId).start",
"$(0,2)": "#(3,restaurantId).end"
}
}
}
}
]
Here is the JSON input:
{
"myRootKey": {
"directMove": "directValue",
"marker": "THE_MARKER",
"someTextField": "someString",
"someObject": {
"someKey": "value"
}
}
}
the output should be:
{
"myRootKey": {
"subKey": {
"directMove": "directValue"
},
"THE_MARKER": {
"someTextField": "someString",
"someObject": {
"someKey": "value"
}
}
}
}
With direct moving it is clear, but how rest of the input to the marker object value?
You match down to "someTextField" and "someObject", but use the new "#" / look up the tree logic to find the "marker" to use as an ouput path.
Spec
[
{
"operation": "shift",
"spec": {
"myRootKey": {
"directMove": "myRootKey.subKey.directValue",
"someTextField": "#(1,marker).someTextField",
"someObject": "#(1,marker).someObject"
}
}
}
]
#(1,marker) allows you to retrieve the value of the marker field
&1 retrieves the value of the matching node
So the spec you are looking for looks like :
[
{
"operation": "shift",
"spec": {
"myRootKey": {
"directMove": "myRootKey.subKey.directValue",
"someTextField": "&1.#(1,marker).someTextField",
"someObject": "&1.#(1,marker).someObject"
}
}
}
]
You can use this spec fully dynamically:
[
{
"operation": "shift",
"spec": {
"*": {
"directMove": "&1.subKey.&",
"*TextField": "&1.#(1,marker).&",
"*Object": "&1.#(1,marker).&"
}
}
}
]