I have complex (EDI) input that looks like this:
{
"INS-2000_loop": [
{
"INS_05": "A",
"INS_06": "D",
"INS_08": "AC"
},
{
"REF_01": "0F",
"REF_02": "111111111"
},
{
"NM1-2100A_loop": [
{
"NM1_01": "IL",
"NM1_02": "1",
"NM1_03": "aaa",
"NM1_04": "bbb"
},
{
"PER_03": "HP",
"PER_04": "1234567890"
}
]
},
{
"HD-2300_loop": [
{
"HD_01": "001",
"HD_03": "HMO",
"HD_04": "485-014-000",
"HD_05": "IND"
},
{
"DTP_01": "348",
"DTP_02": "D8",
"DTP_03": "20240401"
}
]
},
{
"HD-2300_loop": [
{
"HD_01": "001",
"HD_03": "HMO",
"HD_04": "485-014-000",
"HD_05": "IND"
},
{
"DTP_01": "348",
"DTP_02": "D8",
"DTP_03": "20240101"
},
{
"DTP_01": "349",
"DTP_02": "D8",
"DTP_03": "20240331"
}
]
}
]
}
The input can have any number of HD-2300_loop elements (typically less than 5) and only one INS-2000_loop element. I want the HD-2300_loop elements to "drive" the output so that output is an array of complex objects of the same length as the number of HD-2300_loop elements. The complex objects of the output array need to have all the data in the HD-2300_loop elements and the data of all the children of the INS-2000_loop element except the HD-2300_loop looking like this:
[
{
"L2300": [
{
"HD_01": "001",
"HD_03": "HMO",
"HD_04": "485-014-000",
"HD_05": "IND"
},
{
"DTP_01": "348",
"DTP_02": "D8",
"DTP_03": "20240101"
},
{
"DTP_01": "349",
"DTP_02": "D8",
"DTP_03": "20240331"
}
],
"L2000": [
{
"INS_05": "A",
"INS_06": "D",
"INS_08": "AC"
},
{
"REF_01": "0F",
"REF_02": "111111111"
},
{
"NM1-2100A_loop": [
{
"NM1_01": "IL",
"NM1_02": "1",
"NM1_03": "aaa",
"NM1_04": "bbb"
},
{
"PER_03": "HP",
"PER_04": "1234567890"
}
]
}
]
},
...
]
I've started with the following spec which isn't giving what I want - I haven't figured out how to inject the @3 selection into the result of the @1 selection. Also, I haven't figure out how to rename the selection results (to L2300 and L2000) and how to skip the HD-2300_loop child in the @3 selection result.
[
{
"operation": "shift",
"spec": {
"*": {
"*": {
"HD-2300_loop": {
"@1": "",
"@3": ""
}
}
}
}
}
]
I would also greatly appreciate if anyone could point me a Jolt doc or cheat sheet that explains how the & operators work. Thanks very much.
Here's the final solution I came up with (includes some transformations not mentioned in the original problem):
[
{
"operation": "shift",
"spec": {
"INS-2000_loop": {
"*": {
"INS*": "L2000.&",
"NM1-2100A_loop": {
"*": {
"D*|L*|N*": "L2000.L2100A.&",
"PER_04": {
"@": "L2000.L2100A.PER[].value",
"@(1,PER_03)": "L2000.L2100A.PER[#2].type"
},
"PER_06": {
"@": "L2000.L2100A.PER[].value",
"@(1,PER_05)": "L2000.L2100A.PER[#2].type"
},
"PER_08": {
"@": "L2000.L2100A.PER[].value",
"@(1,PER_07)": "L2000.L2100A.PER[#2].type"
}
}
},
"NM1-2100G_loop": {
"*": {
"N*": "L2000.L2100G.&",
"PER_04": {
"@": "L2000.L2100G.PER[].value",
"@(1,PER_03)": "L2000.L2100G.PER[#2].type"
},
"PER_06": {
"@": "L2000.L2100G.PER[].value",
"@(1,PER_05)": "L2000.L2100G.PER[#2].type"
},
"PER_08": {
"@": "L2000.L2100G.PER[].value",
"@(1,PER_07)": "L2000.L2100G.PER[#2].type"
}
}
},
"REF_01": {
"*": {
"@(2,REF_02)": "L2000.@(2)"
}
},
"HD-2300_loop": {
"*": {
"HD*": "L2300s.[&3].L2300.&",
"DTP_01": {
"*": {
"@(2,DTP_03)": "L2300s.[&5].L2300.@(2)"
}
},
"LX-2310_loop": {
"*": {
"N*": "L2300s.[&5].L2300.L2310.&",
"PER_04": {
"@": "L2300s.[&6].L2300.L2310.PER[].value",
"@(1,PER_03)": "L2300s.[&6].L2300.L2310.PER[#2].type"
},
"PER_06": {
"@": "L2300s.[&6].L2300.L2310.PER[].value",
"@(1,PER_05)": "L2300s.[&6].L2300.L2310.PER[#2].type"
},
"PER_08": {
"@": "L2300s.[&6].L2300.L2310.PER[].value",
"@(1,PER_07)": "L2300s.[&6].L2300.L2310.PER[#2].type"
}
}
},
"COB-2320_loop": {
"*": {
"COB*": "L2300s.[&5].L2300.L2320[&5].&",
"DTP_01": {
"*": {
"@(2,DTP_03)": "L2300s.[&7].L2300.L2320[&7].@(2)"
}
},
"NM1-2330_loop": {
"*": {
"N*": "L2300s.[&7].L2300.L2320[&7].L2330.&"
}
}
}
}
}
}
}
}
}
},
{
"operation": "shift",
"spec": {
"L2000": "&",
"L2300s": {
"*": { "L2300": "L2300s[]" }
}
}
},
{
"operation": "shift",
"spec": {
"L2300s": {
"*": {
"@(3,L2000)": "[&].L2000",
"@": "[&].L2300"
}
}
}
}
]