Problems with rule-based mapping in Azure Dataflow - azure

Question

I have multiple XMLs in an Azure Blob. I want to extract certain datapoints and sort them into an Azure SQL db.
For this I use Data Flow. The name of certain elements changes sometimes [random name] so i would like to set up a rule-based mapping, that fetches the right values every time .
I want to retrieve IMPORTANT INFORMATION, which is always located in the first randomly named child of category_a.
Apart from the randomly named object the rest of the structure always stays the same.
This is about the structure:
<title>
<category_a>
<random_name_1>
<object_a>
<subobject_object_a>
<p>IMPORTANT INFORMATION</p>
</subobject_object_a>
</object_a>
</random_name_1>
<random_name_2>
<object_a>
<subobject_object_a>
<p>IRRELEVANT INFORMATION</p>
</subobject_object_a>
</object_a>
</random_name_2>
</category_a>
<category_b></category_b>
How do I need to write the rule based mapping so that I always fetch this value no matter the random name in the middle of the path?
Thanks for your help

Answer 1

There might be no option to find or use rule-based mapping to retrieve the important information. As a work around, I have used lookup and string manipulation activities to get the result. The following is the pipeline JSON:
{
"name": "pipeline1",
"properties": {
"activities": [
{
"name": "Lookup1",
"type": "Lookup",
"dependsOn": [],
"policy": {
"timeout": "0.12:00:00",
"retry": 0,
"retryIntervalInSeconds": 30,
"secureOutput": false,
"secureInput": false
},
"userProperties": [],
"typeProperties": {
"source": {
"type": "XmlSource",
"storeSettings": {
"type": "AzureBlobFSReadSettings",
"recursive": true,
"enablePartitionDiscovery": false
},
"formatSettings": {
"type": "XmlReadSettings",
"validationMode": "none",
"namespaces": true
}
},
"dataset": {
"referenceName": "Xml1",
"type": "DatasetReference"
},
"firstRowOnly": false
}
},
{
"name": "Set variable1",
"type": "SetVariable",
"dependsOn": [
{
"activity": "Lookup1",
"dependencyConditions": [
"Succeeded"
]
}
],
"userProperties": [],
"typeProperties": {
"variableName": "tp",
"value": {
"value": "#replace(replace(split(string(activity('Lookup1').output.value[0].title.category_a),':')[0],'\"',''),'{','')",
"type": "Expression"
}
}
},
{
"name": "Set variable2",
"type": "SetVariable",
"dependsOn": [
{
"activity": "Set variable1",
"dependencyConditions": [
"Succeeded"
]
}
],
"userProperties": [],
"typeProperties": {
"variableName": "tp2",
"value": {
"value": "#replace(replace(split(string(activity('Lookup1').output.value[0].title.category_a[variables('tp')]),':')[0],'\"',''),'{','')",
"type": "Expression"
}
}
},
{
"name": "Set variable3",
"type": "SetVariable",
"dependsOn": [
{
"activity": "Set variable2",
"dependencyConditions": [
"Succeeded"
]
}
],
"userProperties": [],
"typeProperties": {
"variableName": "tp3",
"value": {
"value": "#replace(replace(split(string(activity('Lookup1').output.value[0].title.category_a[variables('tp')][variables('tp2')]),':')[0],'\"',''),'{','')",
"type": "Expression"
}
}
},
{
"name": "Set variable4",
"type": "SetVariable",
"dependsOn": [
{
"activity": "Set variable3",
"dependencyConditions": [
"Succeeded"
]
}
],
"userProperties": [],
"typeProperties": {
"variableName": "tp4",
"value": {
"value": "#replace(replace(split(string(activity('Lookup1').output.value[0].title.category_a[variables('tp')][variables('tp2')][variables('tp3')]),':')[0],'\"',''),'{','')",
"type": "Expression"
}
}
},
{
"name": "Set variable5",
"type": "SetVariable",
"dependsOn": [
{
"activity": "Set variable4",
"dependencyConditions": [
"Succeeded"
]
}
],
"userProperties": [],
"typeProperties": {
"variableName": "final",
"value": {
"value": "#activity('Lookup1').output.value[0].title.category_a[variables('tp')][variables('tp2')][variables('tp3')][variables('tp4')]",
"type": "Expression"
}
}
}
],
"variables": {
"tp": {
"type": "String"
},
"tp2": {
"type": "String"
},
"tp3": {
"type": "String"
},
"tp4": {
"type": "String"
},
"final": {
"type": "String"
}
},
"annotations": []
}
}
The lookup has the source xml file. Since we know the category_a is a child element, I have started from there to obtain the child element names (in each set variable activity).
The following is the output image for reference: