I have the following document:
[
{
"_id": "ObjectId(6448cae9aa015ee5fc8892a8)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": [],
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892ab)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": [
{
"start": 1019,
"end": 1033,
"text": "F10-749-113245",
"tag": "ACCOUNT_NO",
"color": "rgb(244, 10, 10, 0.527)"
}
],
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": "BL"
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892ae)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b1)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": [
{
"start": 65,
"end": 81,
"text": "Nowhere BANK LTD ",
"tag": "DRAWER",
"color": "rgb(85, 239, 239, 0.527)"
}
],
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": "APPLICATION"
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b4)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": [],
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": "APPLICATION"
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b7)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892bf)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c2)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c5)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c8)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
},
"doc_type_class_labels": ""
},
"category": "TEXT"
}
]
I ran the following query to get my required data:
{
'$unwind': {
'path': '$annotations.ner_labels.value',
"preserveNullAndEmptyArrays": True
}
},
I get the following output:
[
{
"_id": "ObjectId(6448cae9aa015ee5fc8892a8)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892ab)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": {
"start": 1019,
"end": 1033,
"text": "F10-749-113245",
"tag": "ACCOUNT_NO",
"color": "rgb(244, 10, 10, 0.527)"
},
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": "BL"
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892ae)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b1)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": {
"start": 65,
"end": 81,
"text": "nowhere BANK LTD ",
"tag": "DRAWER",
"color": "rgb(85, 239, 239, 0.527)"
},
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": "APPLICATION"
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b4)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"tag": "ACCOUNT_NO"
},
"doc_type_class_labels": "APPLICATION"
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b7)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892bf)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c2)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c5)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c8)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
}
]
The preserveNullAndEmptyArrays is creating an empty object. Now i need to add a temporary value inside the empty object. So I used the following code:
{
'$project': {
'annotations.ner_labels.value': {
'$ifNull': [
'$annotations.ner_labels.value', [],
]
},
'category': 1,
'job_id': 1
}
},
After this the entire object is modified instead if the particular field, the following is the output:
[
{
"_id": "ObjectId(6448cae9aa015ee5fc8892a8)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892ab)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": {
"start": 1019,
"end": 1033,
"text": "F10-749-113245",
"tag": "ACCOUNT_NO",
"color": "rgb(244, 10, 10, 0.527)"
}
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892ae)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b1)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": {
"start": 65,
"end": 81,
"text": "nowhere BANK LTD ",
"tag": "DRAWER",
"color": "rgb(85, 239, 239, 0.527)"
}
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b4)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cae9aa015ee5fc8892b7)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892bf)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c2)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c5)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
},
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c8)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
}
]
The following
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c2)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {},
"doc_type_class_labels": ""
},
"category": "TEXT"
}
is changed to the following after ifnull
{
"_id": "ObjectId(6448cbfdaa015ee5fc8892c2)",
"job_id": "ObjectId(6448cac2aa015ee5fc8892a0)",
"annotations": {
"ner_labels": {
"value": []
}
},
"category": "TEXT"
}
The field doc_type_class_labels is missing. How to change only the missing field instead of the entire object?
Kindly advise. Thank you.
you can use $addFields instead of $project since project will not show the fields that you do not include.
The $addFields stage is equivalent to a $project stage that explicitly specifies all existing fields in the input documents and adds the new fields.
db.collection.aggregate([
{
$unwind: {
path: "$annotations.ner_labels.value",
preserveNullAndEmptyArrays: true
}
},
{
$addFields: {
"annotations.ner_labels.value": {
$ifNull: [
"$annotations.ner_labels.value",
[]
]
}
}
}
])