Group by age groups in mongo db - php

Question

I am using following mongo db query . and it is showing me only age groups of my personas data.
db.amplifyindex.aggregate([
{ $unwind: "$demographic" },
{ $match : { 'demographic.is_latest':"active",
'demographic.date_of_birth' : { $exists : true} } },
{ $project : {"ageInMillis" : {$subtract : [new Date(),
"$demographic.date_of_birth"] } } },
{ $project : {"age" : {$divide : ["$ageInMillis", 31558464000] }}},
{ $project : {"age" : {$subtract : ["$age", {$mod : ["$age",1]}]}}},
])
The result object by this query is:
{
"_id" : ObjectId("58a42cbbdb5d880c1e000029"),
"age" : 29.0
}
I want to get data in a way as :
{
"_id" : ObjectId("58a42cbbdb5d880c1e000029"),
"age" : 10-30
"personas" : 10
}
Complete data structure in my scenario is given as :
{
"_id" : ObjectId("58a42cbbdb5d880c1e000029"),
"persona_email" : "milton.ullrich#gmail.com",
"company_id" : "1",
"date_added" : ISODate("2017-02-15T10:23:15.000Z"),
"demographic" : [
{
"persona_fname" : "Hayden",
"middle_name" : "Jacobs",
"persona_lname" : "Schmeler",
"gender" : "male",
"date_of_birth" : ISODate("1987-06-16T19:00:00.000Z"),
"marital_status" : "single",
"height" : "2.1 feet",
"weight" : "5 kg",
"measurement" : {
"waist" : "34 inch"
},
"disabilities" : "No",
"race" : "Asian",
"nationality" : "Jordan",
"life_cycle" : "empty nest",
"children" : {
"gender" : "female",
"date_of_birth" : ISODate("1987-06-16T19:00:00.000Z")
},
"medicare_no" : "4916725587565",
"driving_licence_no" : "PACBALM420M",
"id_no" : "4485872783336",
"passport_no" : "OCRFHCNR",
"residential_address" : {
"unit_no" : 603,
"street_no" : "34863 Ondricka Viaduct Apt. 154",
"street_name" : "84799 Little Wall",
"suburb" : "West Virginia",
"postal_code" : "23873",
"state" : "Dominica",
"country" : "Cook Islands"
},
"work_address" : {
"unit_no" : "6011865161287875",
"street_no" : "630 Beer Underpass Suite 372",
"street_name" : "87672 Lind Burg",
"suburb" : "West Virginia",
"postal_code" : "84356-3662",
"state" : "Qatar",
"country" : "British Indian Ocean Territory (Chagos Archipelago)"
},
"shipping_address" : {
"unit_no" : "6011865161287875",
"street_no" : "8292 Langosh Drive Suite 065",
"street_name" : "9844 Nicolas Mount",
"suburb" : "West Virginia",
"postal_code" : "07014",
"state" : "Togo",
"country" : "Tunisia"
},
"job_title" : "Geoscientists",
"employer_name" : "Aufderhar Group",
"income" : 72577,
"phone_numbers" : {
"work_phone" : "993-783-7499",
"home_phone" : "701.546.7016",
"mobile" : "1-346-729-4392"
},
"emails" : {
"work_emails" : "nikolaus.asa#gmail.com",
"personal_emails" : "emmitt68#gmail.com"
},
"languages" : "italian",
"data_source" : "soldi",
"date_added" : ISODate("2017-02-15T10:23:15.000Z"),
"source" : "soldi",
"is_latest" : "inactive"
},
{
"persona_fname" : "Julien",
"middle_name" : "Hirthe",
"persona_lname" : "Schaefer",
"gender" : "female",
"date_of_birth" : ISODate("1987-06-16T19:00:00.000Z"),
"marital_status" : "single",
"height" : "2.1 feet",
"weight" : "5 kg",
"measurement" : {
"waist" : "34 inch"
},
"disabilities" : "No",
"race" : "Asian",
"nationality" : "Sudan",
"life_cycle" : "single",
"children" : {
"gender" : "female",
"date_of_birth" : ISODate("1987-06-16T19:00:00.000Z")
},
"medicare_no" : "4024007131689860",
"driving_licence_no" : "OGOSUOIFSDN",
"id_no" : "5523297913341227",
"passport_no" : "WWZYTE489ZR",
"residential_address" : {
"unit_no" : 603,
"street_no" : "338 Mueller Gardens Suite 397",
"street_name" : "6302 Catalina Isle",
"suburb" : "West Virginia",
"postal_code" : "97534",
"state" : "Antarctica (the territory South of 60 deg S)",
"country" : "Turks and Caicos Islands"
},
"work_address" : {
"unit_no" : "6011865161287875",
"street_no" : "8561 Jesus Ridges Apt. 662",
"street_name" : "1869 Josiah Wall Apt. 347",
"suburb" : "West Virginia",
"postal_code" : "14810",
"state" : "Zambia",
"country" : "Mongolia"
},
"shipping_address" : {
"unit_no" : "6011865161287875",
"street_no" : "514 Heller Center",
"street_name" : "835 Paxton Cliffs Suite 040",
"suburb" : "West Virginia",
"postal_code" : "48942-3845",
"state" : "Palau",
"country" : "Belize"
},
"job_title" : "Geoscientists",
"employer_name" : "Aufderhar Group",
"income" : 72577,
"phone_numbers" : {
"work_phone" : "[phone] x573",
"home_phone" : "796.308.3001 x88799",
"mobile" : "[phone]"
},
"emails" : {
"work_emails" : "prohaska.katelynn#hotmail.com",
"personal_emails" : "ozulauf#jenkins.com"
},
"languages" : "german",
"data_source" : "soldi",
"date_added" : ISODate("2017-02-15T10:23:15.000Z"),
"source" : "soldi",
"is_latest" : "active"
}
]
}

Answer 1

Within your group pipeline, create the age ranges as part of the _id key and this can be done through the use of the $concat and "$cond" operators. Consider running the following pipeline to get the desired result:
db.amplifyindex.aggregate([
{ "$unwind": "$demographic" },
{ "$match": { "demographic.is_latest": "active" } },
{
"$project": {
"age": {
"$divide": [
{
"$subtract": [
new Date(),
{ "$ifNull": ["$demographic.date_of_birth", new Date()] }
]
},
1000 * 86400 * 365
]
}
}
},
{
"$group": {
"_id": {
"$concat": [
{ "$cond": [ { "$lte": [ "$age", 0 ] }, "Unknown", ""] },
{ "$cond": [ { "$and": [ { "$gt": ["$age", 0 ] }, { "$lt": ["$age", 10] } ]}, "Under 10", ""] },
{ "$cond": [ { "$and": [ { "$gte": ["$age", 10] }, { "$lt": ["$age", 31] } ]}, "10 - 30", ""] },
{ "$cond": [ { "$and": [ { "$gte": ["$age", 31] }, { "$lt": ["$age", 51] } ]}, "31 - 50", ""] },
{ "$cond": [ { "$and": [ { "$gte": ["$age", 51] }, { "$lt": ["$age", 71] } ]}, "51 - 70", ""] },
{ "$cond": [ { "$gte": [ "$age", 71 ] }, "Over 70", ""] }
]
},
"personas": { "$sum": 1 }
}
},
{ "$project": { "_id": 0, "age": "$_id", "personas": 1 } }
])
In the above pipeline, I've taken out the query for null date fields and replaced it with the $ifNull operator within the $project pipeline. This will return the current date where the birthday field is null and thus yields an age of 0 years which will fall into a special age range bracket, thus the need to include all the documents, including the ones where the date of birth field missing.
In the case where the $concat operator is used for all $cond expressions, the combination is crucial as it acts as a case statement where otherwise there will be nested $cond operators.
To understand this design, take the third $cond expression
{
"$cond": [
{
"$and": [
{ "$gte": ["$age", 10] },
{ "$lt": ["$age", 31] }
]
},
"10 - 30", // matching if
"" // else
]
},
which essentially expresses the logic
if ( "$age" >= 10 && "$age" < 31 ) { return "10 - 30"; }
else return "";
$concat is useful when used with the $cond operators as the expression will return an empty value except the matching one, which in turn will return the age group.
For example, an age of 30.645411 would return
"$concat": [ "", "", "10 - 30", "", "", "" ]
which then gives you an _id with value of "10 - 30".