MONGODB Aggregate query to give $min value of array field by element - mongodb

Question

Hi I am converting my existing website from php / mysql to node / mongodb , it is a golf society site which I use to log scores for each members rounds and provide results and statistics into their games. The main collection contains an array of 18 scores which is great for my results queries but I am having a problem with the statistical side, ie Avg score by Hole by Course, Lowest score ever by Hole by Course (eclectic) . I have come up with this aggregate query which works and gives me the result I require but it is ugly !! I am sure there must be a more elegant solution out there and feel I am missing a trick somewhere, I have looked at $map as I thought that might help but don't think it will. I would appreciate it if someone could offer any suggestions in tidying this code up, Thx.
{ _id:
{ date_played: 2019-06-21T00:00:00.000Z,
course_played: 1,
player_id: 1 },
score: [ 8, 4, 7, 4, 7, 1, 7, 5, 6, 4, 5, 7, 6, 4, 7, 5, 6, 7 ],
handicap: 23,
cash_won: 0,
sort_order: 2,
gross_score: 100,
gross_sfpts: 31,
skins_group: 1,
score_differential: 26.2,
pcc_adjustment: 0 }
{ _id:
{ date_played: 2016-08-14T00:00:00.000Z,
course_played: 1,
player_id: 1},
score: [ 5, 4, 5, 6, 5, 4, 8, 6, 1, 3, 3, 4, 3, 6, 3, 6, 4, 5 ],
handicap: 18,
cash_won: 14,
sort_order: 4,
gross_score: 81,
gross_sfpts: 44,
skins_group: 1,
score_differential: 12.1,
pcc_adjustment: 0 }
[
{
'$match': {
'_id.course_played': 1
}
}, {
'$project': {
'player_name': 1,
'hole01': {
'$arrayElemAt': [
'$score', 0
]
},
'hole02': {
'$arrayElemAt': [
'$score', 1
]
},
'hole03': {
'$arrayElemAt': [
'$score', 2
]
},
'hole04': {
'$arrayElemAt': [
'$score', 3
]
},
'hole05': {
'$arrayElemAt': [
'$score', 4
]
},
'hole06': {
'$arrayElemAt': [
'$score', 5
]
},
'hole07': {
'$arrayElemAt': [
'$score', 6
]
},
'hole08': {
'$arrayElemAt': [
'$score', 7
]
},
'hole09': {
'$arrayElemAt': [
'$score', 8
]
},
'hole10': {
'$arrayElemAt': [
'$score', 9
]
},
'hole11': {
'$arrayElemAt': [
'$score', 10
]
},
'hole12': {
'$arrayElemAt': [
'$score', 11
]
},
'hole13': {
'$arrayElemAt': [
'$score', 12
]
},
'hole14': {
'$arrayElemAt': [
'$score', 13
]
},
'hole15': {
'$arrayElemAt': [
'$score', 14
]
},
'hole16': {
'$arrayElemAt': [
'$score', 15
]
},
'hole17': {
'$arrayElemAt': [
'$score', 16
]
},
'hole18': {
'$arrayElemAt': [
'$score', 17
]
}
}
}, {
'$sort': {
'_id.player_id': 1
}
}, {
'$group': {
'_id': '$_id.player_id',
'name': {
'$first': '$player_name'
},
'hole1': {
'$min': '$hole01'
},
'hole2': {
'$min': '$hole02'
},
'hole3': {
'$min': '$hole03'
},
'hole4': {
'$min': '$hole04'
},
'hole5': {
'$min': '$hole05'
},
'hole6': {
'$min': '$hole06'
},
'hole7': {
'$min': '$hole07'
},
'hole8': {
'$min': '$hole08'
},
'hole9': {
'$min': '$hole09'
},
'hole10': {
'$min': '$hole10'
},
'hole11': {
'$min': '$hole11'
},
'hole12': {
'$min': '$hole12'
},
'hole13': {
'$min': '$hole13'
},
'hole14': {
'$min': '$hole14'
},
'hole15': {
'$min': '$hole15'
},
'hole16': {
'$min': '$hole16'
},
'hole17': {
'$min': '$hole17'
},
'hole18': {
'$min': '$hole18'
},
'rounds': {
'$sum': 1
}
}
}, {
'$addFields': {
'total': {
'$add': [
'$hole1', '$hole2', '$hole3', '$hole4', '$hole5', '$hole6', '$hole7', '$hole8', '$hole9', '$hole10', '$hole11', '$hole12', '$hole13', '$hole14', '$hole15', '$hole16', '$hole17', '$hole18'
]
}
}
}, {
'$sort': {
'total': 1
}
}, {
'$limit': 10
}
]
Which gives this as an example when run against the total database, which is the result I want but I would like all the "hole" fields to be returned in an Array as per the original score field.
{ _id: 1,
hole1: 5,
hole2: 4,
hole3: 5,
hole4: 4,
hole5: 5,
hole6: 2,
hole7: 3,
hole8: 3,
hole9: 3,
hole10: 3,
hole11: 2,
hole12: 3,
hole13: 4,
hole14: 2,
hole15: 3,
hole16: 3,
hole17: 3,
hole18: 3,
rounds: 562,
total: 53 }

Answer 1

You might $unwind the scores array, keeping the index as the hole number, then $group by player, course, and hole to get the score for each hole, $sort by hole number to make sure of the order, and then $group by player and course, pushing the scores back into an array.
db.collection.aggregate([
{$match: {"_id.player_id": 1}},
{$unwind: {
path: "$score",
includeArrayIndex: "hole"
}},
{$group: {
_id: {
course_played: "$_id.course_played",
player_id: "$_id.player_id",
hole: "$hole"
},
minScore: {$min: "$score"},
rounds: {$sum: 1}
}},
{$sort: {"_id.hole": 1}},
{$group: {
_id: {
course_played: "$_id.course_played",
player_id: "$_id.player_id"
},
score: {$push: "$minScore"},
total: {$sum: "$minScore"},
rounds: {$first: "$rounds"}
}}
])
Playground