mongodb shell to java driver aggregate statement - driver

Question

I'm able to run an aggregate MongoDb (3.8) request in the mongo shell but I can't get it out working using the java driver (3.4 java 8).
here's the input source structure. I want to aggregate informations to provide company activities stats. note that the ape document states for the company registered activity. I want to group on the first 3 digits of the ape code
{
"_id" : "00552017600016",
"sirene" : "005520176",
"nic" : "00016",
"natureCode" : "5710",
"natureLabel" : "SAS, société par actions simplifiée",
"name" : "HERNAS CARTONNAGE",
"brand" : "",
"registerDate" : ISODate("1954-12-31T23:00:00Z"),
"startDate" : ISODate("1954-12-31T23:00:00Z"),
"headquaters" : true,
"address" : {
"number" : 50,
"ext" : null,
"lane" : "RUE PASTEUR",
"zipcode" : "80210",
"city" : "FEUQUIERES EN VIMEU",
"country" : "FRANCE",
"location" : {
"type" : "Point",
"coordinates" : [
1.60319694244078,
50.0537636973407
],
"accuracy" : 0,
"ban" : "ADRNIVX_0000000260779642"
}
},
"ape" : {
"code" : "1721A",
"label" : "Fabrication de carton ondulé"
},
"staff" : {
"count" : 50,
"year" : 2016
}
}
running this commands in the shell works like a charm
db.companies_geo.aggregate([ {$match:{"address.city":"LE MANS"}},
{$group:{_id:{$substrBytes:["$ape.code",0,2]}, count:{$sum:1}, code:{$last:"$ape.code"}, label:{$last:"$ape.label"}}},
{$project:{count:1, code:1, label:1}},
{$sort:{count:-1}},
{$limit:20}
])
it produces this result
{ "_id" : "68", "count" : 4603, "code" : "6832B", "label" : "Supports juridiques de gestion de patrimoine immobilier" }
{ "_id" : "47", "count" : 1929, "code" : "4799B", "label" : "Vente par automate, aut. com. dét. hors mag., éventaire ou marché" }
{ "_id" : "94", "count" : 1874, "code" : "9499Z", "label" : "Autres organisations fonctionnant par adhésion volontaire" }
{ "_id" : "86", "count" : 1704, "code" : "8690F", "label" : "Activités de santé humaine non classées ailleurs" }
{ "_id" : "56", "count" : 832, "code" : "5630Z", "label" : "Débits de boissons" }
{ "_id" : "90", "count" : 816, "code" : "9004Z", "label" : "Gestion de salles de spectacles" }
{ "_id" : "85", "count" : 769, "code" : "8560Z", "label" : "Activités de soutien à l'enseignement" }
{ "_id" : "70", "count" : 592, "code" : "7022Z", "label" : "Conseil pour les affaires et autres conseils de gestion" }
{ "_id" : "96", "count" : 585, "code" : "9609Z", "label" : "Autres services personnels n.c.a." }
{ "_id" : "81", "count" : 582, "code" : "8130Z", "label" : "Services d'aménagement paysager" }
on the java side I've tried this but the code line bellow the while iteration raises the exception org.bson.codecs.configuration.CodecConfigurationException: Can't find a codec for class com.mongodb.client.model.BsonField.
private List<Document> cityActivityGraph() {
Bson filter = Filters.eq("address.city", "LE MANS");
BsonField id = new BsonField("_id", Filters.eq(new Document("$substrBytes", Arrays.asList("$ape.code", 0, 2))));
BsonField count = Accumulators.sum("count", 1);
BsonField label = Accumulators.last("ape.label", 1);
BsonField code = Accumulators.last("ape.code", 1);
Bson group = Aggregates.group(id, count, code, label);
Bson match = Aggregates.match(filter);
Bson projections = Projections.include("code", "label", "count");
Bson project = Aggregates.project(projections);
Bson sort = Sorts.descending("count");
MongoCursor<Document> cursor = getTable().aggregate(Arrays.asList(
Aggregates.match(match),
Aggregates.group(group),
Aggregates.project(project),
Aggregates.sort(sort))).iterator(); <--- exception here
List<Document> result = new ArrayList<>();
while (cursor.hasNext()) {
Document doc = cursor.next();
System.out.println(doc.toJson());
result.add(doc);
}
return result;
}
any idea ? thanks in advance

Answer 1

found a solution, perhaps not the most efficient but it works
Bson id = new Document("_id", new Document("$substrBytes", Arrays.asList("$ape.code", 0, 3)));
BsonField code = Accumulators.last("code", "$ape.code");
BsonField label = Accumulators.last("label", "$ape.label");
BsonField count = Accumulators.sum("count", 1);
Bson projections = Projections.include("code", "label", "count");
Bson sort = Sorts.descending("count");
AggregateIterable<Document> stats = getTable().aggregate(Arrays.asList(
Aggregates.match(filter),
Aggregates.group(id, code, label, count),
Aggregates.project(projections),
Aggregates.limit(limit),
Aggregates.sort(sort)
));
List<Document> result = new ArrayList<>();
for (Document d:stats) {
result.add(d);
}
return result;