I am trying to validate an inserted document against a schema, and was trying to find a way to validate the inserted document.
There are libraries like MongoEngine that say they do the work, but is there a way to do document validation directly via pymongo ?
The python driver docs are indeed a little light on how to use the db.command. Here is a complete working example:
from pymongo import MongoClient
from collections import OrderedDict
import sys
client = MongoClient() # supply connection args as appropriate
db = client.testX
db.myColl.drop()
db.create_collection("myColl") # Force create!
# $jsonSchema expression type is prefered. New since v3.6 (2017):
vexpr = {"$jsonSchema":
{
"bsonType": "object",
"required": [ "name", "year", "major", "gpa" ],
"properties": {
"name": {
"bsonType": "string",
"description": "must be a string and is required"
},
"gender": {
"bsonType": "string",
"description": "must be a string and is not required"
},
"year": {
"bsonType": "int",
"minimum": 2017,
"maximum": 3017,
"exclusiveMaximum": False,
"description": "must be an integer in [ 2017, 3017 ] and is required"
},
"major": {
"enum": [ "Math", "English", "Computer Science", "History", None ],
"description": "can only be one of the enum values and is required"
},
"gpa": {
# In case you might want to allow doubles OR int, then add
# "int" to the bsonType array below:
"bsonType": [ "double" ],
"minimum": 0,
"description": "must be a double and is required"
}
}
}
}
# Per the docs, args to command() require that the first kev/value pair
# be the command string and its principal argument, followed by other
# arguments. There are two ways to do this: Using an OrderDict:
cmd = OrderedDict([('collMod', 'myColl'),
('validator', vexpr),
('validationLevel', 'moderate')]
db.command(cmd)
# Or, use the kwargs construct:
# db.command('collMod','myColl', validator=vexpr, validationLevel='moderate')
try:
db.myColl.insert({"x":1})
print "NOT good; the insert above should have failed."
except:
print "OK. Expected exception:", sys.exc_info()
try:
okdoc = {"name":"buzz", "year":2019, "major":"Math", "gpa":3.8}
db.myColl.insert(okdoc)
print "All good."
except:
print "exc:", sys.exc_info()
MongoDB supports document validation at the engine level so you'll pick it up via pymongo. You declare your "schema" (rules actually) to the engine. Here's a great place to start: https://docs.mongodb.com/manual/core/document-validation/
You can make a separated JSON file for your Document Validations Schema, like this:
{
"collMod": "users",
"validator": {
"$jsonSchema": {
"bsonType": "object",
"required": ["email", "password","name"],
"properties": {
"email": {
"bsonType": "string",
"description": "Correo Electrónico"
},
"password": {
"bsonType": "string",
"description": "Una representación Hash de la contraseña"
},
"name": {
"bsonType": "object",
"required": ["first", "last"],
"description": "Objeto que separa los nombres y apellidos",
"properties": {
"first": {
"bsonType": "string",
"description": "Primer y segundo nombre"
},
"last": {
"bsonType": "string",
"description": "Primer y segundo apellido"
}
}
},
}
}
}
}
Then you can use in python script, example:
from pymongo import MongoClient
import json #parse JSON file as dict
from collections import OrderedDict #preserve the order (key, value) in the gived insertions on the dict
client = MongoClient("your_mongo_uri")
db = client.your_db_name
with open('your_schema_file.json', 'r') as j:
d = json.loads(j.read())
d = OrderedDict(d)
db.command(d)
I know 2 options to deal with:
import pymongo
mongo_client = MongoClient(url=...,
port=...,
username=...,
password=...,
authSource=...,
authMechanism=...,
connect=True, )
mongo_client.server_info()
db = mongo_client.your_db
users = db.create_collection(name="users",
validator={"$jsonSchema": {
"bsonType": "object",
"required": ["username"],
"properties": {
"username": {
"bsonType": "string",
"pattern": "[a-z0-9]{5,15}",
"description": "user name (required), only lowercase letters "
"and digits allowed, from 5 to 15 characters long"
},
"email": {
"bsonType": "string",
"description": "User's email (optional)"
},
}
}},
validationAction="error",
)
# Inserting user document that fits the scheme
users.insert_one({"username": "admin", "email": "some_admin_mail"})
# Insertion below will be rejected with "pymongo.errors.WriteError: Document failed validation, full error"
# caused by too short username (root)
users.insert_one({"username": "root", "email": "some_root_mail"})
from jsonschema import validate
from jsonschema.exceptions import ValidationError
db = MongoClient(...).your_db
schema = {
"type": "object",
"required": ["username"],
"properties": {
"username": {"type": "string", "pattern": "[a-z0-9]{5,15}"},
"email": {"type": "string"},
},
}
try:
new_user = {"username": "admin", "email": "some_admin_mail"}
# No exception will be raised in validation below
validate(instance=new_user, schema=schema)
db.users.insert_one(new_user)
new_user = {"username": "root", "email": "some_root_mail"}
# Exception <ValidationError: 'root' does not match '[a-z0-9]{5,15}'> will be raised
validate(instance=new_user, schema=schema)
db.users.insert_one(new_user)
except ValidationError:
# Performing error
