I have a this code:
import logging
import json
import os
from azure.cosmos import CosmosClient
import azure.functions as func
url = os.environ["ACCOUNT_URI"]
key = os.environ["ACCOUNT_KEY"]
client1 = CosmosClient(url, key)
client = CosmosClient.from_connection_string(os.environ["CosmosDBConnStr"])
database_name = ""
container_name = ""
database = client.get_database_client(database_name)
container = database.get_container_client(container_name)
logging.info(f"container: {url}")
def main(myblob: func.InputStream, doc: func.Out[func.Document]):
logging.info(f"Python blob trigger function processed blob \n"
f"Name: {myblob.name}\n")
#reading file from blob
contract_data=myblob.read()
try:
logging.info(f"contract data: {contract_data}")
contract_json = json.loads(contract_data)
version = contract_json.get("version")
name = contract_json.get("name")
title = contract_json.get("title")
logging.info(f"contract json: {contract_json}")
query = "SELECT c.version,c.name,c.title,c.Theme,c.description,c['data owner'],c.confidentiality,c.table1 FROM c "
items = list(container.query_items(
query=query,
enable_cross_partition_query=True
))
logging.info(f"item: {items[0]}")
for item in items:
if item["name"] == name and item["version"] == version:
if item["title"] == title:
logging.info(f"Skipping, item already exists: {item}")
return # Skip saving the document
container.upsert_item(body=contract_json,pre_trigger_include =
None,post_trigger_include= None)
return
doc.set(func.Document.from_json(contract_data))
except Exception as e:
logging.info(f"Error: {e}")
I added one document in my cosmosdb, I would like to replace same document has different title. but i could not do that i am getting Code: BadRequest Message: Message: {"Errors":["One of the specified inputs is invalid"]} this is an example of item from my query:
{'version': 'V1', 'name': 'demo_contract2', 'title': 'title122', 'Theme': 'Theme12', 'description': 'test data contract management2', 'data owner': '[email protected]', 'confidentiality': 'open', 'table1': {'description:': 'testen', 'attribute_1': {'type': 'int', 'description:': 'testen', 'identifiability': 'identifiable'}}}
this is an example of contract_json from my file:
{'version': 'V1', 'name': 'demo_contrac1', 'title': 'title1', 'Theme': 'Theme1', 'description': 'test data contract management2', 'data owner': '[email protected]', 'confidentiality': 'open', 'table1': {'description:': 'testen', 'attribute_1': {'type': 'int', 'description:': 'testen', 'identifiability': 'identifiable'}}}
they are matching. How should I regulate my upsert_item function in my code?
You are missing the id
in the Upsert content. A document's identity is defined by the combination of id
and Partition Key value, if you don't specify the id
then Upsert will always behave as a Create operation.
Because you are getting the items through a query, just add the id
:
query = "SELECT c.id, c.version,c.name,c.title,c.Theme,c.description,c['data owner'],c.confidentiality,c.table1 FROM c "
You can now get the id
from item["id"]
to use if needed. The body of the Upsert operation should contain id
, and if that matches with an existing document, then the document will get updated.