Files
hpos-data/scripts/combine_users_testdata.py

56 lines
1.6 KiB
Python
Raw Normal View History

2023-12-10 13:54:26 +05:30
import os
import firebase_admin
from firebase_admin import credentials
from firebase_admin import firestore
from google.cloud.firestore_v1.base_query import FieldFilter
import pandas as pd
import datetime
import sys
cred = credentials.Certificate(os.getcwd() + '/' + 'keys/hpos-prod-firebase-adminsdk.json')
firebase_admin.initialize_app(cred)
db = firestore.client()
patient_collection = db.collection("patientData")
test_collection = db.collection("testData")
start_date = sys.argv[1] # '2023-01-12'
end_date = sys.argv[2] #'2023-07-16'
query = patient_collection.where(filter=FieldFilter("createdAt", ">=", start_date)).where(filter=FieldFilter("createdAt", "<", end_date))
patient_docs = query.stream()
data = []
for patient_doc in patient_docs:
patient_data = patient_doc.to_dict()
patient_id = patient_data["_id"]
test_docs = test_collection.where("_id", "==", patient_id).stream()
for test_doc in test_docs:
test_data = test_doc.to_dict()
combined_data = {**patient_data, **test_data}
for key, value in combined_data.items():
if value == "" and key in patient_data:
combined_data[key] = patient_data[key]
data.append(combined_data)
data.append(patient_data)
df = pd.DataFrame(data)
print(df)
print(df.size)
output_filename = "users_test_data.csv"
df.to_csv(output_filename, index=False)
downloads_dir = os.path.join(os.path.expanduser("~"), "Downloads")
output_path = os.path.join(downloads_dir, output_filename)
df.to_csv(output_path, index=False)
print(f"Data saved to '{output_path}'")
firebase_admin.delete_app(firebase_admin.get_app())