Python Code/Notebook for getting metrics from Azure Trusted Signing using Diagnostic Settings exported to Azure Blob Storage (Codesign Transactions)

azure-codesign-metrics.ipynb

azure-codesign-metrics.py

# -*- coding: utf-8 -*-
"""azure-codesign-metrics.ipynb

Automatically generated by Colab.

Original file is located at
    <REDACTED>
# Azure Codesigning Metrics

Because Microsoft is a bit slow with actually implementing a way to query the metrics of Azure Trusted Signing accounts, I had to make this Python notebook to do it for us. What they do expose is a way to let Diagnostic Settings log towards a Storage Accounts' Blob Container. This notebook uses a connection string to the Azure Storage Account and downloads all blobs in the container and measures their frequency based on the month.

We can expand this functionality to retrieve other data columns from the entries in the blobs (each instance of codesign).

## Install dependencies
"""

!pip install azure-storage-blob pandas

"""## Connect to Azure Blob Storage"""

from azure.storage.blob import BlobServiceClient
from google.colab import userdata

# Connection string from Azure Storage
CONNECTION_STRING = userdata.get('connection_string')

# Create a blob service client
blob_service_client = BlobServiceClient.from_connection_string(CONNECTION_STRING)

# Set container name
CONTAINER_NAME = userdata.get('container_name')

# Connect to the container
container_client = blob_service_client.get_container_client(CONTAINER_NAME)

# List blobs (files) in the container
blobs = [blob.name for blob in container_client.list_blobs()]

# Show the first 10 blob names
blobs[:10]

"""## Download and Read a File"""

import json

# Pick the first blob file
blob_name = blobs[0]  # Change index if needed

# Get blob client and download file
blob_client = container_client.get_blob_client(blob_name)
blob_data = blob_client.download_blob().readall()

# Decode JSONL (JSON lines) file
lines = blob_data.decode("utf-8").split("\n")
sample_entries = [json.loads(line) for line in lines if line.strip()]

# Print first few entries
sample_entries[:3]

"""## Count Transactions per Month"""

import re
from collections import Counter

def count_transactions(blobs):
    transaction_counts = Counter()

    for blob_name in blobs:
        # Extract the year and month from the blob path
        match = re.search(r"y=(\d{4})/m=(\d{2})", blob_name)
        if not match:
            continue  # Skip if no match is found

        year, month = match.groups()
        month_key = f"{year}-{month}"

        # Download and read blob data
        blob_client = container_client.get_blob_client(blob_name)
        blob_data = blob_client.download_blob().readall()

        # Count number of lines (each line is a transaction)
        num_transactions = sum(1 for line in blob_data.decode("utf-8").split("\n") if line.strip())
        transaction_counts[month_key] += num_transactions

    return transaction_counts

transaction_counts = count_transactions(blobs)

print(transaction_counts)

"""## Visualization"""

import pandas as pd
import matplotlib.pyplot as plt

# Convert to DataFrame
df = pd.DataFrame(transaction_counts.items(), columns=["Month", "TransactionCount"])
df = df.sort_values("Month")

# Plot
plt.figure(figsize=(10,5))
plt.bar(df["Month"], df["TransactionCount"], color="blue")
plt.xlabel("Month")
plt.ylabel("Transaction Count")
plt.title("Transactions per Month")
plt.xticks(rotation=45)
plt.show()