PBU Log Inspector

Flask (pip) and React (npm) integration for python-basic-utils logs.

Installation

pip3 install pbu-log-inspector
npm install pbu-log-inspector

Integration

Backend

from flask import Flask
from loginspect import register_endpoint

app = Flask(__name__)
register_endpoint(app)
app.run(host="0.0.0.0", port=5444)

The register_endpoint function will register an endpoint GET /api/_logs, which will can be used to serve daily logs. There are additional parameters available for this function:

register_endpoint(app, log_folder="_logs", api_prefix="/api", log_file_mapping=None, login_check=None)

• The default log_folder is relative to the start script of your application, since that is the pbu default.
• If you provide a different api_prefix this will just be the prefix, not the /_logs part. E.g. /rest would result in an endpoint /rest/_logs
• The log_file_mapping is optional and only necessary, if you instructed pbu to write to different file names than the defaults. When provided, the mapping needs to provide a dictionary with keys representing the default pbu log file names (info.log, debug.log, warning.log, error.log) and their respective values are the name of the log file in your app.
• If your application includes authentication, provide a function via the login_check parameter that performs the authentication. No parameters are passed to the function. You can use from flask import request and access all request data

Restrictions

• The date needs to be the first part of a log message
• The date needs to be in the format %Y-%m-%d %H:%M:%S.%s, e.g. 2019-12-25 13:37:01.567
• There is no limit on how many logs the endpoint delivers. This can easily cause large payloads and potentially exceed operational limits (e.g. browser performance, response size, ...)

Frontend

import React from "react"
import LogInspectorContainer from "pbu-log-inspector"

const MyContainer = props => (
    <div>
        <LogInspectorContainer />
    </div>
)

The container provides the following properties:

• apiPrefix - default "/api" - corresponds to the api_prefix parameter configured on the Flask backend
• applyRequestOptions - default null - a function that takes the base request parameters and has to return the parameters for the fetch request, in case authentication is required and headers need to be set
• errorHandler - default null - a function to handle an error during fetching logs (e.g. show an alert), no return is expected. The parameter is the error thrown.