Python Trending Weekly
ISSUE 67

uv's Major Update

2024-09-01 Season 3

TL;DR

uv's major release aims to unify Python packaging, PortaOne shows how PyPy connects over a billion calls a month, and one 'farewell pandas' post stirs the pot.

Articles 12 Projects 12
🦄

Articles & Tutorials

× 12
01 uv: Unified Python packaging astral.sh

uv just dropped another major update! Not only as a pip replacement, but now it's expanded into an end-to-end solution for Python project management, command-line tools, single-file scripts, and Python itself. (Also: UV Updates and PEP 723: Simplifying Python Packaging and Scripting)

02 Differences between PyJWT and python-jose in handling JWT tokens cnblogs.com

PyJWT and python-jose are Python libraries for handling JWT, but they have some important differences in functionality scope and design philosophy. The article introduces their differences and some examples of using FastAPI + python-jose to handle JWT in projects.

03 Kindle + Python = e-Ink Dashboard (part 1) blog.4dcu.be

How to install Python on a Kindle? The article introduces a hardware jailbreaking method that successfully uses KUAL to install Python on a Kindle Paperwhite 3, runs scripts on it, and successfully creates an e-ink dashboard.

04 Farewell pandas, and thanks for all the fish ibis-project.org

Ibis is an open-source data analysis and visualization library that will remove pandas and dask backends starting from version 10.0, switching to recommend the DuckDB backend instead. The article explains why they initially chose pandas and the pain points pandas brought. (Note: ibis means a type of wading bird)

05 CPython Compiler Hardening nohlson.com

The author joined the Google Summer of Code program. This post reviews what he did while hardening the CPython compiler, including enabling compiler options, developing tools to track warnings, plans to resolve existing warnings, and lessons learned.

06 Achieving High Performance with Code Generation medium.com

Dynamic code generation, or codegen for short, is a technique for generating new code at runtime. The article covers when you should use it, a real example that doubled the speed of a production system, and wrong vs. right ways to use it.

07 How PortaOne uses PyPy for high-performance processing, connecting over 1B of phone calls every month pypy.org

PortaOne is a company serving telecom operators, using Python to develop their core middleware. The article briefly introduces its functionality implementation, using PyPy to improve performance, and major bugs discovered during usage.

08 10 Python programming optimisation techniques dev.to

Introduces 10 methods for optimizing Python code to write more efficient and high-performance code, such as variable packing, using more memory to reduce disk I/O, fixed-length vs. variable-length variables, internal vs. public functions, etc.

09 NTLM Credential Theft in Python Windows Applications horizon3.ai

NTLMv2 hash theft is a common credential harvesting technique. The article discloses SSRF vulnerabilities in 3 popular Python libraries (Gradio, Jupyter Notebook, and Streamlit) that could lead to credential leakage, with detailed analysis of the code causing these vulnerabilities.

10 Flask Source Code Analysis Series - 7 Articles cizixs.com

A Flask source code analysis series published in 2017, analyzing version 0.12. Topics covered include application startup flow, routing, context, requests, responses, and sessions, helpful for deep understanding of Flask. (Shared by @un4gt) (Also: the sharer's Flask project scaffolding tool flasky-cli)

11 What's new in pip 24.2 — or why legacy editable installs are deprecated ichard26.github.io

A pip core team member introduces notable changes in version 24.2, including deprecating editable installs, defaulting to system HTTPS certificates, multiple performance optimizations, stricter pip check rules, and configuring pip to only work in activated virtual environments.

12 Following up "Mother of all htmx demos" david.guillot.me

HTMX is appearing more and more frequently in the Python community - we've shared quite a bit in previous issues. This author played a key role in HTMX's popularity. The author reviews their experience with and promotion of HTMX, the impact it had, and answers some questions people raised.

htmx GitHub Stars Trend
🐿️

Projects & Resources

× 12
01 theine: high performance in-memory cache github.com

A high-performance cache library inspired by Caffeine, with a Rust core, using W-TinyLFU and Clock-Pro eviction policies for high hit rates, and hierarchical timers for automatic expiration of expired data.

02 picows: Ultra-fast websocket client and server for asyncio github.com

A library implemented with Cython, designed for building asyncio WebSocket clients and servers, with performance nearly 2x faster than aiohttp.

03 microrabbit: A lightweight, asynchronous Python framework for RabbitMQ github.com

A lightweight RabbitMQ handling framework that simplifies building microservices and distributed systems, with features including simple message routing, plugin support, and intuitive client configuration.

04 wave: Realtime Web Apps and Dashboards for Python and R github.com

Build beautiful, low-latency, real-time, browser-based apps and dashboards in pure Python or R, without HTML, Javascript, or CSS. (3.9K stars)

wave: Realtime Web Apps and Dashboards for Python and R
05 repo2vec: Chat with your codebase with 2 commands github.com

A simple and easy-to-use library, similar to GitHub Copilot, that lets you chat with any public or private code repository.

06 django-currentuser: Conveniently store reference to request user on thread/db level github.com

Conveniently manage references to the current request user, supporting both thread and database storage methods.

07 intelligent-trading-bot: Automatically generating signals and trading based on machine learning and feature engineering github.com

Automatically trade cryptocurrencies using state-of-the-art machine learning algorithms and feature engineering, analyzing historical data and training models in batch offline mode.

08 rerun: Visualize streams of multimodal data github.com

Record multimodal data like images, tensors, point clouds, and text, with real-time visualization, useful for robotics, simulation, computer vision, and other scenarios involving lots of sensors or time-varying signals. (6.2K stars)

09 VideoSys: An easy and efficient system for video generation github.com

Provides user-friendly and high-performance infrastructure for video generation, supporting the entire pipeline of training, inference, serving, and compression. (1.6K stars)

10 KnowledgeGraph: Building Knowledge Graphs from Scratch github.com

This repository collects papers/blog materials on topics like foundational knowledge, construction theory, and practical implementation for building knowledge graphs, from theory to implementation. (1K stars)

11 terminaltexteffects: a terminal visual effects engine, application, and Python library github.com

A terminal visual effects engine that can be installed as a system program to generate effects in the terminal, or as a Python library to enable effects in Python scripts/applications. (2.7K stars)

12 kotaemon: An open-source RAG-based tool for chatting with your documents github.com

A minimalist, customizable RAG UI (based on Gradio) for chatting with documents, with support for viewing how the RAG pipeline runs. (5.8K stars)

kotaemon: An open-source RAG-based tool for chatting with your documents
ESC
↑↓ select · ↵ open175 issues