LPP dev guide¶

Kick Starter for efficient scientific coding¶

You will find here material to help you get started with your coding. Depending on whether you mostly do data analysis or numerical modeling, you'll need slightly different tools and methods. the advice we give you will help to be efficient, rigorous and to write code than you'll be able to use, maintain and share in the long run. Now remember, If you're a PhD or a post-doc, following these advice will not only help you improving the quality and reproducibility of your science, but also will make all your coding efforts reusable by people in the lab once you're gone, and more importantly will give you the basic knowledge you need to legitimately claim for a data science / computing science position in the private sector.

• The commandments of programming
• Common tools and methodology you'll need to get started.

Now, what are you doing?

• You're mostly coding for data analysis
• You're mostly coding for numerical modeling

Code review and analysis¶

Code review¶

• Code Review for Teams Too Busy to Review Code (youtube video)
• Nice tutorial on code review with Rhodecode

Code analysis¶

• CPP Check

Performance¶

Videos

• Modern C++

Optimizations

• Instruction tables

Setting up a clean Python environment¶

• Python and virtualenv

C++ development¶

• The C++ language
• Design Pattern in general and in C++
• C++ Gurus

Code Design and Architecture¶

• The S.O.L.I.D. principles

Writing code¶

• Code review
• Useful resources

Documentation¶

• [[hyb-par: Documentationtools| Documentation Tools]]