You can't have the foundation with words, and you can see the common things clearly only by keeping in mind the familiarity. If you are still worried about the return value of a basic function parameter, then you have to memorize more basic knowledge and there is no other way.

Second, learn to find wheels instead of making them.

In fact, this view is controversial. If you don't build wheels, you need to use some molding library functions and APIs. Don't be exclusive, write everything yourself. In this case, you can't get rid of low taste. For example, message queuing is implemented in c++. Do you write a message queue for a message that only stores a few string messages?

Third, find a good IDE.

Get eclipse for java development, vs for window development and vim for linux development. A good IDE can make you get twice the result with half the effort. What's more, in China, most of them are ctrl+c and CTRL+V, a good IDE, with automatic completion, function jump search, automatic typesetting and error detection. When you decide to use an ide, you should seriously learn its shortcut keys and a lot of usage methods.

Fourth, think again.

A good piece of code is not as complex as possible. Simply implementing complex functions is what we need most. Although the code of liunx kernel is very large, the code realized by Jin Dian algorithm can't be refined any more. Jade Bird of Changping Peking University suggested that it is more important to simplify the carding program flow and improve the coding speed to complete the function with the minimum code amount. So the most important thing in programming is thinking.