Z's Blog

More documentation:Memory Model and Default Constructor Construction.

How to delineate the most important 20% in an unfamiliar field or book? How to identify trivial details?
I hadn’t deeply contemplated these methodologies in my previous reading and studying, thinking they were merely empty talks. However, I now feel that mastering a suitable technique is like having a sharp blade in hand, allowing one to navigate obstacles with ease.

Spent a month reading TC++PL4E. Since I had previously read C++ Primer and understood most of the C++ syntax, the reading speed was quite fast. However, by reading alongside the C++ standard, I discovered many things in C++ that I previously didn’t know. From the perspective of the standard and the father of C++, C++ is indeed comprehensive enough. Thus, I’ll compare C++ Primer, TC++PL4E, and the ISO C++ documentation.

POD is Plain Old Data. In C++, it refers to objects that can be treated as “data only,” where programmers do not need to worry about the complexity of class layout, as well as user-defined constructors, copy, and move semantics.

ADL is the abbreviation for Argument-dependent lookup, which in Chinese translates to 参数依赖查找. ADL is useful for avoiding verbose code, but it can also cause some ambiguity.

最近在读The C++ Programming Language Fourth Edition(简称TC++PL4E)，由于我入的实体书是中译本(原版好贵)，所以我是对照着英文版PDF看的，发现了一些原版和中译本中的勘误，还有一些我觉得书中歧义的地方，查阅标准(ISO/IEC 14882:2014(E))之后的定义也一并贴出，汇总在这里列出来。

There are countless resources on C/C++ available online, but I often find that, after reading some articles, it seems the authors themselves don’t fully understand what they wrote (looking back at my own earlier articles, my perspective was too one-sided and superficial at that time). Therefore, for C/C++ materials, I believe it’s essential to refer directly to the standard documentation, as standards do not introduce ambiguities. One should not blindly search the internet and trust second-hand digested materials.

I think consulting these four documents is enough for understanding the features of the C/C++ languages (click to preview online or download):

• ISO/IEC 9899:1999 (E) (C99 standard)
• The C Programming Language Second Edition (The major work of C language creators Dennis Ritchie and Brian Kernighan)
• ISO/IEC 14882:2014(E) (C++14 standard)
• The C++ Programming Language Fourth Edition (Written by the father of C++, based on the C++11 standard)

The reason the C language standard does not follow the latest C11 standard is that the current C++ standard (C++14) references ISO/IEC 9899:1999 in its Normative references, which means TCPL and TC++PL can serve as applicable descriptions for the C/C++ standards and can corroborate each other.

For more about C++ Normative references, refer to ISO/IEC 14882:2014(E) §1.2 Normative references.

I will gradually extract some commonly ambiguous language features from the standard norms here to ensure that what I write is backed by the standard documentation.

CppQuiz is a simple online quiz that you can use to test your knowledge of the C++ programming language.
It’s quite interesting; I wrote down a few questions I encountered today. I’ll add more here whenever I have time. In fact, many questions in CppQuiz can be explained by “Deep Exploration of C++ Object Model”… If you find yourself struggling with many questions, I recommend buying a copy of “Deep Exploration of C++ Object Model” and reading it thoroughly!
In addition, I will try to find relevant descriptions in the C++ standard (ISO/IEC 14882:2014) while answering questions.

This blog has switched to Github Action for automated deployment.

Using Hexo to generate a static blog and then deploying it to Github/Coding Pages is currently how my blog is hosted. It’s quite enjoyable to use, but there are several issues:

1. The environment configuration for Hexo is too cumbersome; it’s almost impossible to update the blog when changing to a new computer.
2. Each time I modify an article, I have to regenerate everything.
3. Additionally, the number of command executions for submitting the blog source files is too high (mainly the first step).

I wasted a lot of time submitting articles, so I optimized the process to almost achieve fully automated submissions.

Recently, I’ve been reading the Boost code and writing some analyses on the usage and implementation of useful modules in the Boost libraries, with updates to come periodically.