Shulou(Shulou.com)06/02 Report--

It is believed that many inexperienced people are at a loss to analyze the process of compiling Python source code. Therefore, this paper summarizes the causes and solutions of the problem. Through this article, I hope you can solve this problem.

The re module provided in the Python language can support regular expressions and also provides a SGML,XML analysis module. Most developers use Python source code to develop and run XML programs. Here we share it with you.

There are a lot of similarities, so in this order, Python's GC chapter, I'm going to focus more on implementation and my questions, Java's GC chapter, more on usage. Python takes the route of combining a variety of GC technology routes, which I think has its merits.

First of all, Python uses the original Ref Counting technology, and for circular references that can not be solved by reference counting, the Python source code also uses Mark-Sweeping to GC. This seems to have two benefits, a large amount of memory recycling. Apportioned to the reference count.

Reduce the burden of the Mark process, will not cause the program to pause, but can really eliminate the real memory leakage caused by circular references. PyObject_GC_New will call _ PyObject_GC_Malloc, where the return value of the former.

It focuses on the object itself, while the latter focuses on memory. In fact, on a newly allocated piece of memory, the object has the following relationship with the memory it is locked in: from the process of object creation, Python has the following key C implementation functions and structures:

Typedef union _ gc_head {

Struct {

Union _ gc_head * gc_next

Union _ gc_head * gc_prev

Py_ssize_t gc_refs

} gc

Long double dummy; / * force worst-case alignment * /

} PyGC_Head

In fact, I am a little disappointed with this structure because it takes up too many resources to recover a piece of memory. Maybe I'm too stingy. I think 8 bytes may be just right. To be honest, I have a preliminary idea in my mind that an object's management memory requires only 8 bytes, and the whole GC process does not need to be copied and compressed.

When I look at the code, I do not know if I do not understand some skills, or LOCK did not achieve, I feel that Python malloc and free put a pair of useless LOCK and UNLOCK, [Python 2.5.2], I do not know if it is because I do not have the actual debugging reasons, have not found the mystery of this macro.

To be honest, I've been fighting memory leaks for years, and this time I've learned a lot from it (and from other sources). Combined with the memory pool used in the Ref class I've written, I've constructed a new memory pool, hoping to have a chance to use my talents.

Note:

[1] I haven't examined the original Python source code, but I remember that the original Python introduced garbage collection only through the reference counting mechanism, especially Ruby 1.9, which used to use reference counting technology.

[2] it almost forced me to look at the source code of JVM, but on a 64-bit platform, this structure could change even more.

[3] wait until I have finished the code before I can honor this sentence, and then I will Open Source.

After reading the above, have you mastered the method of dissecting the Python source code programming process? If you want to learn more skills or want to know more about it, you are welcome to follow the industry information channel, thank you for reading!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.