search
HomeBackend DevelopmentPython Tutorial使用Protocol Buffers的C语言拓展提速Python程序的示例

 Protocol Buffers (类似XML的一种数据描述语言)最新版本2.3里,protoc—py_out命令只生成原生的Python代码。 尽管PB(Protocol Buffers)可以为C++语言生成快速解析和序列化代码,但是这种方式对于Python不适用,并且手动生成的已包装的代码需要非常大的维护工作。在讨论组里,这是一个常见的功能要求,由于一个必备的客户端组件—AppEngine(根据团队介绍名称为AppEngine),生成原生的Python代码有更高的优先级。

幸运的是, PB 2.4版本中本地化代码已被提名,在 svn的分支中已经可以下载,因此你能够使用快速的 PB有一段时间了。 (我们使用 r352版本有一段时间了,还没有遇到任何问题。) PB团队一直不愿轻易指定任何发布日期,在我的威胁下, Kenton Varda提到日期初步定在 2011年初。


我没有在其它地方看见过这个文档,希望它能对其他人有所帮助.

如何做能让它快起来

安装好新的PB库之后并使用 protoc --py_out=...  重新构建好你的PB之后,你需要在运行你的Python程序之前进行环境变量 PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=cpp 的设置,以便于选择C++的,或者PB默认使用的Python实现.

就这样了!这至少就能在可以动态转化/序列化消息的PB运行时库用通用的C++代码了. (注意我们还没有生成任何C++代码.)

它能有多快呢? 我编写了一个简单的程序来获得性能在我们的应用程序中的提升感观:
 

nruns = 1000nwarmups = 100xs = ... # your protobufsdef ser(): return [x.SerializeToString() for x in xs]def parse(ys): for y in ys: pb.Email().ParseFromString(y)
 
t = timeit.Timer(lambda:None)
t.timeit(nwarmups)print 'noop:', t.timeit(nruns) / nruns
 
t = timeit.Timer(ser)
t.timeit(nwarmups)print 'ser:', t.timeit(nruns) / nruns / len(xs)
 
ys = ser()
t = timeit.Timer(lambda: parse(ys))
t.timeit(nwarmups)print 'parse:', t.timeit(nruns) / nruns / len(xs)print 'msg size:', sum(len(y) for y in ys) / len(ys)

以秒为单位,这段程序在我的桌面上给出了如下几个时间结果:
 

$ python sandbox/pbbench.py out.ini
ser: 0.000434461673101
parse: 0.000602062404156
msg size: 10730
 
$ PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=cpp \
> python sandbox/pbbench.py out.ini
ser: 2.86788344383e-05
parse: 7.63910810153e-05
msg size: 10730

这显示出在序列化和转化方面分别有15和8被的速度提升。不坏!但还可以更快.

如何做让它更快

现在我们实际上只是特地针对你的PB生成了一个C++实现,而我们从来没有使用过运行时反射。首先,为你的Python项目添加一个C扩展,不如,通过修改如下的 setup.py:
 

setup(
  ...
  ext_modules=[Extension('podpb',
sources=['cpp/podpb.c','cpp/main.pb.cc'], libraries=['protobuf'])],
  ...
  )

使用 protoc --cpp_out=cpp 生成main.pb.c, 并按如下所示创建 podpb.c 来设置一个空的 Python C 模块:
 

#include <python.h>
 
static PyMethodDef PodMethods[] = {
 {NULL, NULL, 0, NULL}    /* Sentinel *
    </python.h>
Statement
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Python's Execution Model: Compiled, Interpreted, or Both?Python's Execution Model: Compiled, Interpreted, or Both?May 10, 2025 am 12:04 AM

Pythonisbothcompiledandinterpreted.WhenyourunaPythonscript,itisfirstcompiledintobytecode,whichisthenexecutedbythePythonVirtualMachine(PVM).Thishybridapproachallowsforplatform-independentcodebutcanbeslowerthannativemachinecodeexecution.

Is Python executed line by line?Is Python executed line by line?May 10, 2025 am 12:03 AM

Python is not strictly line-by-line execution, but is optimized and conditional execution based on the interpreter mechanism. The interpreter converts the code to bytecode, executed by the PVM, and may precompile constant expressions or optimize loops. Understanding these mechanisms helps optimize code and improve efficiency.

What are the alternatives to concatenate two lists in Python?What are the alternatives to concatenate two lists in Python?May 09, 2025 am 12:16 AM

There are many methods to connect two lists in Python: 1. Use operators, which are simple but inefficient in large lists; 2. Use extend method, which is efficient but will modify the original list; 3. Use the = operator, which is both efficient and readable; 4. Use itertools.chain function, which is memory efficient but requires additional import; 5. Use list parsing, which is elegant but may be too complex. The selection method should be based on the code context and requirements.

Python: Efficient Ways to Merge Two ListsPython: Efficient Ways to Merge Two ListsMay 09, 2025 am 12:15 AM

There are many ways to merge Python lists: 1. Use operators, which are simple but not memory efficient for large lists; 2. Use extend method, which is efficient but will modify the original list; 3. Use itertools.chain, which is suitable for large data sets; 4. Use * operator, merge small to medium-sized lists in one line of code; 5. Use numpy.concatenate, which is suitable for large data sets and scenarios with high performance requirements; 6. Use append method, which is suitable for small lists but is inefficient. When selecting a method, you need to consider the list size and application scenarios.

Compiled vs Interpreted Languages: pros and consCompiled vs Interpreted Languages: pros and consMay 09, 2025 am 12:06 AM

Compiledlanguagesofferspeedandsecurity,whileinterpretedlanguagesprovideeaseofuseandportability.1)CompiledlanguageslikeC arefasterandsecurebuthavelongerdevelopmentcyclesandplatformdependency.2)InterpretedlanguageslikePythonareeasiertouseandmoreportab

Python: For and While Loops, the most complete guidePython: For and While Loops, the most complete guideMay 09, 2025 am 12:05 AM

In Python, a for loop is used to traverse iterable objects, and a while loop is used to perform operations repeatedly when the condition is satisfied. 1) For loop example: traverse the list and print the elements. 2) While loop example: guess the number game until you guess it right. Mastering cycle principles and optimization techniques can improve code efficiency and reliability.

Python concatenate lists into a stringPython concatenate lists into a stringMay 09, 2025 am 12:02 AM

To concatenate a list into a string, using the join() method in Python is the best choice. 1) Use the join() method to concatenate the list elements into a string, such as ''.join(my_list). 2) For a list containing numbers, convert map(str, numbers) into a string before concatenating. 3) You can use generator expressions for complex formatting, such as ','.join(f'({fruit})'forfruitinfruits). 4) When processing mixed data types, use map(str, mixed_list) to ensure that all elements can be converted into strings. 5) For large lists, use ''.join(large_li

Python's Hybrid Approach: Compilation and Interpretation CombinedPython's Hybrid Approach: Compilation and Interpretation CombinedMay 08, 2025 am 12:16 AM

Pythonusesahybridapproach,combiningcompilationtobytecodeandinterpretation.1)Codeiscompiledtoplatform-independentbytecode.2)BytecodeisinterpretedbythePythonVirtualMachine,enhancingefficiencyandportability.

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

Video Face Swap

Video Face Swap

Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Tools

Dreamweaver CS6

Dreamweaver CS6

Visual web development tools

VSCode Windows 64-bit Download

VSCode Windows 64-bit Download

A free and powerful IDE editor launched by Microsoft

SublimeText3 Linux new version

SublimeText3 Linux new version

SublimeText3 Linux latest version

Zend Studio 13.0.1

Zend Studio 13.0.1

Powerful PHP integrated development environment

Safe Exam Browser

Safe Exam Browser

Safe Exam Browser is a secure browser environment for taking online exams securely. This software turns any computer into a secure workstation. It controls access to any utility and prevents students from using unauthorized resources.