README

简体中文

phpy

A library for inter-calling Python and PHP. You can use Python functions and libraries in PHP, or use PHP packages in Python.

• See documents: docs/en/README.md
• Supports Linux/Windows/macOS
• Not support Python threading or async-io features
• Require PHP 8.1 or later version

py2php

py2php is online utility that will auto-translate python code into PHP code.

Calling Python from PHP

Compile and install phpy.so as an extension, and append extension=phpy.so to php.ini.

PHP Example:

$os = PyCore::import("os");
echo $os->uname();

Transformers

$transformers = PyCore::import('transformers');
$AutoTokenizer = $transformers->AutoTokenizer;
$AutoModelForSequenceClassification = $transformers->AutoModelForSequenceClassification;

$os = PyCore::import('os');
$os->environ['https_proxy'] = getenv('https_proxy');

$tokenizer = $AutoTokenizer->from_pretrained("lxyuan/distilbert-base-multilingual-cased-sentiments-student");
$model = $AutoModelForSequenceClassification->from_pretrained("lxyuan/distilbert-base-multilingual-cased-sentiments-student");

Calling PHP from Python

Simply import it as a C++ Mudule.

Python Example:

import phpy
content = phpy.call('file_get_contents', 'test.txt')

o = phpy.Object('redis')
assert o.call('connect', '127.0.0.1', 6379)
rdata = phpy.call('uniqid')
assert o.call('set', 'key', rdata)
assert o.call('get', 'key') == rdata

Implementation

It creates ZendVM and CPython VM in the process at the same time, and directly uses C functions to call each other in the process stack space.

The overhead is only the conversion of zval <-> PyObject structures, so the performance is very high.

In the benchmark test, we created a PyDict and executed PHP code and Python code to read and write 10 million times respectively.

The performance of phpy writing PyDict with PHP code is 14% higher than the native Python, and the read performance is 25% higher.

More details: docs/en/benchmark.md