示例
一个简单的示例. 比如写了一个名为 saudade 的包. 目录结构为(用 tree 生成, 不错的程 序, 之前没用过)
.
├── demo
│ └── __init__.py
├── saudade
│ ├── coding_func.py
│ ├── __init__.py
│ └── physics_func.py
└── setup.py
直接执行
python setup.py install --user
即可安装完成. --user
安装给当前用户, 直接就放在 ~/.local/lib/python_xx
目录下
了, 不影响其它用户, 也不需要管理员权限.
pip uninstall saudade
即可卸载. 当然也可以传到 github, 直接从 github 上安装.
结果
安装完以后就可以调用了. 如
>>> import saudade as sau
>>> @sau.timer
... def foo(x):
... return x**2
...
>>> foo(3)
Finished 'foo' in 0.0000 secs
9
>>>
可以看出, 可以调用包的中计时器了.
而 demo
目录下的函数, 调用时要调用 demo , 如
>>> import demo
>>> demo.hello()
hello
>>>
源码说明
file:./2021-01-13-coding-setup_tools/setup.py
from setuptools import setup, find_packages
setup(
name = 'saudade',
version ='0.1',
packages = find_packages(),
description = "This is ZQW's first python package",
author = 'ZQW',
author_email = "zeqing6688@126.com",
)
此文件最重要, 它说明这个目录是一个包, 可以安装.
demo/file:./2021-01-13-coding-setup_tools/demo/__init__.py
def hello():
print('hello')
这是一个测试函数
saudade/file:./2021-01-13-coding-setup_tools/saudade/coding_func.py
import numpy as np
#import matplotlib.pyplot as plt
import functools
import time
def timer(func):
"""
Print the runtime of the decorated function.
参考自: https://realpython.com/primer-on-python-decorators/
"""
@functools.wraps(func)
def wrapper_timer(*args, **kwargs):
start_time = time.perf_counter()
value = func(*args, **kwargs)
end_time = time.perf_counter()
run_time = end_time - start_time
print(f"Finished {func.__name__!r} in {run_time:.4f} secs")
return value
return wrapper_timer
@timer
def get_data(parameters, func, x):
'''
计算画图所需要的点, 并将参数和结果保存到文件.
由于自己写的这些函数不能输入数组然后输出数组, 所以只能用循环一个一个算
'''
num_points = len(x)
y = []
pm = parameters.get_parameters()
# 计算每一个数据点
for i in range(num_points):
print(f"正在计算第 {i+1:d}/{num_points:d}个点...")
yi = func(x[i])
y.append(yi)
# 保存数据和参数到文件
np.savetxt('./data/x.csv', x, delimiter=',')
np.savetxt('./data/y.csv', y, delimiter=',')
np.savetxt('./data/paramaters.csv', pm, delimiter=',')
return x, y, pm
saudade
目录的名字与包的名字相同, 包含了包的主要内容. 此文件是包中的一些函数
saudade/./2021-01-13-coding-setup_tools/saudade/__init__.py
from .coding_func import *
from .physics_func import *
整个包的初始化位置.
saudade/./2021-01-13-coding-setup_tools/saudade/physics_func.py
import numpy as np
def ts(f, a, b, n=51):
"""Tanh-sinh quadrature 方法. 适用于端点发散的情况."""
up = 4
h = 2*up / (n-1)
t = np.linspace(-up, up, n, endpoint=True)
x = np.tanh(1/2*np.pi*np.sinh(t))
w = 1/2*h*np.pi*np.cosh(t)
w = w/(np.cosh(1/2*np.pi*np.sinh(t))**2)
gc = 0
for i in range(n):
p = (x[i]*(b-a) + a + b)/2
gc = gc + f(p)*w[i]
err = 0
gc = gc * (b-a)/2
return gc, err
def bose(beta, energy):
"""Bose 分布函数"""
x = -beta * energy
return np.exp(x) / (1 - np.exp(x))
def cos_theta_kq(theta_k, phi_k, theta_q, phi_q):
"""k, q 夹角的余弦值"""
x = (np.sin(theta_k)*np.sin(theta_q) * np.cos(phi_k - phi_q)
+ np.cos(theta_k)*np.cos(theta_q))
return x
class PrincipalValueInt():
"""分母带有无穷小的那种积分"""
def __init__(self, numerator, coeff, down_bound, upbound):
"""初始化, numerator 都是函数. 分母为 a*x**2 + b*x + c"""
self.numerator = numerator
self.down_bound = down_bound
self.upbound = upbound
self.coeff = coeff
a = coeff[0]
b = coeff[1]
c = coeff[2]
self.delta = b**2 - 4*a*c
def get_imag(self):
"""计算积分的虚部."""
root_exist = self.delta > 0
if root_exist:
# 如果根存在, 计算两根
root1 = (-self.coeff[1] - np.sqrt(self.delta)) / (2 * self.coeff[0])
root2 = (-self.coeff[1] + np.sqrt(self.delta)) / (2 * self.coeff[0])
# 判断两根是否位于积分区间内
root1_in = self.down_bound < root1 and root1 < self.upbound
root2_in = self.down_bound < root2 and root2 < self.upbound
# 计算积分结果
imag = (root1_in) * self.numerator(root1)
imag += (root2_in) * self.numerator(root2)
imag *= -np.pi / np.abs(root2 - root1)
else:
# 根不存在, 虚部为 0
imag = 0
imag *= 1/self.coeff[0] # bug No.2 分子要除以 a 才行.
return imag
包中的另一些函数.