1 import csv
 2 with open(r"thermistor.txt","rb") as f:
 3       reader = csv.reader(f,delimiter='\t')
 4       number=[]
 5       time = []
 6       data=[]
 7       for row in reader:
 8            number.append(row[0])
 9            time.append(row[1])
10            data.append(float(row[2])) 

>>> print number[0], time[0], data[0] 

>>> 2018-02-15 21:31:08.781 49.9492

>>> print number[0:3], time[0:3], data[0:3] 

>>> ['\xef\xbb\xbf1', '2', '3']
['2018-02-15 21:31:08.781', '2018-02-15 21:31:09.296', '2018-02-15 21:31:09.811']
[49.9492, 49.9589, 49.964] 

1 import csv
2 import codecs
3 with codecs.open(r"thermistor.txt","rb","utf-8-sig") as f:
4     reader = csv.reader(f,delimiter='\t')
5     number=[] time=[] data=[]
6     for row in reader:
7         number.append(row[0])
8         time.append(row[1])
9         data.append(float(row[2])

>>> ['1', '2', '3']
['2018-02-15 21:31:08.781', '2018-02-15 21:31:09.296', '2018-02-15 21:31:09.811']
[49.9492, 49.9589, 49.964]

Type: raw
Points: 16200
Count: 1
XInc: 1.23457E-013
XOrg: 2.4000000000E-008
YData range: 1.48000E-001
YData center: 5.00000E-004
Coupling: 50
Ohms XRange: 2.00000E-009
XOffset: 2.4000000000E-008
YRange: 1.44000E-001
YOffset: 5.00000E-004
Date: 15 APR 2018
Time: 16:00:54:74
Frame: 86100C:MY46520443
X Units: second
Y Units: Volt
XY Data:
2.4000000E-008, 1.4349E-002
2.4000123E-008, 1.6005E-002
2.4000247E-008, 1.5455E-002
2.4000370E-008, 1.5702E-002
2.4000494E-008, 1.5147E-002
... 

1 with open(r"waveform.txt","rb") as f:
2     lines = f.readlines() x=[] y=[]
3     for line in lines[18:]:
4         x.append(float(line.replace("\r\n","").split(",")[0]))
5         y.append(float(line.replace("\r\n","").split(",")[1]))

>>> [2.4e-08, 2.4000123e-08, 2.4000247e-08]
[0.014349, 0.016005, 0.015455] # 数据读取正确

1 import re
2 with open(r"waveform.txt","rb") as f:
3     lines = f.readlines()
4     x=[]
5     y=[]
6     for line in lines:
7         if re.search('E-002',line):
8             x.append(float(line.replace("\r\n","").split(",")[0]))
9             y.append(float(line.replace("\r\n","").split(",")[1]))

>>> [2.4e-08, 2.4000123e-08, 2.4000247e-08]
[0.014349, 0.016005, 0.015455] # 数据读取正确

注：具体的数据读取方式要根据具体文本文件的特征决定，运用合适的方法才能得到更好的结果。

案例三：二维数据写入

很多时候，经过 process( ) 后的数据，需要备份留用或者供其他程序调用，因此，将处理后的数据写入文本文件也将是关键的一步。根据数据读入的经验，被读入的数据经常存储在 list 中，那么处理后数据也通常存储在 list 中，因此，以 list 的写入作为例子：

x = [1, 2, 3, 4]
y = [2.0, 4.0, 6.0, 8.0] # 参考数据

1 xy = {}
2 for i in range(len(x)):
3     xy[x[i]] = y[i]
4 with open(r"15.txt", 'wb') as f:
5     writer = csv.writer(f,delimiter='\t')
6     for x, y in xy.items():
7         writer.writerow([x, y]) 

1   2.0
2   4.0
3   6.0
4   8.0

x = [1, 2, 3, 4]
y = [2.0, 4.0, 6.0, 8.0]
z = [3.0, 6.0, 9.0, 12.0]

1 xyz = []
2 for i in range(len(x)):
3     xyz.append([x[i],y[i],z[i]])
4 with open(r"15.txt", 'wb') as f:
5     writer = csv.writer(f,delimiter='\t')
6     for x, y, z in xyz:
7         writer.writerow([x, y, z]) 

1   2.0 3.0
2   4.0 6.0
3   6.0 9.0
4   8.0 12.0

我们已经提到了两种方法读取上述的数据，它们共同点是将数据存储在列表中，正如开头所说，列表在处理大量数据时是非常缓慢的。那么，我们就来看一看 numpy.genfromtxt 如何大显身手。

代码示例

为了得到我们需要的有用数据，我们有两个硬的要求： (1) 跳过表头信息；(2) 区分横纵坐标。

import numpy as np
data = np.genfromtxt('waveform.txt',delimiter=',',skip_header=18)

print data[0:3,0], data[0:3,1]

因为读入的是二维数据，因此利用 numpy 二维数据的切片方式 (Index slicing) 输出各自的前三个数据验证是否读取正确：

[  2.40000000e-08   2.40001230e-08   2.40002470e-08]
[ 0.014349  0.016005  0.015455]

对数据进行归一化处理后，调用 Matplotlib 画图命令，就可得到图像如下：

1 import matplotlib.pyplot as plt
2 fig, axes = plt.subplots(figsize=(8,6))
3 axes.plot(x, y, 'r', linewidth=3)
4 axes.set_xlabel('Time(ps)')
5 axes.set_ylabel('Amplitude[a.u.]')
6 fig.savefig("triangular.png", dpi=600)