update

Former-commit-id: 6b3ec58

Author: wangzhankun

实验楼nlp/1.1.py

@@ -0,0 +1,148 @@
+#%% [markdown]
+## 字符串操作
+### count()返回特定子串在字符串中出现的次数
+# In[]
+seq = '12345,1234,123,12,1'
+seql = '1'
+a = seq.count(seql)
+a
+
+#%% [markdown]
+### strip()去除字符串首尾的指定符号。无指定时，默认去除空格符和换行符
+
+# In[]
+seq = '我们正在使用实验楼，实验楼学到很多！'
+seq.strip()
+seq.strip('！')
+seq.strip('我们')
+
+#### lstrip()去除字符串左边指定的符号
+#### rstrip()去除字符串右边指定的符号
+
+# In[]
+seq = '12321'
+print(seq.lstrip('1'))
+print(seq.strip('1'))
+print(seq.rstrip('1'))
+
+### 字符串拼接
+#### '+'直接拼接
+#%%
+seq1 = '实'
+seq2 = '验'
+seq3 = '楼'
+seq = seq1 + seq2 + seq3
+print(seq)
+
+### 字符串拼接
+
+#### join()方法
+#%%
+seq = ['2018', '10', '31']
+seq = '-'.join(seq)
+print(seq)
+
+### 字符串比较
+#### 这里需要加载operator工具
+#%%
+import operator
+seq1 = '字符串1号'
+seq2 = '字符串2号'
+print(operator.gt(seq1,seq2))
+
+##### 直接使用运算符
+#%%
+print(seq1)
+print(seq2)
+print(seq1 < seq2)
+
+### 大小写转换
+#### upper(),lower()
+#%%
+seq = 'appLE'
+print(seq.upper())
+print(seq.lower())
+
+### 查找字符串
+#### find未找到则返回-1
+#%%
+seq = '这是一段字符串'
+seq1 = '字符串'
+print(seq.find(seq1))
+print(seq.find('无'))
+
+### 字符串截取
+#%%
+seq = '这是字符串'
+seq1 = seq[0:4]
+print(seq1)
+print(seq[0])
+print(seq[2:4])
+
+
+### 字符串切分
+#%%
+seq = '今天天气很好，我们出去玩'
+print(seq.split('，'))
+seq = '2018-11-11'
+print(seq.split('-'))
+seq = 'I have an apple'
+print(seq.split(' '))
+
+### 字符串翻转
+#%%
+seq = '12345'
+print(seq[::-1])
+
+
+### 字符串代替
+#%%
+seq = '2018-11-11'
+print(seq.replace('-','/'))
+
+### 以某字符串开头结尾
+#%%
+seq = 'abcdefg'
+print(seq.startswith('a'))
+print(seq.endswith('f'))
+
+
+
+## 正则表达式
+#### 不规则日期年份的提取
+#%%
+import re
+# 连续四个字符，每个字符是0-9
+pattern = re.compile(r'[0-9]{4}')
+time = '2018-01-01'
+match = pattern.search(time)
+match.group()
+
+#%%
+import re
+pattern = re.compile(r'[0-9]{4}')
+times = ('2018-01-01', '01/01/2019', '01.2017.01')
+for time in times:
+    match = pattern.search(time)
+    if match:
+        print('年份有：', match.group())
+
+### findall
+#%%
+import re
+# 识别数字
+pattern = re.compile(r'\d')
+print(pattern.findall('o1n2m3k4'))
+# 识别非数字
+pattern = re.compile(r'\D')
+print(pattern.findall('o1j2k3n4'))
+
+### match
+# match 与 search功能一样，但是只匹配一次,并且从开头开始匹配
+#%%
+import re
+pattern = re.compile('c')
+print(pattern.match('comcdc').group())
+pattern = re.compile('1')
+# 无法匹配1，因为match从开头匹配
+print(pattern.match('abcdefg1').group())