day13

Author: congboqiu

19100401/congboqiu/d13/d13/main.py

@@ -0,0 +1,47 @@
+import re
+import jieba
+import getpass
+import requests
+import wxpy
+from pyquery import PyQuery
+from mymodule import stats_word
+
+#导入模块wxpy，建微信机器人
+from wxpy import *
+bot=Bot(cache_path=True)
+my_friend=bot.friends().search('翕羊羊')[0]
+my_friend.send('你好，请分享篇文章给我')#发消息给朋友
+
+
+@bot.register(my_friend,msg_types='Sharing')
+def auto_reply(msg):
+    print(msg)
+    print(msg.url)
+    response = requests.get(msg.url)
+    document = PyQuery(response.text)
+    content = document('#js_content').text()
+    text=content
+    
+    #引入变量words和count，为图像化函数做好参数引入
+    dict_text=dict(stats_word.stats_text_cn(text,10)) #把结果转化字典
+    key_words=list(dict_text.keys())#把字典的key和对应值转换为列表
+    count_values=list(dict_text.values()) #把字典的value和对应值转换为列表
+   
+    #用matplotlib把words和count图像化
+    import matplotlib.pyplot as plt
+    import numpy as np
+    plt.rcdefaults()
+    fig,ax = plt.subplots()
+    plt.rcParams['font.sans-serif']=['SimHei'] 
+    ax.barh(key_words,count_values, align = "center", color = "blue")
+    ax.set_yticks(key_words)
+    ax.set_yticklabels(key_words)
+    ax.invert_yaxis()
+    ax.set_xlabel(count_values)
+    ax.set_title('你刚才所发文章的词频前10统计')
+    
+    plt.savefig('result.png')
+    msg.reply_image('result.png')
+    
+embed()
+

19100401/congboqiu/d13/d13/mymodule/stats_word.py

@@ -0,0 +1,51 @@
+
+import re
+import jieba
+from collections import Counter
+counter=20
+def stats_text_en(text, counter):
+
+    if not isinstance(text, str):
+        raise ValueError("I can only handle a string type text")
+
+    # first removing "-", useless space, then changing all words into lower-case.
+    text_p = text.replace("\-", " ").strip().lower()
+    # don't forgeting the "\n", Otherwise it will bring some awkawk words
+    i = re.compile("[^a-z \n]")
+    text_en = i.sub("", text_p).split()
+    
+    # using Sounter to create a dictionary sorted by frequency
+    sort_en = Counter(text_en).most_common(counter)
+    return sort_en
+
+
+def stats_text_cn(text, counter):
+
+    if not isinstance(text, str):
+        raise ValueError("I can only handle a string type text")
+
+    # only chinese left and change into a list.
+    text_cn = ""
+    for t in text:
+        if ord(t) > 256:#ASCII编码，去除英文及符号
+            text_cn = text_cn + t
+        else:
+            text_cn = text_cn + "，"
+    list_cn_word = ([x for x in jieba.cut(text_cn, cut_all=False) if len(x) >= 2])
+
+    # using Sounter to create a dictionary sorted by frequency
+    sort_cn = Counter(list_cn_word).most_common(counter)
+    return sort_cn
+    print(sort_cn)
+
+def stats_text(text, counter):
+    if not isinstance(text, str):
+        raise ValueError("I can only handle a string type text")
+
+    sort_en = stats_text_en(text, counter)
+    sort_cn = stats_text_cn(text, counter)
+
+
+    
+
+