BeautifulSoup，一碗美麗的湯，一個隱藏的大坑
阿新 • • 發佈：2018-10-19
amp 標簽 隱藏 網頁 all process print輸出 網絡爬蟲 tree
python 網絡爬蟲常用的4大解析庫助手：re正則、etree xpath、scrapy xpath、BeautifulSoup。（因為etree xpath和scrapy xpath用法上有較大的不同，故沒有歸為一類），本文來介紹BeautifulSoup一個少為人知的坑，見示例：
例1(它是長得不一樣， 柬文勿怪)：
content = """
<html>
<body>
<div class="td-post-content td-pb-padding-side">
<p>
<img class="alignnone size-full wp-image-122426"
data-recalc-dims="1" height="352"
src="https://i2.wp.com/img.postnews.com.kh/2017/01/Anal-Itching.jpg?resize=630%2C352&ssl=1"
width="630"/>
</p>
<p>
<img class="alignnone size-full wp-image-122427"
data-recalc-dims="1" height="473"
src="https://i1.wp.com/img.postnews.com.kh/2017/01/Anal-Itching1.jpg?resize=630%2C473&ssl=1"
width="630"/>
</p>
<p>
????????????????? ?????????????????????????????
?????????????????????????????????????????????????????????????????
????????????????????????? ??????????????????????????
</p>
<p>
<img class="alignnone size-full wp-image-122427"
data-recalc-dims="1" height="473"
src="https://i1.wp.com/img.postnews.com.kh/2017/01/Anal-Itching1.jpg?resize=630%2C473&ssl=1"
width="630"/>
</p>
<p>
<img class="alignnone size-full wp-image-122428"
data-recalc-dims="1" height="473"
src="https://i2.wp.com/img.postnews.com.kh/2017/01/Anal-Itching2.jpg?resize=630%2C473&ssl=1"
width="630"/>
<br/>
<em>
<br/>
??????
</em>
??????????????????????? ???????????? ??????????????????
?????????????????????????????????????
</p>
</div>
</body>
</html>
"""
soup = BeautifulSoup(content)
img_lst = []
inner_src_list = soup.find_all(‘img‘, src=True)
for i, src in enumerate(inner_src_list):
url=src["src"].replace("&ssl", "&ssl")
print(url)
print(soup.prettify())
# content = soup.prettify() # src的打印結果一樣
img_tags = soup.find_all(‘img‘)
for img in img_tags:
print(img[‘src‘])
控制臺打印輸出如下：
怎麽會這樣：文本中的‘amp;’字符怎麽消失了？
解釋如下：BeautifulSoup在提取src時內部會自動把符號‘&’轉義成‘&‘，【網頁解析有時不一定要眼前的直覺】【不僅bs如此， etree xpath和scrapy xpath也是一樣】
例2：
文本同上
soup = BeautifulSoup(content)
img_lst = []
inner_src_list = soup.find_all(‘img‘, src=True) # 註意比較
for i, src in enumerate(inner_src_list):
url=src["src"].replace("&ssl", "&ssl")
print(url)
inner_src_list = soup.find_all(‘img‘, attr={‘src‘:True}) # 註意比較
for i, src in enumerate(inner_src_list):
url=src["src"].replace("&ssl", "&ssl")
print(url)
這裏不作打印了，直接說明現象，第一個print正常打印，第二個print輸出為空，為什麽？
解釋如下： 第一個find_all，把src=True視為存在src屬性的img標簽，第二個find_all，把attr={‘src‘, True}視為存在src且屬性值為True的img標簽，所以結果可想而知！
上述如有不正之處，歡迎指出，謝謝！BeautifulSoup，一碗美麗的湯，一個隱藏的大坑