 
                        要读取一个xml文件中每个item节点的review_id、summary等的子结点,样例如下:
<item>
<review_id>0095693</review_id>
<summary>书本内容</summary>
<polarity>P</polarity>
<text>书本的内容很好,对我很有帮助,就是字体的颜色是紫色的,看就了会觉得不清晰。</text>
<category>book</category>
</item>
完整例子可以从这里下载。编程环境为Mac 10.9.2,Python 2.7.6,代码如下:
import sys
import os
from xml.etree.ElementTree import iterparse, tostring
def count_pos_neg(itemfile):
    pos_count = 0
    neg_count = 0
    try:
        for event, elem in iterparse(itemfile, events=["start",]):
            if elem.tag == "item":
                try:
                    if processItem(elem)['polarity'] == "P":
                        pos_count += 1
                    else:
                        neg_count += 1
                except Exception, e:
                    print >> sys.stderr, "Ignoring item: %s" % e
                elem.clear()
    except SyntaxError, se:
        print >> sys.stderr, se
    return pos_count, neg_count
def processItem(item):
    """ Process a review.
    Implement custom code here. Use 'item.find('tagname').text' to access the properties of a review. 
    """
    category = item.find("category").text
    polarity = item.find("polarity").text
    text = item.find("text").text
    summary = item.find("summary").text
    return {'polarity':polarity,
            'summary':summary,
            'text':text,
            'category':category }
if __name__ == "__main__":
    pc, nc = count_pos_neg(itemfile)
问题在于,每碰到第55个item节点,就会发生一次AttruibuteError,错误信息为
Ignoring item: 'NoneType' object has no attribute 'text'
我在使用evens=('end',)进行解析时,没有发生错误。这是否说明之前的错误与使用start解析有关?
Copyright 2014-2025 https://www.php.cn/ All Rights Reserved | php.cn | 湘ICP备2023035733号
文档说:
start事件发生时这个元素的子元素还没有解析,所以你应该用end事件。