lxml再探

常用缩略语

1. API：应用程序编程接口
2. DOM：文档对象模型
3. HTML：超文本标记语言
4. SAX：用于XML的简单API
5. XML：可扩展标记语言
6. XPath：XML路径语言
7. XSLT：可扩展样式表语言转换

使用iterparse方法

lxml parse 方法读取整个文档并在内存中构建一个树。相对于 cElementTree，lxml 树的开销要高一些，因为它保持了更多有关节点上下文的信息，包括对其父节点的引用。使用这种方法解析一个 2G 的文档时，会使一个具有 2G RAM 的机器进入交换，这会大大影响性能。

lxml的iterparse方法是ElementTree API的扩展。 iterparse返回用于所选元素上下文的Python迭代器。 它接受两个有用的参数：要监视的事件的元组和标记名。

iterparse中的events参数start和end的用法

“首先访问外层 elements” 或“首先访问内层 elements”。

<level-1>
    <level-2-1>
        <level-3-1>level-3-1>
        <level-3-2>level-3-2>
    level-2-1>
    <level-2-2>
        <level-3-3>level-3-3>
        <level-3-4>level-3-4>
    level-2-2>
level-1>

1. start

from lxml.etree import iterparse

with open('foo.xml', 'r') as xml:
    for event, element in iterparse(xml, events=['start']):
        print(element.tag)
-----------------------------------------
level-1
level-2-1
level-3-1
level-3-2
level-2-2
level-3-3
level-3-4

2. end

with open('foo.xml', 'r') as xml:
    for event, element in iterparse(xml, events=['end']):
        print(element.tag)
---------------------------------------------
level-3-1
level-3-2
level-2-1
level-3-3
level-3-4
level-2-2
level-1

3. [start,end]

<?xml version="1.0"?>
<data>
    <country name="china">
        <rank updated="yes">2</rank>
        <year>2008</year>
        <gdppc>141100</gdppc>
        <neighbor name="Austria" direction="E"/>
        <neighbor name="Switzerland" direction="W"/>
    </country>
    <country name="hangkong">
        <rank updated="yes">5</rank>
        <year>2011</year>
        <gdppc>59900</gdppc>
        <neighbor name="Malaysia" direction="N"/>
    </country>
    <country name="Panama">
        <rank updated="yes">69</rank>
        <year>2011</year>
        <gdppc>13600</gdppc>
        <neighbor name="Costa Rica" direction="W"/>
        <neighbor name="Colombia" direction="E"/>
    </country>
</data>
------------------------------------------------------------
('start', <Element 'data' at 0x5d11828>
('start', <Element 'data' at 0x5d116a0>)
('start', <Element 'country' at 0x5d11c88>)
('start', <Element 'rank' at 0x5d11cc0>)
('end', <Element 'rank' at 0x5d11cc0>)
('start', <Element 'year' at 0x5d11d30>)
('end', <Element 'year' at 0x5d11d30>)
('start', <Element 'gdppc' at 0x5d11d68>)
('end', <Element 'gdppc' at 0x5d11d68>)
('start', <Element 'neighbor' at 0x5d11da0>)
('end', <Element 'neighbor' at 0x5d11da0>)
('start', <Element 'neighbor' at 0x5d11dd8>)
('end', <Element 'neighbor' at 0x5d11dd8>)
('end', <Element 'country' at 0x5d11c88>)
('start', <Element 'country' at 0x5d11e10>)
('start', <Element 'rank' at 0x5d11e48>)
('end', <Element 'rank' at 0x5d11e48>)
('start', <Element 'year' at 0x5d11e80>)
('end', <Element 'year' at 0x5d11e80>)
('start', <Element 'gdppc' at 0x5d11eb8>)
('end', <Element 'gdppc' at 0x5d11eb8>)
('start', <Element 'neighbor' at 0x5d11ef0>)
('end', <Element 'neighbor' at 0x5d11ef0>)
('end', <Element 'country' at 0x5d11e10>)
('start', <Element 'country' at 0x5d11f28>)
('start', <Element 'rank' at 0x5d11f60>)
('end', <Element 'rank' at 0x5d11f60>)
('start', <Element 'year' at 0x5d11f98>)
('end', <Element 'year' at 0x5d11f98>)
('start', <Element 'gdppc' at 0x5d11fd0>)
('end', <Element 'gdppc' at 0x5d11fd0>)
('start', <Element 'neighbor' at 0x43a3390>)
('end', <Element 'neighbor' at 0x43a3390>)
('start', <Element 'neighbor' at 0x43a3cc0>)
('end', <Element 'neighbor' at 0x43a3cc0>)
('end', <Element 'country' at 0x5d11f28>)
('end', <Element 'data' at 0x5d116a0>)

start就是一个标签的开始，end就是一个标签的结尾

Python：使用基于事件驱动的SAX解析XML