Question

我的xml中有一些部分需要重新订购，我知道xml不需要重新订购，但这是我需要做的，但不能弄清楚＆＃39;正确＆＃39;这样做的方式。我正在使用lxml并且一直使用.insert命令重新排序。我需要重新排序每个<asset type="preview">内的每个标签，以便它看起来像这样：

    <asset type="preview">
        <territories>
            <territory>SE</territory>
        </territories>
        <data_file role="source">
            <locale name="es"/>
            <file_name>some_name_nor-preview-sv.mov</file_name>
            <size>1715119116</size>
            <checksum type="md5">55cd94d051700be34014b2892e925fa1</checksum>
            <attribute name="crop.top">25</attribute>
            <attribute name="crop.bottom">25</attribute>
            <attribute name="crop.left">4</attribute>
            <attribute name="crop.right">4</attribute>
            <attribute name="image.burned_subtitles.locale">sv</attribute>
            <attribute name="image.textless_master">false</attribute>
        </data_file>
    </asset>

我有时会有多个<asset type="preview">，有时不会。有时，每个<asset type="preview">都不包含此处列出的所有标记。以下是我试图按上述方式重新排序的xml部分。

    <asset type="preview">
        <data_file role="source">
            <size>1657800204</size>
            <file_name>some_name_nor-preview.mov</file_name>
            <checksum type="md5">c61dfa7139ab04560cac41cf5ba8a1f2</checksum>
            <locale name="es"/>
            <attribute name="crop.top">25</attribute>
            <attribute name="crop.right">4</attribute>
            <attribute name="crop.bottom">25</attribute>
            <attribute name="crop.left">4</attribute>
        </data_file>
        <territories>
            <territory>WW</territory>
        </territories>
        <data_file role="notes">
            <size>9642</size>
            <file_name>some_name_nor-preview-notes.pdf</file_name>
            <checksum type="md5">4d0dc3534cd1d0f9885afbfda9be8b71</checksum>
        </data_file>
    </asset>
    <asset type="preview">
        <data_file role="source">
            <size>1715119116</size>
            <file_name>some_name_nor-preview-sv.mov</file_name>
            <checksum type="md5">55cd94d051700be34014b2892e925fa1</checksum>
            <locale name="es"/>
            <attribute name="image.burned_subtitles.locale">sv</attribute>
            <attribute name="crop.top">25</attribute>
            <attribute name="crop.right">4</attribute>
            <attribute name="image.textless_master">false</attribute>
            <attribute name="crop.left">4</attribute>
            <attribute name="crop.bottom">25</attribute>
        </data_file>
        <territories>
            <territory>SE</territory>
        </territories>
    </asset>
    <asset type="preview">
        <data_file role="source">
            <size>1709158524</size>
            <file_name>some_name_nor-preview-fi.mov</file_name>
            <checksum type="md5">58c5fcfa718393f76cb9b2d8f7c10362</checksum>
            <locale name="es"/>
            <attribute name="crop.bottom">25</attribute>
            <attribute name="crop.top">25</attribute>
            <attribute name="crop.left">4</attribute>
            <attribute name="image.textless_master">false</attribute>
            <attribute name="crop.right">4</attribute>
            <attribute name="image.burned_subtitles.locale">fi</attribute>
        </data_file>
        <territories>
            <territory>FI</territory>
        </territories>
    </asset>
    <asset type="preview">
        <territories>
            <territory>NO</territory>
        </territories>
        <data_file role="source">
            <size>1718632572</size>
            <file_name>some_name_nor-preview-no.mov</file_name>
            <checksum type="md5">41734d9d8dd4165416a4369f4ce9c8e1</checksum>
            <locale name="es"/>
            <attribute name="crop.left">4</attribute>
            <attribute name="crop.top">25</attribute>
            <attribute name="crop.bottom">25</attribute>
            <attribute name="image.textless_master">false</attribute>
            <attribute name="image.burned_subtitles.locale">no</attribute>
            <attribute name="crop.right">4</attribute>
        </data_file>
    </asset>
    <asset type="preview">
        <territories>
            <territory>DK</territory>
        </territories>
        <data_file role="source">
            <size>1721312028</size>
            <file_name>some_name_nor-preview-da.mov</file_name>
            <checksum type="md5">919abd17baf680161a220dbae8409918</checksum>
            <locale name="es"/>
            <attribute name="image.textless_master">false</attribute>
            <attribute name="crop.bottom">25</attribute>
            <attribute name="image.burned_subtitles.locale">da</attribute>
            <attribute name="crop.right">4</attribute>
            <attribute name="crop.left">4</attribute>
            <attribute name="crop.top">25</attribute>
        </data_file>
    </asset>

这是我目前的工作＆＃39;代码，它没有重新排序attribute[@name=代码，不确定这是正确的方法：

        a = 0
        b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag='locale'):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/locale")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag='file_name'):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/file_name")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1           
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag='size'):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/size")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag='checksum'):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/checksum")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='crop.top']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='crop.top']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='crop.bottom']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='crop.bottom']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='crop.left']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='crop.left']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='crop.right']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='crop.right']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='image.burned_forced_narrative.locale']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='image.burned_forced_narrative.locale']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='image.burned_subtitles.locale']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='image.burned_subtitles.locale']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0
        for node_search in tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']"):
            for element in node_search.iter(tag="attribute[@name='image.textless_master']"):
                node_products = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']")[b]
                node_type = tree.xpath("//video/assets/asset[@type='preview']/data_file[@role='source']/attribute[@name='image.textless_master']")[b]
                node_products.insert(a, node_type)
                b = b+1
            a = a+1
            b = 0

Answer 1

我对你的要求并不完全清楚。以下代码按此顺序对每个asset_preview进行排序：

unknown tags
<territories>
unknown <data_file> roles
<data_file role=source>
<data_file role=notes>

并对每个data_file进行排序：

unknown tags
<locale>
<file_name>
<size>
<checksum>
unknown attributes
<attribute name="crop.top">
other <attributes>, in a specific order.

理解这种技术的关键是要意识到节点是一个列表，并且可以按照重新排序任何列表的方式重新排序。就我而言，我使用sorted()和自定义密钥。

你走了：

from lxml import etree

def preview_key(et):
    major_ordering = ['territories', 'data_file']
    minor_ordering = ['source', 'notes']
    try:
        major = major_ordering.index(et.tag)
    except ValueError:
        major = -1
    try:
        minor = minor_ordering.index(et.get('role', None))
    except ValueError:
        minor = -1
    return major, minor

def data_file_key(et):
    major_ordering = ['locale', 'file_name', 'size', 'checksum', 'attribute']
    minor_ordering = [
            "crop.top",
            "crop.bottom",
            "crop.left",
            "crop.right",
            "image.burned_subtitles.locale",
            "image.textless_master"]
    try:
        major = major_ordering.index(et.tag)
    except ValueError:
        major = -1
    try:
        minor = minor_ordering.index(et.get('name', None))
    except ValueError:
        minor = -1
    return major, minor



with open('input.xml') as input_file:
    parser = etree.XMLParser(remove_blank_text=True)
    tree = etree.parse(input_file, parser)
root = tree.getroot()

for preview in tree.xpath("//asset[@type='preview']"):
    preview[:] = sorted(preview, key=preview_key)

for data_file in tree.xpath("//data_file"):
    data_file[:] = sorted(data_file, key=data_file_key)

with open('output.xml', 'w') as output_file:
    output_file.write(etree.tostring(tree, pretty_print = True))

Python - lxml重新排序xml标签

1 个答案: