Python 解析包含许多子代和子代的XML文件

我对python非常陌生。我有一个非常大的xml文件，我想从中提取一些数据。以下是摘录：

<program>
    <id>38e072a7-8fc9-4f9a-8eac-3957905c0002</id>
    <programID>3853</programID>
    <orchestra>New York Philharmonic</orchestra>
    <season>1842-43</season>
    <concertInfo>
        <eventType>Subscription Season</eventType>
        <Location>Manhattan, NY</Location>
        <Venue>Apollo Rooms</Venue>
        <Date>1842-12-07T05:00:00Z</Date>
        <Time>8:00PM</Time>
    </concertInfo>
    <worksInfo>
        <work ID="52446*">
            <composerName>Beethoven,  Ludwig  van</composerName>
            <workTitle>SYMPHONY NO. 5 IN C MINOR, OP.67</workTitle>
            <conductorName>Hill, Ureli Corelli</conductorName>
        </work>
        <work ID="8834*4">
            <composerName>Weber,  Carl  Maria Von</composerName>
            <workTitle>OBERON</workTitle>
            <movement>"Ozean, du Ungeheuer" (Ocean, thou mighty monster), Reiza (Scene and Aria), Act II</movement>
            <conductorName>Timm, Henry C.</conductorName>
            <soloists>
                <soloist>
                    <soloistName>Otto, Antoinette</soloistName>
                    <soloistInstrument>Soprano</soloistInstrument>
                    <soloistRoles>S</soloistRoles>
                </soloist>
            </soloists>
        </work>
        <work ID="3642*">
            <composerName>Hummel,  Johann</composerName>
            <workTitle>QUINTET, PIANO, D MINOR, OP. 74</workTitle>
            <soloists>
                <soloist>
                    <soloistName>Scharfenberg, William</soloistName>
                    <soloistInstrument>Piano</soloistInstrument>
                    <soloistRoles>A</soloistRoles>
                </soloist>
                <soloist>
                    <soloistName>Hill, Ureli Corelli</soloistName>
                    <soloistInstrument>Violin</soloistInstrument>
                    <soloistRoles>A</soloistRoles>
                </soloist>
                <soloist>
                    <soloistName>Derwort, G. H.</soloistName>
                    <soloistInstrument>Viola</soloistInstrument>
                    <soloistRoles>A</soloistRoles>
                </soloist>
                <soloist>
                    <soloistName>Boucher, Alfred</soloistName>
                    <soloistInstrument>Cello</soloistInstrument>
                    <soloistRoles>A</soloistRoles>
                </soloist>
                <soloist>
                    <soloistName>Rosier, F. W.</soloistName>
                    <soloistInstrument>Contrabass</soloistInstrument>
                    <soloistRoles>A</soloistRoles>
                </soloist>
            </soloists>
        </work>
        <work ID="0*">
            <interval>Intermission</interval>
        </work>
        <work ID="8834*3">
            <composerName>Weber,  Carl  Maria Von</composerName>
            <workTitle>OBERON</workTitle>
            <movement>Overture</movement>
            <conductorName>Etienne, Denis G.</conductorName>
        </work>
        <work ID="8835*1">
            <composerName>Rossini,  Gioachino</composerName>
            <workTitle>ARMIDA</workTitle>
            <movement>Duet</movement>
            <conductorName>Timm, Henry C.</conductorName>
            <soloists>
                <soloist>
                    <soloistName>Otto, Antoinette</soloistName>
                    <soloistInstrument>Soprano</soloistInstrument>
                    <soloistRoles>S</soloistRoles>
                </soloist>
                <soloist>
                    <soloistName>Horn, Charles Edward</soloistName>
                    <soloistInstrument>Tenor</soloistInstrument>
                    <soloistRoles>S</soloistRoles>
                </soloist>
            </soloists>
        </work>
        <work ID="8837*6">
            <composerName>Beethoven,  Ludwig  van</composerName>
            <workTitle>FIDELIO, OP. 72</workTitle>
            <movement>"In Des Lebens Fruhlingstagen...O spur ich nicht linde," Florestan (aria)</movement>
            <conductorName>Timm, Henry C.</conductorName>
            <soloists>
                <soloist>
                    <soloistName>Horn, Charles Edward</soloistName>
                    <soloistInstrument>Tenor</soloistInstrument>
                    <soloistRoles>S</soloistRoles>
                </soloist>
            </soloists>
        </work>
        <work ID="8336*4">
            <composerName>Mozart,  Wolfgang  Amadeus</composerName>
            <workTitle>ABDUCTION FROM THE SERAGLIO,THE, K.384</workTitle>
            <movement>"Ach Ich liebte," Konstanze (aria)</movement>
            <conductorName>Timm, Henry C.</conductorName>
            <soloists>
                <soloist>
                    <soloistName>Otto, Antoinette</soloistName>
                    <soloistInstrument>Soprano</soloistInstrument>
                    <soloistRoles>S</soloistRoles>
                </soloist>
            </soloists>
        </work>
        <work ID="5543*">
            <composerName>Kalliwoda,  Johann  W.</composerName>
            <workTitle>OVERTURE NO. 1, D MINOR, OP. 38</workTitle>
            <conductorName>Timm, Henry C.</conductorName>
        </work>
    </worksInfo>
</program>
<program>

当我运行这段代码时，我只得到最后一个soloistName和soloistInstrumet。我心目中的结果有点像对每个项目的重复观察。所以我会有这样的想法：

13918年，纽约爱乐乐团，1842-1843年，订阅季，52446*，奥托，安托瓦内特，女高音，S

13918，…，3642*，沙尔芬伯格，威廉，钢琴，A

13918，…，3642*，希尔，乌雷利·科雷利，小提琴，A

依此类推，直到最后一个工作ID：

13918，…，8336*4，奥托，安托瓦内特，女高音，S

我得到的只是最后的工作：

13918年，纽约爱乐乐团，1842-1843年，订阅季，8336*，奥托，安托瓦内特，女高音，S

---

文件中有超过15000个程序，如我发布的示例。我想解析所有这些，并提取上面提到的信息。我不完全确定如何去做这件事，我已经在互联网上搜索了一种方法来做这件事，但我尝试的一切都不起作用

您的问题在于您误解了循环的工作方式。具体来说，这些值仅在循环中更改：

for x in range(10):
    pass

print(x) # prints 9

vs

这是两件不同的事情。你在做前者。您需要做的是这样的事情：

with open('nyphil.txt', 'w') as f:
    nyphilwriter = csv.writer(f)        
    for program in root.iter('program'):
        id_ = program.findtext('id')
        program_id = program.findtext('programID')
        orchestra = program.findtext('orchestra')
        season = program.findtext('season')
        for concert in program.findall('concertInfo'):
            event = concert.findtext('eventType')
        for info in program.findall('worksInfo'):
            for work in info.iter('work'):
                work_id = work.get('ID')
                for soloists in work.iter('soloists'):
                    for soloist in soloists.iter('soloist'):
                        # Change this line to whatever you want to write out
                        nyphilwriter.writerow([id, program_id, orchestra, season, event, work_id, soloist.findtext('soloistName')])

---

这里的问题是您误解了循环的工作方式。具体来说，这些值仅在循环中更改：

for x in range(10):
    pass

print(x) # prints 9

vs

这是两件不同的事情。你在做前者。您需要做的是这样的事情：

with open('nyphil.txt', 'w') as f:
    nyphilwriter = csv.writer(f)        
    for program in root.iter('program'):
        id_ = program.findtext('id')
        program_id = program.findtext('programID')
        orchestra = program.findtext('orchestra')
        season = program.findtext('season')
        for concert in program.findall('concertInfo'):
            event = concert.findtext('eventType')
        for info in program.findall('worksInfo'):
            for work in info.iter('work'):
                work_id = work.get('ID')
                for soloists in work.iter('soloists'):
                    for soloist in soloists.iter('soloist'):
                        # Change this line to whatever you want to write out
                        nyphilwriter.writerow([id, program_id, orchestra, season, event, work_id, soloist.findtext('soloistName')])

---

13918不会出现在您的数据中。撇开这一点不谈，下面是我写的，它似乎成功地处理了您的数据

from lxml import etree

tree = etree.parse('test.xml')
programs = tree.xpath('.//program')

for program in programs:
    programID, orchestra, season = [program.xpath(_)[0].text for _ in ['programID', 'orchestra', 'season']]
    print (programID, orchestra, season)
    works = program.xpath('worksInfo/work')
    for work in works:
        workID = work.attrib['ID']
        soloistItems = work.xpath('soloists/soloist')
        for soloistItem in soloistItems:
            print (workID, soloistItem.find('soloistName').text, soloistItem.find('soloistInstrument').text, soloistItem.find('soloistRoles').text)

该脚本生成以下输出

3853 New York Philharmonic 1842-43
8834*4 Otto, Antoinette Soprano S
3642* Scharfenberg, William Piano A
3642* Hill, Ureli Corelli Violin A
3642* Derwort, G. H. Viola A
3642* Boucher, Alfred Cello A
3642* Rosier, F. W. Contrabass A
8835*1 Otto, Antoinette Soprano S
8835*1 Horn, Charles Edward Tenor S
8837*6 Horn, Charles Edward Tenor S
8336*4 Otto, Antoinette Soprano S

---

还有一件事需要注意：我在XML的开头放了一个标记，在末尾放了一个标记，因为实际数据将包含多个元素。

13918不会出现在数据中。撇开这一点不谈，下面是我写的，它似乎成功地处理了您的数据

from lxml import etree

tree = etree.parse('test.xml')
programs = tree.xpath('.//program')

for program in programs:
    programID, orchestra, season = [program.xpath(_)[0].text for _ in ['programID', 'orchestra', 'season']]
    print (programID, orchestra, season)
    works = program.xpath('worksInfo/work')
    for work in works:
        workID = work.attrib['ID']
        soloistItems = work.xpath('soloists/soloist')
        for soloistItem in soloistItems:
            print (workID, soloistItem.find('soloistName').text, soloistItem.find('soloistInstrument').text, soloistItem.find('soloistRoles').text)

该脚本生成以下输出

3853 New York Philharmonic 1842-43
8834*4 Otto, Antoinette Soprano S
3642* Scharfenberg, William Piano A
3642* Hill, Ureli Corelli Violin A
3642* Derwort, G. H. Viola A
3642* Boucher, Alfred Cello A
3642* Rosier, F. W. Contrabass A
8835*1 Otto, Antoinette Soprano S
8835*1 Horn, Charles Edward Tenor S
8837*6 Horn, Charles Edward Tenor S
8336*4 Otto, Antoinette Soprano S

---

还有一件事需要注意：我在XML的开头放了一个标记，在末尾放了一个标记，因为实际数据将包含多个元素。

非常感谢！！！这正是我需要的。我对这一切都很陌生，我确实对循环的实际工作方式非常困惑。这是一个巨大的帮助，谢谢！！如果这个答案是最能解决您问题的答案，您应该通过按下左边的复选标记将其标记为接受

嗨，韦恩，我对它投了更高的票，但我的声誉不足15，因此没有记录：/但是您的答案非常有用！非常感谢你！！！这正是我需要的。我对这一切都很陌生，我确实对循环的实际工作方式非常困惑。这是一个巨大的帮助，谢谢！！如果这个答案是最能解决您问题的答案，您应该通过按下左边的复选标记将其标记为接受
嗨，韦恩，我对它投了更高的票，但我的声誉不足15，因此没有记录：/但是您的答案非常有用！