如何从xml或tcx文件中获取数据系列

I want to treat data from .tcx file (xml form) between specific tags with Python.
File format is like as follows.

我想用Python处理特定标签之间的.tcx文件（xml格式）数据。文件格式如下。

 <Track>
      <Trackpoint>
        <Time>2015-08-29T22:04:39.000Z</Time>
        <Position>
          <LatitudeDegrees>37.198049426078796</LatitudeDegrees>
          <LongitudeDegrees>127.07204628735781</LongitudeDegrees>
        </Position>
        <AltitudeMeters>34.79999923706055</AltitudeMeters>
        <DistanceMeters>7.309999942779541</DistanceMeters>
        <HeartRateBpm>
          <Value>102</Value>
        </HeartRateBpm>
        <Cadence>76</Cadence>
        <Extensions>
          <TPX xmlns="http://www.garmin.com/xmlschemas/ActivityExtension/v2">
            <Watts>112</Watts>
          </TPX>
        </Extensions>
      </Trackpoint>
....Lots of <Trackpoint> ... </Trackpoint>
</Track>

Eventually, I'll make Data table with columns of 'Lattitude, Altitude, ... Watts'.
First I tried to make a list from taged data (like Watts ... /Watts) with BeautifulSoup, xpath etc. But I'm a newbie to deal with these tools. How can I grab data between tags in xml file with Python?

最后，我将使用“Lattitude，Altitude，... Watts”列创建数据表。首先，我尝试使用BeautifulSoup，xpath等从托管数据（如Watts ... / Watts）制作一个列表。但我是处理这些工具的新手。如何使用Python在xml文件中的标签之间获取数据？

3 个解决方案

#1

You could use the lxml module, along with XPath. lxml is good for parsing XML/HTML, traversing element trees and returning element text/attributes. You can select particular elements, sets of elements or attributes of elements using XPath. Using your example data:

您可以使用lxml模块和XPath。 lxml适用于解析XML / HTML，遍历元素树和返回元素文本/属性。您可以使用XPath选择特定元素，元素集或元素属性。使用您的示例数据：

content = '''
<Track>
      <Trackpoint>
        <Time>2015-08-29T22:04:39.000Z</Time>
        <Position>
          <LatitudeDegrees>37.198049426078796</LatitudeDegrees>
          <LongitudeDegrees>127.07204628735781</LongitudeDegrees>
        </Position>
        <AltitudeMeters>34.79999923706055</AltitudeMeters>
        <DistanceMeters>7.309999942779541</DistanceMeters>
        <HeartRateBpm>
          <Value>102</Value>
        </HeartRateBpm>
        <Cadence>76</Cadence>
        <Extensions>
          <TPX xmlns="http://www.garmin.com/xmlschemas/ActivityExtension/v2">
            <Watts>112</Watts>
          </TPX>
        </Extensions>
      </Trackpoint>
....Lots of <Trackpoint> ... </Trackpoint>
</Track>
'''

from lxml import etree

tree = etree.XML(content)
time = tree.xpath('Trackpoint/Time/text()')

print(time)

Output

产量

['2015-08-29T22:04:39.000Z']

#2

You can even use lxml module to convert XML to CSV (for later import into a dataframe, spreadsheet, or database table) using an iterated Python list across various XPaths.

您甚至可以使用lxml模块将XML转换为CSV（以便以后导入到数据框，电子表格或数据库表中），使用跨各种XPath的迭代Python列表。

Notice the very last Watts node is a special, longer XPath due escaping the special namespace, xlmns not registered in sample XML.

请注意，最后一个Watts节点是一个特殊的，更长的XPath，因为它逃避了特殊的命名空间，xlmns没有在样本XML中注册。

import os, csv
import lxml.etree as ET

# SET DIRECTORY
cd = os.path.dirname(os.path.abspath(__file__))

# LOAD XML FILE
xmlfile = 'trackXML.xml'
dom = ET.parse(os.path.join(cd, xmlfile))

# DEFINING COLUMNS
columns = ['latitude', 'longitude', 'altitude', 'distance', 'watts']

# OPEN CSV FILE
with open(os.path.join(cd,'trackData.csv'), 'w') as m:
    writer = csv.writer(m)    
    writer.writerow(columns)

    nodexpath = dom.xpath('//Trackpoint')

    dataline = []   # FOR ONE-ROW CSV APPENDS
    datalines = []  # FOR FINAL OUTPUT 
    for j in range(1,len(nodexpath)+1):        
        dataline = []

        # LOCATE PATH OF EACH NODE VALUE
        latitudexpath = dom.xpath('//Trackpoint[{0}]/Position/LatitudeDegrees/text()'.format(j))
        dataline.append('') if latitudexpath == [] else dataline.append(latitudexpath[0])

        longitudexpath = dom.xpath('//Trackpoint[{0}]/Position/LongitudeDegrees/text()'.format(j))
        dataline.append('') if longitudexpath == [] else dataline.append(longitudexpath[0])

        altitudexpath = dom.xpath('//Trackpoint[{0}]/AltitudeMeters/text()'.format(j))
        dataline.append('') if altitudexpath == [] else dataline.append(altitudexpath[0])

        distancexpath = dom.xpath('//Trackpoint[{0}]/DistanceMeters/text()'.format(j))
        dataline.append('') if distancexpath == [] else dataline.append(distancexpath[0])

        wattsxpath = dom.xpath("//Trackpoint[{0}]/*[name()='Extensions']/*[name()='TPX']/*[name()='Watts']/text()".format(j))
        dataline.append('') if wattsxpath == [] else dataline.append(wattsxpath[0])

        datalines.append(dataline)
        writer.writerow(dataline)

print(datalines)

In addition to CSV file, below is the datalines list output of selected columns:

除CSV文件外，以下是所选列的数据表列表输出：

[['37.198049426078796', '127.07204628735781', '34.79999923706055', '7.309999942779541', '112']]

#3

The Python program https://github.com/cast42/vpower/blob/master/vpower.py iterates over the TCX file specified at the command line and add a power field for all measurements of the cycling activity. It uses the lxml library for speed and because it deals with namespaces. In previous versions of this program I used xml.etree.ElementTree but ran into problems with the namespaces.

Python程序https://github.com/cast42/vpower/blob/master/vpower.py遍历命令行中指定的TCX文件，并为循环活动的所有测量添加一个功率字段。它使用lxml库来提高速度，因为它处理命名空间。在此程序的先前版本中，我使用了xml.etree.ElementTree但遇到了名称空间的问题。