Heavily nested XML to CSV using Python

Question

I have following XML file which I want to convert to CSV using Python.

<?xml version="1.0" encoding="UTF-8"?><households xmlns:s="http://www.mediametrie.fr/nge/  " xmlns:xalan="http://xml.apache.org/xalan" date="2015-04-06" creation_date="2015-04-08T03:48:34">
    <household id="10003456">
        <destinations/>
        <members>
            <member id="1">
                <member_process result="KO" vacation="undefined">
                    <individual_audience>
                        <individual_audience_tvset id="1">
                            <channel session="5647128" begin="56435" end="76896"/>
                        </individual_audience_tvset>
                    </individual_audience>
                    <alarms>
                        <alarm id="Alarm_id_1" rule_id="Rule_id_1">
                            <parameters>
                                <parameter name="tvset_id" value="1"/>
                                <parameter name="length" value="46384"/>
                                <parameter name="end" value="2017-04-06T20:30:00"/>
                                <parameter name="channel" value="1010128"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </member_process>
            </member>
            <member id="2">
                <member_process result="KO" vacation="undefined">
                    <individual_audience>
                        <individual_audience_tvset id="1">
                            <channel session="5674897" begin="98765" end="76543"/>
                        </individual_audience_tvset>
                    </individual_audience>
                    <alarms>
                        <alarm id="Alarm_id_2" rule_id="Rule_id_2">
                            <parameters>
                                <parameter name="tvset_id" value="1"/>
                                <parameter name="length" value="56745"/>
                                <parameter name="end" value="2017-04-06T20:30:00"/>
                                <parameter name="channel" value="4563256"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </member_process>
            </member>
            <member id="3">
                <member_process result="KO" vacation="undefined">
                    <individual_audience>
                        <individual_audience_tvset id="1">
                            <channel session="1010128" begin="47218" end="93600"/>
                        </individual_audience_tvset>
                    </individual_audience>
                    <alarms>
                        <alarm id="AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT" rule_id="R_INDP_AUDIENCE_TOO_HIGH_LIMIT">
                            <parameters>
                                <parameter name="tvset_id" value="1"/>
                                <parameter name="length" value="46382"/>
                                <parameter name="end" value="2015-04-06T20:30:00"/>
                                <parameter name="channel" value="1010128"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </member_process>
            </member>
            <member id="4">
                <member_process result="KO" vacation="undefined">
                    <individual_audience>
                        <individual_audience_tvset id="1">
                            <channel session="1010128" begin="47219" end="93600"/>
                        </individual_audience_tvset>
                    </individual_audience>
                    <alarms>
                        <alarm id="AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT" rule_id="R_INDP_AUDIENCE_TOO_HIGH_LIMIT">
                            <parameters>
                                <parameter name="tvset_id" value="1"/>
                                <parameter name="length" value="46381"/>
                                <parameter name="end" value="2015-04-06T20:30:00"/>
                                <parameter name="channel" value="1010128"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </member_process>
            </member>
            <member id="5">
                <member_process result="KO" vacation="undefined">
                    <individual_audience>
                        <individual_audience_tvset id="1">
                            <channel session="1010128" begin="47220" end="93600"/>
                        </individual_audience_tvset>
                    </individual_audience>
                    <alarms>
                        <alarm id="AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT" rule_id="R_INDP_AUDIENCE_TOO_HIGH_LIMIT">
                            <parameters>
                                <parameter name="tvset_id" value="1"/>
                                <parameter name="length" value="46380"/>
                                <parameter name="end" value="2015-04-06T20:30:00"/>
                                <parameter name="channel" value="1010128"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </member_process>
            </member>
            <member id="6">
                <member_process result="KO" vacation="undefined">
                    <individual_audience>
                        <individual_audience_tvset id="1">
                            <channel session="1010128" begin="47221" end="93600"/>
                        </individual_audience_tvset>
                    </individual_audience>
                    <alarms>
                        <alarm id="AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT" rule_id="R_INDP_AUDIENCE_TOO_HIGH_LIMIT">
                            <parameters>
                                <parameter name="tvset_id" value="1"/>
                                <parameter name="length" value="46379"/>
                                <parameter name="end" value="2015-04-06T20:30:00"/>
                                <parameter name="channel" value="1010128"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </member_process>
            </member>
        </members>
        <regular_guests/>
        <occasional_guests/>
        <tvsets>
            <tvset id="1">
                <tvset_process result="OK">
                    <tvset_audience>
                        <channel session="47" begin="46304" end="46384"/>
                        <channel session="1010483" begin="46384" end="46419"/>
                        <channel session="235" begin="46419" end="46424"/>
                        <channel session="1010128" begin="46424" end="93600"/>
                    </tvset_audience>
                    <alarms>
                        <alarm id="AL_T_P_VALID_LAST_HOUR_REBOOT" rule_id="T_P_METER_STOPPING_TIMESTAMPING">
                            <parameters>
                                <parameter name="unique_id" value="4547"/>
                                <parameter name="reboot_date" value="2015-04-06T07:17:44"/>
                                <parameter name="length" value="1.6221180555555557"/>
                            </parameters>
                        </alarm>
                        <alarm id="AL_T_P_VALID_LAST_HOUR_REBOOT" rule_id="T_P_METER_STOPPING_TIMESTAMPING">
                            <parameters>
                                <parameter name="unique_id" value="4566"/>
                                <parameter name="reboot_date" value="2015-04-07T13:17:54"/>
                                <parameter name="length" value="1.2313657407407406"/>
                            </parameters>
                        </alarm>
                        <alarm id="AL_T_P_TECH_ID_RESOL_FALSE_POSITIVE" rule_id="T_P_TECH_ID_RESOL">
                            <parameters>
                                <parameter name="channel_id" value="194"/>
                                <parameter name="unique_id" value="4549"/>
                            </parameters>
                        </alarm>
                    </alarms>
                </tvset_process>
            </tvset>
        </tvsets>
        <household_process result="KO" vacation="no">
            <alarms>
                <alarm id="AL_T_FP_AUDIENCE_WITHOUT_PRESENCE" rule_id="T_FP_AUDIENCE_WITHOUT_PRESENCE">
                    <parameters>
                        <parameter name="tvset_id" value="1"/>
                        <parameter name="length" value="80"/>
                        <parameter name="start" value="2015-04-06T07:21:44"/>
                    </parameters>
                </alarm>
                <alarm id="AL_T_FP_AUDIENCE_WITHOUT_PRESENCE" rule_id="T_FP_AUDIENCE_WITHOUT_PRESENCE">
                    <parameters>
                        <parameter name="tvset_id" value="1"/>
                        <parameter name="length" value="792"/>
                        <parameter name="start" value="2015-04-06T07:23:44"/>
                    </parameters>
                </alarm>
                <alarm id="AL_R_FP_AUDIENCE_TOO_HIGH_LIMIT" rule_id="R_FP_AUDIENCE_TOO_HIGH_LIMIT">
                    <parameters>
                        <parameter name="tvset_id" value="1"/>
                        <parameter name="length" value="47176"/>
                        <parameter name="end" value="2015-04-06T20:30:00"/>
                        <parameter name="channel" value="1010128"/>
                    </parameters>
                </alarm>
                <alarm id="AL_R_FP_AT_LEAST_ONE_MEMBER_OK" rule_id="R_FP_AT_LEAST_ONE_MEMBER_OK">
                    <parameters/>
                </alarm>
            </alarms>
        </household_process>
    </household>
</households>

Output should look something like this

household id, destinations, member id, result, vacation, individual_audience_tvset id, session, begin, end, alarm id, rule_id, name, value
10003456, None, 1, KO, undefined, 1, 5647128, 56435, 76896, Alarm_id_1, Rule_id_1, tvset_id, 1
10003456, None, 1, KO, undefined, 1, 5647128, 56435, 76896, Alarm_id_1, Rule_id_1, length, 46384
10003456, None, 1, KO, undefined, 1, 5647128, 56435, 76896, Alarm_id_1, Rule_id_1, end, 2017-04-06T20:30:00
10003456, None, 1, KO, undefined, 1, 5647128, 56435, 76896, Alarm_id_1, Rule_id_1, channel, 1010128

Similarly for member id = 2 with same household id.

Any help is highly appreciated. Thank you in advance!

Use Beautiful Soup for parsing the XML and the standard library for writing out the CSV. — chb
– chb, Commented Mar 25, 2018 at 8:40
Yeah I have tried, regular for looping but it print unexpected nested XML results. — Siddhesh Chavan
– Siddhesh Chavan, Commented Mar 25, 2018 at 10:59
Can you update your question with the code you tried, so we can build on what you've got? — Tim
– Tim, Commented Mar 26, 2018 at 11:06

Martin Evans · Accepted Answer · 2018-03-26 16:59:14Z

2

This assumes you have your XML in a file called input.xml. BeautifulSoup can be used to help with parsing the XML read from the file. You then just need to create a table containing all the information that you wish to extract:

from bs4 import BeautifulSoup
import csv

fields = [
    "household id",
    "destinations",
    "member id"]

member_fields = [    
    ["result", "member_process", "result"],
    ["vacation", "member_process", "vacation"],
    ["individual_audience_tvset id", "individual_audience_tvset", "id"],
    ["session", "channel", "session"],
    ["begin", "channel", "begin"],
    ["end", "channel", "end"],
    ["alarm id", "alarm", "id"],
    ["rule_id", "alarm", "rule_id"],
    ["name", "parameter", "name"],
    ["value", "parameter", "value"]
    ]

fieldnames = fields + [field for field, _, _ in member_fields]    

with open('input.xml') as f_input, open('output.csv', 'w', newline='') as f_output:
    csv_output = csv.DictWriter(f_output, fieldnames=fieldnames)
    csv_output.writeheader()

    xml = f_input.read()
    soup = BeautifulSoup(xml, "xml")
    household_id = soup.find('household')['id']

    for member in soup.find_all('member'):
        member_id = member['id']
        row = {'household id' : household_id, 'member id' : member_id}

        for field, x, y in member_fields:
            row[field] = member.find(x)[y]

        csv_output.writerow(row)

Which would create output.csv containing:

household id,destinations,member id,result,vacation,individual_audience_tvset id,session,begin,end,alarm id,rule_id,name,value
10003456,,1,KO,undefined,1,5647128,56435,76896,Alarm_id_1,Rule_id_1,tvset_id,1
10003456,,2,KO,undefined,1,5674897,98765,76543,Alarm_id_2,Rule_id_2,tvset_id,1
10003456,,3,KO,undefined,1,1010128,47218,93600,AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT,R_INDP_AUDIENCE_TOO_HIGH_LIMIT,tvset_id,1
10003456,,4,KO,undefined,1,1010128,47219,93600,AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT,R_INDP_AUDIENCE_TOO_HIGH_LIMIT,tvset_id,1
10003456,,5,KO,undefined,1,1010128,47220,93600,AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT,R_INDP_AUDIENCE_TOO_HIGH_LIMIT,tvset_id,1
10003456,,6,KO,undefined,1,1010128,47221,93600,AL_R_INDP_AUDIENCE_TOO_HIGH_LIMIT,R_INDP_AUDIENCE_TOO_HIGH_LIMIT,tvset_id,1

answered Mar 26, 2018 at 16:59

Martin Evans

46.9k17 gold badges88 silver badges104 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

user12325998 Over a year ago

very good asnwer. I tried to implement your question to my xml file but with no success until now even though my xml is similar. If you have time please have a look at my xml file stackoverflow.com/questions/59009797/…

SPJ Over a year ago

This remains an excellent solution. I have searched through the stack over the past few days and this is the easiest solution for complex nested XML

Collectives™ on Stack Overflow

Heavily nested XML to CSV using Python

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related