BUG: iterparse on read_xml overwrites nested child elements

[bama-chi]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd

XML ='''
   <values>
       <guidedSaleKey>
           <code>9023000918982</code>
           <externalReference>0102350511</externalReference>
       </guidedSaleKey>
       <store>
           <code>02300</code>
           <externalReference>1543</externalReference>
           <currency>EUR</currency>
       </store>
   </values>
'''

df = pd.read_xml(XML,iterparse={"values":["code","code"]}, names=["guided_code","store_code"])
print(df)

Issue Description

dataframe will not be able to return value of both code elements from guidedSaleKey and store
this will return this:

     guided_code     store_code
0  9023000918982  9023000918982

Expected Behavior

this should return a dataframe like this:

     guided_code     store_code
0  9023000918982  2300

Installed Versions

INSTALLED VERSIONS

commit : 87cfe4e
python : 3.10.4.final.0
python-bits : 64
OS : Linux
OS-release : 5.4.0-137-generic
Version : #154-Ubuntu SMP Thu Jan 5 17:03:22 UTC 2023
machine : x86_64
processor : x86_64
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 1.5.0
numpy : 1.23.4
pytz : 2022.5
dateutil : 2.8.2
setuptools : 58.1.0
pip : 22.0.4
Cython : None
pytest : 7.1.3
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : 4.9.2
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : None
pandas_datareader: None
bs4 : None
bottleneck : None
brotli : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : None
numba : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : None
snappy :
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
xlwt : None
zstandard : None
tzdata : None

[bama-chi]

there was quite same bug with attributes on this issue

[ParfaitG]

I am not sure this is a bug. Per read_xml docs, the dictionary key in iterparse is intended to be the repeating element which serializes as rows in DataFrame.

Here you are using the root which never repeats in XML. Conceptually, guidedSaleKey and store are two different sets that can repeat in varying numbers. You can run read_xml separately then join DataFrames if their repetition matches.

[ParfaitG]

However, your issue may relate to an iterparse issue regarding same-named elements which I recently observed on this SO post:

from io import BytesIO
import pandas as pd

xml = b'''\
<?xml version="1.0" encoding="UTF-8"?>
<histdata>
   <item>
    <datetime>20-1-2023 00:00:00</datetime>
    <value channel="Traffic Total" chid="-1">92 kbit/s</value>
    <value channel="Traffic In" chid="0">77 kbit/s</value>
    <value channel="Traffic Out" chid="1">16 kbit/s</value>
    <value channel="Downtime" chid="-4">0 %</value>
    <coverage>100 %</coverage>
   </item>
   <item>
    <datetime>20-1-2023 00:05:00</datetime>
    <value channel="Traffic Total" chid="-1">82 kbit/s</value>
    <value channel="Traffic In" chid="0">727 kbit/s</value>
    <value channel="Traffic Out" chid="1">18 kbit/s</value>
    <value channel="Downtime" chid="-4">0 %</value>
    <coverage>100 %</coverage>
   </item>
</histdata>
'''

with BytesIO(xml) as f:            # only latest pandas 2.0 dev supports file-like objects for iterparse
    items_df = pd.read_xml(
        f, 
        iterparse={"item": [
            "datetime", "value", "value", "value", "value", "coverage"
        ]},
        names = [
            "datetime", "value1", "value2",  "value3", "value4",  "coverage"
        ]
    )

items_df
#              datetime     value1     value2     value3     value4 coverage
# 0  20-1-2023 00:00:00  92 kbit/s  92 kbit/s  92 kbit/s  92 kbit/s    100 %
# 1  20-1-2023 00:05:00  82 kbit/s  82 kbit/s  82 kbit/s  82 kbit/s    100 %

As you can see the first <value> element text repeats for all other <value> instances. Interestingly, the repetitious result is not observed when only parsing the repeating attributes suggesting a bug:

with BytesIO(xml) as f:
    attribs_df = pd.read_xml(
        f, 
        iterparse={"item": [
            "datetime",  "channel", "chid", "channel", "chid",  "channel", "chid", "channel", "chid", "coverage"
        ]},
        names = [
            "datetime", "channe11", "chid1",  "channel2", "chid2", "channel3", "chid3", "channel4", "chid4", "coverage"
        ]
    )

attribs_df
#              datetime       channe11  chid1    channel2  chid2     channel3  chid3  channel4  chid4 coverage
# 0  20-1-2023 00:00:00  Traffic Total     -1  Traffic In      0  Traffic Out      1  Downtime     -4    100 %
# 1  20-1-2023 00:05:00  Traffic Total     -1  Traffic In      0  Traffic Out      1  Downtime     -4    100 %