Recently I wanted to try the newly implemented xml_read function within pandas. I thought about testing the feature with SEPA camt-format xml. I'm stuck with the functions parameters, as I'm unfamiliar with the lxml logic. I tried pointing to the transactions values as rows ("Ntry" tag), as I thought this will then loop through those rows and creates the dataframe. Setting xpath to default returns an empty dataframe with the columns "GrpHdr" and "Rpt", but the relevant data is one level below "Rpt". Setting xpath='//*' creates a huge dataframe with every tag as column and values randomly sorted. If anyone is familiar with using the pandas xml_read and nested xmls, I'd appreciate any hints. The xml file looks like this (fake values):

Account
2021-08-05T14:20:23.077+02:00

 Name



Account ID
2021-08-05T14:20:23.077+02:00


DEXXXXX







161651651651
CRDT

2021-08-05T14:20:23.077+02:00



11465165
CRDT
BOOK

2021-08-02


2021-08-02



PMNT

RCDT
ESCT



NTRF+65454
DFE




4945141.0
CRDT


PMNT

RCDT
ESCT



NTRF+55155
DFEsds




Name


Name



Referenz NOTPROVIDED
 Buchug