ENH: add dtype to read_xml

[MehdiALLALA]

Is your feature request related to a problem?

pandas column type detection is not precise. for some columns the type in not the actual data type when reading xml files with read_xml.

Describe the solution you'd like

the solution is to add dtype option in the read_xml utility that will provide pandas with the wanted column types, this feature exists already with read_json, read_csv and many other utilities.

API breaking implications

this will make the read_xml utility detect the right type for each column. to keep the data in the correct format.

Describe alternatives you've considered

creating a script that will use the default pandas.DataFrame(......) constructor that contains dtype option to parse the xml file.

the loaded result from the read_xml is not the same as the xml file. the highlighted column is a string but pandas read it as an integer and with this the data is corrupted.

[ParfaitG]

Thanks, @MehdiALLALA. The dtypes feature for read_xml was raised in a similar issue, #43567, and added to issue tracker, #40131.

[mroeschke]

Closing as a duplicate of #43567