The Python libxml XPath implementation is only XPath 1.0
Fails parsing fully both nodes.csv and fields.csv files
Example Fix:
cac:ProcurementLegislationDocumentReference[not(cbc:ID/text()=('CrossBorderLaw','LocalLegalBasis'))]
Modified to
cac:ProcurementLegislationDocumentReference[not(cbc:ID/text()='CrossBorderLaw')]