Un servicio de Azure para la ingesta, preparación y transformación de datos a gran escala.
The final solution was to leave the datasource schema empty and set Allow schema drift = true in the pipeline source. After applying this change, the pipeline was able to process the source file correctly and load the data. This indicates that the issue was related to schema interpretation or compatibility within Azure Data Factory, and that allowing the source to handle schema changes dynamically resolved the problem without requiring changes to the source file itself.