you might also try the blow option.

1). Use a different file format: You can try using a different file format that supports multi-character delimiters, such as text JSON.

2). Use a custom Row class: You can write a custom Row class to parse the multi-character delimiter yourself, and then use the spark.read.text API to read the file as text. You will then need to apply the custom Row class to each line in the text file to extract the values