0

I am using SuperCsv to process contact csv files from different sources. Number of columns is the same and there is a header in file so I want to use the CsvBeanReader.

Has different sources have different columns and header titles, I am building dynamically the cellProcessors array based on the number of columns identified in the header.

I was struggling for a few hours with a SuperCsvException telling me there was a mismatch between the number of processors and some particular files which happen to all be csv exports from google mail contacts applications before I noticed these files had datarows ending with a useless comma where has the header row has not.

I solved the problem by catching the first SuperCsvException and adding the extra cell processor at this time but I was wondering whether this last comma was present in other types of csv files and whether superCsv had any option that could allow to keep the power of CsvBeanReader allowing for this last comma flexibility.

Carlo V. Dango
  • 13,322
  • 16
  • 71
  • 114
Yves Nicolas
  • 6,901
  • 7
  • 25
  • 40

1 Answers1

0

I would consider using the CsvListReader.Read() to get a list of string values. If you then by the length of the list know what to do, you can apply an array of processors using the Util.executeCellProcessors() which takes as input the list of strings and the cellprocessors.

Carlo V. Dango
  • 13,322
  • 16
  • 71
  • 114