Optional fields in CSV parser-CodePudding

I am writing code to read a CSV file.

A, B, C
1, 1, 1
2, 2, 2

Sometimes column C will be present, and sometimes it will not be. If it's not, I really don't care, but if it is I want to handle it. I'm also using the Univocity parsers, but don't necessarily have to.

CsvParserSettings settings = new CsvParserSettings();
settings.getFormat().setLineSeparator("\n");
settings.setHeaderExtractionEnabled(true);

CsvParser parser = new CsvParser(settings);
parser.beginParsing(file);

Record record;
parser.getRecordMetadata();

while ((record = parser.parseNextRecord()) != null) {
   String a = record.getString("A");
   long b = record.getLong("B");
   long c = record.getLong("C");
}

I can do something like

long c = 1;
try{
  c = record.getLong("C");
} catch (Exception e) {}

But this seems needlessly complex. Is there some kind of setting that makes a field entirely optional?

CodePudding user response：

Record contains a getValues() method that returns all values. You could use this to check the number of values to avoid an exception.

if( record.getValues().length > 2 )
  long c = record.getLong("C");

Untested, I just glanced at the API docs.

(The Apache CSV parser has a CSVRecord with a size() method you can test directly, with less (I assume) overhead. In case you might be interested in switching.)

CodePudding user response：

Maybe you could use the isMapped method from CSVrecord

if(record.isMapped("C")){
 c = record.getLong("C");
} else ...