Java Streams-&gt;从列表中提取特定索引处的值 <List<String> &gt;

Question

I'm reading a CSV file with Java 8 streams and what is the best way to get the value of a particular column in CSV file For eg.

firstName,lastName,age,
tom,abraham,18,
liz,abraham,15,
tonny,paul,25

I want to extract the second column, so result set would be abraham;paul How can this be done by Java 8 lambda and Streams?

Answer 1

Welcome to StackOverflow :)

The CSV file could be read as any other text file, no parser is needed and the String::split with File::readAllLines is sufficient:

Set<String> names = Files.readAllLines(Paths.get("file.csv")) // Read all lines of file
                         .stream()                            // Stream them
                         .skip(1)                             // Omit the column names (if any)
                         .map(s -> s.split(";")[1])           // Split by ; and get the 2nd column
                         .collect(Collectors.toSet());        // Collect Strings to Set

I didn't read the label well. If you already have parsed List<List<String>> , the simplest way to achieve the very same result is:

Set<String> names = parsedList.stream()
                              .map(row -> row.get(1))       // Get the second column
                              .collect(Collectors.toSet())  // collect to Set<String>

Note the following:

• There is not needed the distinct() method since collecting to Set assures the distinct elements by definition. If you insist on collecting to List<String> , replace the last line with:

    .... .distinct() .collect(Collectors.toList()); 

  This way is valid for both of my suggested solutions.

• An exception might appear at the following lines in case of irregular CSV pattern or escaped ; characters:

  • map(s -> s.split(";")[1])
  • map(list -> list.get(1))

  Then you need to use a parser.

Answer 2

Files.readAllLines(Paths.get("my.csv")).stream()
        .skip(1)
        .map(s -> s.split(";")[1])
        .distinct()
        .collect(Collectors.toList()).forEach(System.out::println);

Is the simplest way, but I would prefer to use regular expression and the Matcher for this case.

Answer 3

You can use a map stream.

Files.lines(Paths.get("file path")).map(row -> row.split(";")).map(row -> row[1]).distinct().collect(Collectors.toList());

//first map: transform string to Array.
//Second map: choose first index array.
//distinct: remove duplicates elements.