Hans-Martin Adorf

Hi Scott, I finally know, why I could not make myself understood so far. Lets first look into the JAXB-world. My mental model is embodied in the attached JUnit test and the XML-file that is being read (unmarhalled) in the test. As you know, in the JAXB-context there is no distinction between blanks and other characters. A string can have zero or more characters (or blanks). However, there is a distinction between an empty string and null. In the XML-file the fact that an attribute has no value (null)...

Hi Scott, thanks for your message. I was unaware of this option. However, I am not really interested in null values. I included null values in the discussion only because I wanted to discuss the general case. What I am interested in is being able to put strings with zero or more blanks into the CSV file, and to find in the corresponding bean the string of blanks in the corresponding Java variable for that column. All I need is a recipe how to get that done using OpenCSV. Your examples show a variety...

Hi Scott, thank you once more for devoting your time to the problem that I reported. I should like to restructure the discussion. For the moment let us put aside the question of whether the field in the bean is required to have a value. The salient quesion is how to you get OpenCSV to parse a CSV field that has zero or more blanks and what the result should be. This is a semantic question. The problem arises, because of the weak (sick?) definition of the CSV format. Let us consider a suitably (JAXB)...

Hi Scott, thank you very much for devoting so much time to the problem that I reported. Before getting back to the problem I should like to also thank you for the great work you are doing. I had used OpenCSV a while ago, and I was positively suprised to discover that one can now parse a CSV-File directly into a bean. Here is my use case: I have a text document in which I want to replace certain strings by others. Originally the strings to be searched for where abbreviations, which had to be replaced...

The attribute "required" means the value must be non-null. The empty string (i.e. a string that contains no characters is non-null, and should be allowed in the CSV file and imho should be parsed properly.

Additional files

Problem when parsing a CSV file with empty entries

Problem when parsing a CSV file with empty entries