CsvFileReader#readAllAsSequence fails on non-equal sized rows

[koral--]

Describe the bug
If rows are not equal-sized an exception is thrown:
com.github.doyaaaaaken.kotlincsv.util.CSVFieldNumDifferentException: Fields num seems to be 4 on each row, but on 2th csv row, fields num is 3.

To Reproduce

        csvWriter().open(csvFile) {
            writeRow(listOf("a"))
            writeRow(listOf("a", "b"))
        }
        csvReader().open(csvFile) {
           readAllAsSequence()
        }

Expected behavior
Missing cells are treated as nulls or empty strings.

Environment

• kotlin-csv version 0.11.0
• OS: Android 10

Screenshots
N/A

[doyaaaaaken]

@koral--
Thanks for reporting!

Throwing exception is expected behavior.
If you change behavior, the skipEmptyLine option on CsvReader is available.
See https://github.com/doyaaaaaken/kotlin-csv#customize

[doyaaaaaken]

Any other question on this topic? Feel free to ask.
If not, I'll close this issue.

[koral--]

I don't longer need this library, but can help resolve this issue if needed.

I don't understand your explanation. Could you elaborate?
Is it really intended that CSV written by this library is not readable by itself?

[doyaaaaaken]

The "CSV" definition is defined on RFC4180.
It says here.

Within the header and each record, there may be one or more fields, separated by commas. Each line should contain the same number of fields throughout the file.

So, your sample code doesn't write "CSV" file, so it's not unnatural that kotlin-csv cannot read your writing file.

[koral--]

OK, thanks. Now I understand the explanation.
One little question then: is it intended that library allows writing a file which does not conform to CSV format?
In this particular case the 2nd #writeRow() call may throw an exception (or it can be thrown when closing file, technical details are not important now).

[doyaaaaaken]

One little question then: is it intended that library allows writing a file which does not conform to CSV format?

Yes, it's intended.

In this particular case the 2nd #writeRow() call may throw an exception (or it can be thrown when closing file, technical details are not important now).

Yes, it should be. To throw exception on the 2nd writeRow() call is useful.
At this point, I don't plan to implement this validation by myself, but pull requests are welcomed. 👍

[koral--]

OK so I guess that this issue can be rephrased to something like "Fail fast on writing malformed CSV" and tagged as enhancement., right?

[doyaaaaaken]

Yeah, right. Thank you.
Maybe, it'd be better to create another issue, relate this issue on that, and close this issue.