[
https://issues.apache.org/jira/browse/CSV-138?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14186330#comment-14186330
]
Alan Stewart commented on CSV-138:
----------------------------------
You are probably correct. I was debugging in Intellij and the first call to
resultSet.getString(i) in the for loop on line 480 of CSVPrinter was be
evaluated as "2", being the primary key, which was in fact the 2nd record in
the table in a system test environment. It explained why my local integration
test kept failing with no records returned when I just created the one record
as test data. With the work around above, resultSet.getString(i) brought back
"1", which is what I was expecting.
Our system is so tied to SQLServer it will be difficult to test against another
database like Postgres or MySql. I will need to reproduce with a simple one
table test case and then change drivers to use against different databases.
> CSVPrinter.printRecords(ResultSet) skips first row with SQLServer
> -----------------------------------------------------------------
>
> Key: CSV-138
> URL: https://issues.apache.org/jira/browse/CSV-138
> Project: Commons CSV
> Issue Type: Bug
> Components: Printer
> Affects Versions: 1.0
> Reporter: Alan Stewart
>
> WIth SQLServer's 2012 jdbc driver 4.0, when I call CSVPrinter#printRecords
> with a java.sql.ResultSet, the first row is not printed. It appears that the
> line
> {code}
> final int columnCount = resultSet.getMetaData().getColumnCount();
> {code}
> positions the cursor on the first row and then when
> {code}
> while (resultSet.next()) { ...
> {code}
> is called, the 2nd row onwards is what is getting outputted.
> As a workaround, I used :
> {code}
> final int columnCount = rs.getMetaData().getColumnCount();
> do {
> for (int i = 1; i <= columnCount; i++) {
> csvPrinter.print(rs.getString(i));
> }
> csvPrinter.println();
> } while (rs.next());
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
