[
https://issues.apache.org/jira/browse/DDLUTILS-165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12484047
]
Stefan Huber commented on DDLUTILS-165:
---------------------------------------
The patch works fine with LONGVARCHAR values.
But with LONGVARBINARY fields i have the same effect. So if you add a type
conversion for LONGVARBINARY fields (analog to LONGVARCHAR fields) i would
classify this bug as resolved.
// fix issue DDLUTILS-165 as postgresql-8.2-504-jdbc3.jar seems to
return Integer.MAX_VALUE
// on columns defined as TEXT.
else if (column.getSizeAsInt() == Integer.MAX_VALUE)
{
column.setSize(null);
if (column.getTypeCode() == Types.VARCHAR)
{
column.setTypeCode(Types.LONGVARCHAR);
}
+ else if (column.getTypeCode() == Types.BINARY)
+ {
+ column.setTypeCode(Types.LONGVARBINARY);
+ }
}
> Data type mismatch in PostgreSQL with TEXT columns
> --------------------------------------------------
>
> Key: DDLUTILS-165
> URL: https://issues.apache.org/jira/browse/DDLUTILS-165
> Project: DdlUtils
> Issue Type: Bug
> Components: Core - PostgreSql
> Environment: PostgreSQL v2.8.3 on Windows with JDBC driver
> 'postgresql-8.2-504.jdbc3.jar'
> Reporter: Stefan Huber
> Assigned To: Thomas Dudziak
> Fix For: 1.0
>
> Attachments: patch.txt
>
>
> The PostgreSql jdbc driver returns VARCHAR(2147483647) for columns defined as
> TEXT. The ModelReader expects VARCHAR(-1) for these data type.
> Maybe the behaviour of the jdbc driver changed lately, because there is a
> comment in the PostgreSqlModelReader.readColumn method which says that
> PostgreSQL returns VARCHAR(-1), but in my environment i get
> VARCHAR(2147483647).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
