#1263 DB copy from MS SQL to HSQLDB with NVARCHAR column leads to no dialect mapping error

[Flo]

It is not possible to copy a table with NVARCHAR field(s) to an HSQLDB database. During table creation in the target I get the error:

Copy failed: No Dialect mapping for JDBC type: -9 (NVARCHAR)

This probably also happens for NCHAR, NCLOB and LONGNVARCHAR

I have seen that the mapping is hard-coded in the CommonHibernateDialect implementations.
I fixed this for me by adding following line in the constructor of HSQLDialectExt:

registerColumnType(Types.NVARCHAR, "varchar");

I have mapped this to varchar because HSQLDB uses by default unicode encoding for character strings:

HyperSQL's default character set is Unicode, therefore all possible character strings can be represented by these types.
(http://hsqldb.org/doc/guide/guide.html)

Also the content of the tab data types displays this mapping:

TYPE_NAME DATA_TYPE PRECISION LITERAL_PREFIX LITERAL_SUFFIX CREATE_PARAMS
NVARCHAR 12 [VARCHAR] 2147483647 ' ' LENGTH

Perhaps it is possible to let the user configure this mapping in future.

[Gerd Wagner]

I just released a new snapshot that contains the changes you suggested, see:
https://sourceforge.net/projects/squirrel-sql/files/3-snapshots/snapshot-20160821_2336/

It would be great if you could give it a try and let me know if the changes work.

Thanks
Gerd

[Flo]

Hi Gerd,

Thanks. I have tested it and it works now.

Perhaps it makes sense to adapt the lines

registerColumnType(Types.NCHAR, "varchar");
registerColumnType(Types.LONGNVARCHAR, "varchar");

to

registerColumnType(Types.NCHAR, "char($l)");
registerColumnType(Types.LONGNVARCHAR, "longvarchar");

and add
registerColumnType(Types.NCLOB, "longvarchar");

Regards

Florian