nastra commented on code in PR #6338:
URL: https://github.com/apache/iceberg/pull/6338#discussion_r1037875766
##########
core/src/main/java/org/apache/iceberg/jdbc/JdbcUtil.java:
##########
@@ -72,9 +72,9 @@ final class JdbcUtil {
+ TABLE_NAME
+ " VARCHAR(255) NOT NULL,"
+ METADATA_LOCATION
- + " VARCHAR(5500),"
+ + " VARCHAR(1000),"
Review Comment:
I've rerun some testing today to verify my findings from yesterday when
making this PR. It turns out adjusting `NAMESPACE_PROPERTY_KEY` from 5500 to
255 was the only thing that was necessary. I was mislead by the stack trace
```
at
com.mysql.jdbc.PreparedStatement.executeInternal(PreparedStatement.java:1903)
at com.mysql.jdbc.PreparedStatement.execute(PreparedStatement.java:1242)
at
org.apache.iceberg.jdbc.JdbcCatalog.lambda$initializeCatalogTables$1(JdbcCatalog.java:152)
at org.apache.iceberg.ClientPoolImpl.run(ClientPoolImpl.java:58)
at org.apache.iceberg.ClientPoolImpl.run(ClientPoolImpl.java:51)
at
org.apache.iceberg.jdbc.JdbcCatalog.initializeCatalogTables(JdbcCatalog.java:135)
at org.apache.iceberg.jdbc.JdbcCatalog.initialize(JdbcCatalog.java:103)
... 87 more
org.apache.iceberg.jdbc.UncheckedSQLException: Cannot initialize JDBC catalog
at org.apache.iceberg.jdbc.JdbcCatalog.initialize(JdbcCatalog.java:109)
at org.apache.iceberg.CatalogUtil.loadCatalog(CatalogUtil.java:212)
```
thinking that `JdbcCatalog.initializeCatalogTables(JdbcCatalog.java:135)`
is affecting the
[CREATE_TABLE_STATEMENT](https://github.com/apache/iceberg/blob/383d9caec804293e27b23b50c2cd89168346a848/core/src/main/java/org/apache/iceberg/jdbc/JdbcCatalog.java#L135)
while the issue was in all 3 issues the `CREATE_NAMESPACE_PROPERTIES_TABLE`
stmt.
So should we restore those varchars from 1000 back to 5500?
`Although InnoDB supports row sizes larger than 65,535 bytes internally,
MySQL itself imposes a row-size limit of 65,535 for the combined size of all
columns` <-- we could also try and set them to a higher value by calculating
(65535 - (255 * 3)) / 2 (metadata_location + previous_metadata_location). But
if we ever add another column, we'd have to re-calculate that.
To your second question around truncation. If a value is truncated, then
there will be an exception: `Data truncation: Data too long for column
'metadata_location' at row 1`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
