[ http://issues.apache.org/jira/browse/DERBY-797?page=all ]
Daniel John Debrunner resolved DERBY-797:
-----------------------------------------
Resolution: Invalid
Invalid as Stefan says, since InputStream.read() is not guaranteed to fill the
buffer.
> ResultSet.getBinaryStream() fails to read chunks of size > 32k
> --------------------------------------------------------------
>
> Key: DERBY-797
> URL: http://issues.apache.org/jira/browse/DERBY-797
> Project: Derby
> Type: Bug
> Components: JDBC
> Versions: 10.1.2.1
> Environment: Derby embedded engine
> Windows 2k
> JRE 1.4.2_03
> Reporter: Stefan Guggisberg
>
> assume the following table:
> create table TEST (TEST_ID integer not null, TEST_DATA blob not null);
> insert a record with a blob value larger than 32k, e.g. of size 100000
> read that record using a stmt like "select TEST_DATA from TEST where TEST_ID
> = ?"
> the following code fragment demonstrates the issue:
> InputStream in = resultSet.getBinaryStream(1);
> byte[] buf = new byte[33000];
> int n = in.read(buf);
> ==> n == 32668, i.e. < buf.length !
> the problem occurs with all chunked reads that cross the boundary at offset
> 32668, e.g.
> InputStream in = resultSet.getBinaryStream(1);
> byte[] buf 1= new byte[32660];
> int n = in.read(buf1);
> // ok, n == buf1.length
> byte[] buf 2= new byte[20];
> n = in.read(buf2);
> // n == 8, i.e. < buf2.length !
> workarounds for this bug:
> - read byte by byte i.e. using in.read()
> - use resultSet.getBytes()
> the faulty code seems to be in
> org.apache.derby.impl.store.raw.data.MemByteHolder
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira
