Hi.
We had a problem with OCILobRead when reading a multibyte LOB buffer (Korean).
We have found a workaround that seems to solve this problem, but is somewhat awkward. The workaround is to call OCILobRead with csid parameter set to 1, instead of 0, which explicitly sets a charset ID to be used for the returned buffer.
Our questions are:
1. Is this an appropriate solution?
2. Is there a better approach?
3. What influence will have not using NLS_LANG on the client side?
Thanks.