Is there a way to get a IndexInput with a larger buffer "sometimes"?
Since the fields/documents and terms processing are basically sequential, it would seem that merging would benefit from a much larger buffer size.
The recent changes alleviate this a bit for document merging, since the document will be read as an entire byte array, but is probably only helpful for large documents.
This change would work well in all cases. Thoughts? --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
