Hi David,
If you read through Range::estimate_split_row(), you'll see that it should
return *false* (and subsequently throw Error::RANGESERVER_ROW_OVERFLOW) if
split_row == end_row, which makes the check for split_row.compare(end_row)
<= 0 ok. Also, the start_row is not included in the range, so the
split_row.compare(start_row)
> 0 is correct. If any of the keys in split_row_data is greater than the
end_row, then the assert would fire. I think I need to modify the check in
estimate_split_row() from if (row.compare(end_row) == 0) { to
if(row.compare(end_row) >= 0) {.
I think there might be a situation where the rows in split_row_data may
contain a row that is larger than end_row. I will get this change into
0.9.7.16 which will be released tomorrow. Give that version a try.
Hopefully it will resolve the problem.
- Doug
On Fri, Feb 14, 2014 at 12:44 AM, David <[email protected]> wrote:
> I read the Range::estimate_split_row function, and think that the
> expectation would become "split_row.compare(end_row)<0 &&
> split_row.compare(start_row) >= 0"
>
> --
> You received this message because you are subscribed to the Google Groups
> "Hypertable Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send email to [email protected].
> Visit this group at http://groups.google.com/group/hypertable-dev.
> For more options, visit https://groups.google.com/groups/opt_out.
>
--
Doug Judd
CEO, Hypertable Inc.
--
You received this message because you are subscribed to the Google Groups
"Hypertable Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/hypertable-dev.
For more options, visit https://groups.google.com/groups/opt_out.
