Range
(
mapping
=
None
,
*
,
ignore_unknown_fields
=
False
,
**
kwargs
)
A Range
represents a range of keys in a database. The keys
themselves are encoded in "sortable string format", also known as
ssformat. Consult Spanner's open source client libraries for details
on the encoding.
Each range represents a contiguous range of rows, possibly from
multiple tables/indexes. Each range is associated with a single
paxos group (known as a "group" throughout this API), a split (which
names the exact range within the group), and a generation that can
be used to determine whether a given Range
represents a newer or
older location for the key range.
Attributes
Name
Description
start_key
bytes
The start key of the range, inclusive. Encoded in "sortable string format" (ssformat).
limit_key
bytes
The limit key of the range, exclusive. Encoded in "sortable string format" (ssformat).
group_uid
int
The UID of the paxos group where this range is stored. UIDs are unique within the database. References
Group.group_uid
.split_id
int
A group can store multiple ranges of keys. Each key range is named by an ID (the split ID). Within a group, split IDs are unique. The
split_id
names the exact split in group_uid
where this range is stored.generation
bytes
generation
indicates the freshness of the range
information contained in this proto. Generations can be
compared lexicographically; if generation A is greater than
generation B, then the Range
corresponding to A is newer
than the Range
corresponding to B, and should be used
preferentially.