| 1 |
|
| 2 |
This directory contains an SQLite extension that implements a virtual |
| 3 |
table type that allows users to create, query and manipulate r-tree[1] |
| 4 |
data structures inside of SQLite databases. Users create, populate |
| 5 |
and query r-tree structures using ordinary SQL statements. |
| 6 |
|
| 7 |
1. SQL Interface |
| 8 |
|
| 9 |
1.1 Table Creation |
| 10 |
1.2 Data Manipulation |
| 11 |
1.3 Data Querying |
| 12 |
1.4 Introspection and Analysis |
| 13 |
|
| 14 |
2. Compilation and Deployment |
| 15 |
|
| 16 |
3. References |
| 17 |
|
| 18 |
|
| 19 |
1. SQL INTERFACE |
| 20 |
|
| 21 |
1.1 Table Creation. |
| 22 |
|
| 23 |
All r-tree virtual tables have an odd number of columns between |
| 24 |
3 and 11. Unlike regular SQLite tables, r-tree tables are strongly |
| 25 |
typed. |
| 26 |
|
| 27 |
The leftmost column is always the pimary key and contains 64-bit |
| 28 |
integer values. Each subsequent column contains a 32-bit real |
| 29 |
value. For each pair of real values, the first (leftmost) must be |
| 30 |
less than or equal to the second. R-tree tables may be |
| 31 |
constructed using the following syntax: |
| 32 |
|
| 33 |
CREATE VIRTUAL TABLE <name> USING rtree(<column-names>) |
| 34 |
|
| 35 |
For example: |
| 36 |
|
| 37 |
CREATE VIRTUAL TABLE boxes USING rtree(boxno, xmin, xmax, ymin, ymax); |
| 38 |
INSERT INTO boxes VALUES(1, 1.0, 3.0, 2.0, 4.0); |
| 39 |
|
| 40 |
Constructing a virtual r-tree table <name> creates the following three |
| 41 |
real tables in the database to store the data structure: |
| 42 |
|
| 43 |
<name>_node |
| 44 |
<name>_rowid |
| 45 |
<name>_parent |
| 46 |
|
| 47 |
Dropping or modifying the contents of these tables directly will |
| 48 |
corrupt the r-tree structure. To delete an r-tree from a database, |
| 49 |
use a regular DROP TABLE statement: |
| 50 |
|
| 51 |
DROP TABLE <name>; |
| 52 |
|
| 53 |
Dropping the main r-tree table automatically drops the automatically |
| 54 |
created tables. |
| 55 |
|
| 56 |
1.2 Data Manipulation (INSERT, UPDATE, DELETE). |
| 57 |
|
| 58 |
The usual INSERT, UPDATE or DELETE syntax is used to manipulate data |
| 59 |
stored in an r-tree table. Please note the following: |
| 60 |
|
| 61 |
* Inserting a NULL value into the primary key column has the |
| 62 |
same effect as inserting a NULL into an INTEGER PRIMARY KEY |
| 63 |
column of a regular table. The system automatically assigns |
| 64 |
an unused integer key value to the new record. Usually, this |
| 65 |
is one greater than the largest primary key value currently |
| 66 |
present in the table. |
| 67 |
|
| 68 |
* Attempting to insert a duplicate primary key value fails with |
| 69 |
an SQLITE_CONSTRAINT error. |
| 70 |
|
| 71 |
* Attempting to insert or modify a record such that the value |
| 72 |
stored in the (N*2)th column is greater than that stored in |
| 73 |
the (N*2+1)th column fails with an SQLITE_CONSTRAINT error. |
| 74 |
|
| 75 |
* When a record is inserted, values are always converted to |
| 76 |
the required type (64-bit integer or 32-bit real) as if they |
| 77 |
were part of an SQL CAST expression. Non-numeric strings are |
| 78 |
converted to zero. |
| 79 |
|
| 80 |
1.3 Queries. |
| 81 |
|
| 82 |
R-tree tables may be queried using all of the same SQL syntax supported |
| 83 |
by regular tables. However, some query patterns are more efficient |
| 84 |
than others. |
| 85 |
|
| 86 |
R-trees support fast lookup by primary key value (O(logN), like |
| 87 |
regular tables). |
| 88 |
|
| 89 |
Any combination of equality and range (<, <=, >, >=) constraints |
| 90 |
on spatial data columns may be used to optimize other queries. This |
| 91 |
is the key advantage to using r-tree tables instead of creating |
| 92 |
indices on regular tables. |
| 93 |
|
| 94 |
1.4 Introspection and Analysis. |
| 95 |
|
| 96 |
TODO: Describe rtreenode() and rtreedepth() functions. |
| 97 |
|
| 98 |
|
| 99 |
2. COMPILATION AND USAGE |
| 100 |
|
| 101 |
The easiest way to compile and use the RTREE extension is to build |
| 102 |
and use it as a dynamically loadable SQLite extension. To do this |
| 103 |
using gcc on *nix: |
| 104 |
|
| 105 |
gcc -shared rtree.c -o libSqliteRtree.so |
| 106 |
|
| 107 |
You may need to add "-I" flags so that gcc can find sqlite3ext.h |
| 108 |
and sqlite3.h. The resulting shared lib, libSqliteRtree.so, may be |
| 109 |
loaded into sqlite in the same way as any other dynamicly loadable |
| 110 |
extension. |
| 111 |
|
| 112 |
|
| 113 |
3. REFERENCES |
| 114 |
|
| 115 |
[1] Atonin Guttman, "R-trees - A Dynamic Index Structure For Spatial |
| 116 |
Searching", University of California Berkeley, 1984. |
| 117 |
|
| 118 |
[2] Norbert Beckmann, Hans-Peter Kriegel, Ralf Schneider, Bernhard Seeger, |
| 119 |
"The R*-tree: An Efficient and Robust Access Method for Points and |
| 120 |
Rectangles", Universitaet Bremen, 1990. |