Geografic-Information-System/lib/vector/vectorlib_spatialindex.dox

/*! \page vlibSpatialIndex Spatial index

by GRASS Development Team (http://grass.osgeo.org)

\par Table of contents

- \ref vlibSpidx
 - \ref vlibSidxFileFormat
- \ref vlibCidx
 - \ref vlibCidxFileFormat
  - \ref vlibCidxFileHead


\section vlibSpidx Vector library spatial index management

Spatial index (based on R*-tree) is created with topology, see \ref
RTree data structure.

Spatial index occupies a lot of memory but it is necessary for 
topology building. Also, it takes some time to release the memory
occupied by spatial index (see dig_spidx_free()). The spatial index can
also be built in file to save memory by setting the environment variable
GRASS_VECTOR_LOWMEM.

The function building topology - Vect_build() - is usually called at
the end of modules (before Vect_close()) so it is faster to call
<tt>exit()</tt> and operating system releases all the memory much
faster.  By default the memory is not released.

It is possible to call Vect_set_release_support() before Vect_close()
to enforce memory release, but it takes some time on large files.

The spatial index is stored in file and not loaded for old vectors that
are not updated, saving a lot of memory. Spatial queries are done in 
file.

Currently most of the modules do not release the memory occupied for
spatial index and work like this (pseudocode):

\code
int main
{
     Vect_open_new();
     /* writing new vector */

     Vect_build();
     Vect_close();  /* memory is not released */
}
\endcode

In general it is possible to free the memory with Vect_set_release_support()
such as:

\code
int main
{
     Vect_open_new();
     /* writing new vector */

     Vect_build();
     Vect_set_release_support();
     Vect_close();  /* memory is released */
}
\endcode

but it takes a bit longer. 

It makes sense to release the spatial index if it is used only at the beginning
of a module or in permanently running programs like QGIS. Note that this
applies only when creating a new vector or updating an old vector.
For example:

\code
int main
{
     Vect_open_update();
     /* select features using spatial index, e.g.  Vect_select_lines_by_box() */
     Vect_set_release_support();
     Vect_close();  /* memory is released */

     /* do some processing which needs memory */
}
\endcode

See also \ref spatial_index data structure.

\subsection vlibSidxFileFormat Sidx file format specification

Spatial index file ('sidx') is read by Vect_open_sidx().

\subsubsection vlibSidxFileHead Header

Note: <tt>plus</tt> is instance of \ref Plus_head structure.

<table border="1" style="border-collapse: collapse" cellpadding="5">
<tr><td><b>Name</b></td><td><b>Type</b></td><td><b>Number</b></td><td><b>Description</b></td></tr>

<tr><td>plus->spidx_Version_Major </td><td>C</td><td>1</td><td>file version (major)</td></tr>
<tr><td>plus->spidx_Version_Minor </td><td>C</td><td>1</td><td>file version (minor)</td></tr>
<tr><td>plus->spidx_Back_Major</td><td>C</td><td>1</td><td>supported from GRASS version (major)</td></tr>
<tr><td>plus->spidx_Back_Minor</td><td>C</td><td>1</td><td>supported from GRASS version (minor)</td></tr>

<tr><td>plus->spidx_port->byte_order</td><td>C</td><td>1</td><td>little or big endian
                  flag; files are written in machine native order but
                  files in both little and big endian order may be
                  readl; zero for little endian</td></tr>

<tr><td>plus->spidx_port.off_t_size</td><td>C</td><td>1</td><td>off_t size (LFS)</td></tr>

<tr><td>plus->spidx_head_size</td><td>L</td><td>1</td><td>header size</td></tr>

<tr><td>plus->spidx_with_z</td><td>C</td><td>1</td><td>2D/3D vector data</td></tr>

<tr><td>ndims</td><td>C</td><td>1</td><td>Number of dimensions</td></tr>

<tr><td>nsides</td><td>C</td><td>1</td><td>Number of sides</td></tr>

<tr><td>nodesize</td><td>I</td><td>1</td><td>%Node size</td></tr>

<tr><td>nodecard</td><td>I</td><td>1</td><td>%Node card (?)</td></tr>

<tr><td>leafcard</td><td>I</td><td>1</td><td>Leaf card (?)</td></tr>

<tr><td>min_node_fill</td><td>I</td><td>1</td><td>Minimum node fill (?)</td></tr>

<tr><td>min_leaf_fill</td><td>I</td><td>1</td><td>Minimum leaf fill (?)</td></tr>

<tr><td>plus->Node_spidx->n_nodes</td><td>I</td><td>1</td><td>Number of nodes</td></tr>

<tr><td>plus->Node_spidx->n_leafs</td><td>I</td><td>1</td><td>Number of leafs</td></tr>

<tr><td>plus->Node_spidx->n_levels</td><td>I</td><td>1</td><td>Number of levels</td></tr>

<tr><td>plus->Node_spidx_offset</td><td>O</td><td>1</td><td>%Node offset</td></tr>

<tr><td>plus->Line_spidx->n_nodes</td><td>I</td><td>1</td><td>Number of nodes</td></tr>

<tr><td>plus->Line_spidx->n_leafs</td><td>I</td><td>1</td><td>Number of leafs</td></tr>

<tr><td>plus->Line_spidx->n_levels</td><td>I</td><td>1</td><td>Number of levels</td></tr>

<tr><td>plus->Line_spidx_offset</td><td>O</td><td>1</td><td>Line offset</td></tr>

<tr><td>plus->Area_spidx->n_nodes</td><td>I</td><td>1</td><td>Number of nodes</td></tr>

<tr><td>plus->Area_spidx->n_leafs</td><td>I</td><td>1</td><td>Number of leafs</td></tr>

<tr><td>plus->Area_spidx->n_levels</td><td>I</td><td>1</td><td>Number of levels</td></tr>

<tr><td>plus->Area_spidx_offset</td><td>O</td><td>1</td><td>Area offset</td></tr>

<tr><td>plus->Isle_spidx->n_nodes</td><td>I</td><td>1</td><td>Number of nodes</td></tr>

<tr><td>plus->Isle_spidx->n_leafs</td><td>I</td><td>1</td><td>Number of leafs</td></tr>

<tr><td>plus->Isle_spidx->n_levels</td><td>I</td><td>1</td><td>Number of levels</td></tr>

<tr><td>plus->Isle_spidx_offset</td><td>O</td><td>1</td><td>Isle offset</td></tr>

<tr><td>plus->Face_spidx_offset</td><td>O</td><td>1</td><td>Face offset</td></tr>

<tr><td>plus->Volume_spidx_offset</td><td>O</td><td>1</td><td>Volume offset</td></tr>

<tr><td>plus->Hole_spidx_offset</td><td>O</td><td>1</td><td>Hole offset</td></tr>

<tr><td>plus->coor_size</td><td>O</td><td>1</td><td>Coor file size</td></tr>
</table>

\section vlibCidx Vector library category index management

The category index (stored in the cidx file) improves the performance
of all selections by cats/attributes (SQL, e.g. <tt>d.vect
cats=27591</tt>, <tt>v.extract list=20000-21000</tt>). This avoids
that all selections have to be made by looping through all vector
lines.  Category index is also essential for simple feature
representation of GRASS vectors.

Category index is created for each field. In memory, it is stored in
\ref Cat_index data structure.

Category index is built with topology, but it is <b>not updated</b> if
vector is edited on level 2.  Category index is stored in 'cidx' file,
'cat' array is written/read by one call of dig__fwrite_port_I() or
dig__fread_port_I().

Stored values can be retrieved either by index in 'cat' array (if all
features of given field are required) or by category value (one or few
features), always by <tt>Vect_cidx_*()</tt> functions.

To create category index, it will be necessary to rebuild topology for
all existing vectors.  This is an opportunity to make (hopefully) last
changes in 'topo', 'cidx' formats.

\subsection vlibCidxFileFormat Cidx file format specification

Category index file ('cidx') is read by Vect_cidx_open().

\subsubsection vlibCidxFileHead Header

Note: <tt>plus</tt> is instance of \ref Plus_head structure.

<table border="1" style="border-collapse: collapse" cellpadding="5">
<tr><td><b>Name</b></td><td><b>Type</b></td><td><b>Number</b></td><td><b>Description</b></td></tr>

<tr><td>plus->cpidx_Version_Major </td><td>C</td><td>1</td><td>file version (major)</td></tr>
<tr><td>plus->cpidx_Version_Minor </td><td>C</td><td>1</td><td>file version (minor)</td></tr>
<tr><td>plus->cpidx_Back_Major</td><td>C</td><td>1</td><td>supported from GRASS version (major)</td></tr>
<tr><td>plus->cpidx_Back_Minor</td><td>C</td><td>1</td><td>supported from GRASS version (minor)</td></tr>

<tr><td>plus->cidx_port->byte_order</td><td>C</td><td>1</td><td>little or big endian
                  flag; files are written in machine native order but
                  files in both little and big endian order may be
                  readl; zero for little endian</td></tr>

<tr><td>plus->cidx_head_size</td><td>L</td><td>1</td><td>cidx head size</td></tr>

<tr><td>plus->n_cidx</td><td>I</td><td>1</td><td>number of fields</td></tr>

<tr><td>field</td><td>I</td><td>n_cidx</td><td>field number</td></tr>

<tr><td>n_cats</td><td>I</td><td>n_cidx</td><td>number of categories</td></tr>

<tr><td>n_ucats</td><td>I</td><td>n_cidx</td><td>number of unique categories</td></tr>

<tr><td>n_types</td><td>I</td><td>n_cidx</td><td>number of feature types</td></tr>

<tr><td>rtype</td><td>I</td><td>n_cidx * n_types</td><td>Feature type</td></tr>

<tr><td>type[t]</td><td>I</td><td>n_cidx * n_types</td><td>Number of items</td></tr>

</table>


*/