Library Overview. Hadjieleftheriou, M. In Libspatialindex 1.9.0 Documentation.
Library Overview [link]Paper  abstract   bibtex   
[Excerpt] The library currently consists of six packages: [::] The core spatialindex utilities. [::] The storagemanager files. [::] The spatialindex interfaces. [::] The rtree index. [::] The mvrtree index. [::] The tprtree index. [\n] [...] [::Spatial Index Utilities] To provide common constructors and uniform initialization for all objects provided by the library a PropertySet class is provided. A PropertySet associates strings with Variants. Each property corresponds to one string. [\n] [...] [::Storage Manager] The library provides a common interface for storage management of all indices. It consists of the IStorageManager interface, which provides functions for storing and retrieving entities. An entity is viewed as a simple uint8_t array; hence it can be an index entry, a data entry or anything else that the user wants to store. The storage manager interface is generic and does not apply only to spatial indices. [\n] [...] [::SpatialIndex Interfaces] A spatial index is any index structure that accesses spatial information efficiently. It could range from a simple grid file to a complicated tree structure. A spatial index indexes entries of type IEntry, which can be index nodes, leaf nodes, data etc. depending on the structure characteristics. The appropriate interfaces with useful accessor methods should be provided for all types of entries. [\n] [...] [::The RTree Package] The RTree index [guttman84] is a balanced tree structure that consists of index nodes, leaf nodes and data. Every node (leaf and index) has a fixed capacity of entries, (the node capacity) chosen at index creation An RTree abstracts the data with their Minimum Bounding Region (MBR) and clusters these MBRs according to various heuristics in the leaf nodes. Queries are evaluated from the root of the tree down the leaves. Since the index is balanced nodes can be under full. They cannot be empty though. A fill factor specifies the minimum number of entries allowed in any node. The fill factor is usually close to 70%. [\n] [...] [\n] [...]
@incollection{hadjieleftheriouLibraryOverview2014,
  title = {Library Overview},
  booktitle = {Libspatialindex 1.9.0 Documentation},
  author = {Hadjieleftheriou, Marios},
  date = {2014},
  url = {https://web.archive.org/web/201905/https://libspatialindex.org/overview.html},
  urldate = {2019-05-28},
  abstract = {[Excerpt]
The library currently consists of six packages:
[::] The core spatialindex utilities.
[::] The storagemanager files.
[::] The spatialindex interfaces.
[::] The rtree index.
[::] The mvrtree index.
[::] The tprtree index.

[\textbackslash n] [...]

[::Spatial Index Utilities]
To provide common constructors and uniform initialization for all objects provided by the library a PropertySet class is provided. A PropertySet associates strings with Variants. Each property corresponds to one string.
[\textbackslash n] [...]

[::Storage Manager]
The library provides a common interface for storage management of all indices. It consists of the IStorageManager interface, which provides functions for storing and retrieving entities. An entity is viewed as a simple uint8\_t array; hence it can be an index entry, a data entry or anything else that the user wants to store. The storage manager interface is generic and does not apply only to spatial indices.
[\textbackslash n] [...]

[::SpatialIndex Interfaces]
A spatial index is any index structure that accesses spatial information efficiently. It could range from a simple grid file to a complicated tree structure. A spatial index indexes entries of type IEntry, which can be index nodes, leaf nodes, data etc. depending on the structure characteristics. The appropriate interfaces with useful accessor methods should be provided for all types of entries.
[\textbackslash n] [...]

[::The RTree Package]
The RTree index [guttman84] is a balanced tree structure that consists of index nodes, leaf nodes and data. Every node (leaf and index) has a fixed capacity of entries, (the node capacity) chosen at index creation An RTree abstracts the data with their Minimum Bounding Region (MBR) and clusters these MBRs according to various heuristics in the leaf nodes. Queries are evaluated from the root of the tree down the leaves. Since the index is balanced nodes can be under full. They cannot be empty though. A fill factor specifies the minimum number of entries allowed in any node. The fill factor is usually close to 70\%.
[\textbackslash n] [...]

[\textbackslash n] [...]},
  keywords = {~INRMM-MiD:z-IQU69UMD,documentation,free-scientific-software,free-software,geospatial,reference-manual}
}

Downloads: 0