Skip to content

Commit

Permalink
deegree#1543 - Remove chapter 6.5.8. and references to deprecated fun…
Browse files Browse the repository at this point in the history 
…ctionalities
  • Loading branch information
@julianzz98
julianzz98 committed Oct 31, 2024
1 parent 0dd47e8 commit d8902a1
Show file tree
Hide file tree
Showing 2 changed files with 9 additions and 222 deletions.
4 changes: 1 addition & 3 deletions deegree-documentation/src/main/asciidoc/basics.adoc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -513,11 +513,9 @@ chapters, but here's a short overview:
service metadata ("Edit metadata"), edit controller configuration ("Edit
global config")
* Feature Stores: Display feature types and number of stored features
("Info"), Import GML feature collections ("Loader"), Mapping wizard
("Create new" SQL feature store)
("Info"), Import GML feature collections ("Loader")
* Metadata Stores: Import metadata sets ("Loader"), create database
tables ("Setup tables")
* Server Connections (JDBC): Test database connection ("Test")

=== Best practices for creating workspaces

Expand Down
227 changes: 8 additions & 219 deletions deegree-documentation/src/main/asciidoc/featurestores.adoc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -984,13 +984,8 @@ GML application schema files:

===== Recommended workflow

NOTE: This section assumes that you already have an existing database that you
want to map to a GML application schema. If you want to derive a
database model from a GML application schema, see
<<anchor-mapping-wizard>>.

Manually creating a mapping for a rich GML application schema may appear
to be a dauting task at first sight. Especially when you are still
to be a daunting task at first sight. Especially when you are still
trying to figure out how the configuration concepts work, you will be
using a lot of trial-and-error. Here are some general practices to make
this as painless as possible.
Expand Down Expand Up @@ -1029,13 +1024,6 @@ data model of an application schema:
* Manually (or with the help of a generic XML tool such as XMLSpy)
analyze the GML application schema to determine the feature types and
understand their data model
* Use the services console to auto-generate a mapping configuration (see
<<anchor-mapping-wizard>>). It should reflect the structure and
datatypes correctly. Auto-generate the mapping, create a copy of the
file and start with a minimal version (_FeatureTypeMapping_ by
_FeatureTypeMapping_, property by property). Adapt it to your own
database tables and columns and remove optional elements and attributes
that you don't want to map.
* Use the deegree support options (mailing lists, commercial support) to
get help.

Expand All @@ -1050,7 +1038,7 @@ In schema-driven mode, the _<FeatureTypeMapping>_ element basically
works as in table-driven mode (see
<<anchor-configuration-tabledriven>>). It defines a mapping between a
table in the database and a feature type. However, there are additional
possibilities and it's usually more suitable to focus on feature types
possibilities, and it's usually more suitable to focus on feature types
and XML nodes instead of tables and table columns. Here's an overview of
the _<FeatureTypeMapping>_ options and their meaning in schema-driven
mode:
Expand Down Expand Up @@ -1119,7 +1107,7 @@ element and the current table context is a row of table _ad_address_.
The first (required) property that we're going to map is
_ad:inspireId_. The schema defines that _ad:inspireId_ has as child
element named _base:Identifier_ which in turn has two child elements
named _base:localId_ and _base:namespace_. Lets's assume that we
named _base:localId_ and _base:namespace_. Let's assume that we
have a column _localid_ in our table, that we want to map to
_base:localId_, but for _base:namespace_, we don't have a
corresponding column. We want this element to have the fixed value
Expand Down Expand Up @@ -1552,7 +1540,7 @@ to _true_.
* In all other cases, the _NULL_ is escalated to the container element
using the same strategy (until the feature level has been reached).

This works well most of the times, but sometimes, it can be handy to
This works well most of the time, but sometimes, it can be handy to
override this behaviour. For that, each _<Primitive>_, _<Complex>_,
_<Geometry>_ or _<Feature>_ configuration element supports the
optional attribute _nullEscalation_. The following config snippet
Expand Down Expand Up @@ -1644,14 +1632,10 @@ type

NOTE: In order for _<BLOBMapping>_ to work, you need to have the correct
tables in your database and initialize the feature type table with the
names of all feature types you want to use. We recommend not to do this
manually, see <<anchor-mapping-wizard>>. The wizard will also create
suitable indexes to speed up queries.
names of all feature types you want to use.

NOTE: You may wonder how to get data into the database in BLOB mode. As for
standard mapping, you can do this by executing WFS-T requests or by
using the feature store loader. Its usage is described in the last steps
of <<anchor-mapping-wizard>>.
standard mapping, you can do this by executing WFS-T requests.

NOTE: In BLOB mode, only spatial and feature id queries can be mapped to SQL
WHERE-constraints. All other kinds of filter conditions are performed in
Expand Down Expand Up @@ -1775,8 +1759,8 @@ column _attr_gml_id_.
The sequence id generator queries a database sequence to generate new
and unique identifiers. This requires that the database column for the
id is compatible with the values generated by the sequence and that the
database does not perform any kind of automatical value insertion for
this column (e.g triggers).
database does not perform any kind of automatic value insertion for
this column (e.g. triggers).

*SQL feature store: Database sequence generator example*

Expand Down Expand Up @@ -1823,201 +1807,6 @@ server. If you want to turn off in-memory filtering completely, use
_<DisablePostFiltering>_. If this option is specified and a filter
requires in-memory filtering, the query will be rejected.

[[anchor-mapping-wizard]]
==== Auto-generating a mapping configuration and tables

Although this functionality is still in beta stage, the services console
can be used to automatically derive an SQL feature store configuration
and set up tables from an existing GML application schema. If you don't
have an existing database structure that you want to use, you can use
this option to create a working database set up very quickly. And even
if you have an existing database you need to map manually, this
functionality can be prove very helpful to generate a valid mapping
configuration to start with.

NOTE: As every (optional) attribute and element will be considered in the
mapping, you may easily end up with hundreds of tables or columns.

This walkthrough is based on the INSPIRE Annex I schemas, but you should
be able to use these instructions with other GML application schemas as
well. Make sure that the INSPIRE workspace has been downloaded and
activated as described in <<anchor-workspace-inspire>>. As another
prerequisite, you will have to create an empty, spatially-enabled
PostGIS database that you can connect to from your deegree installation.

TIP: Instead of PostGIS, you can also use an Oracle Spatial or an Microsoft
SQL Server database. In order to enable support for these databases, see
<<anchor-db-libraries>>.

NOTE: If the application schema contains UTF-8 characters which are not part
of the 7-bit ASCII subset they are normalised during the generation of
the feature store configuration for the database mapping (but kept for
the feature type names). So the mapping to table and column names
contains only 7-bit ASCII character and it is no requirement to the
database to use UTF-8.

As a first step, create a JDBC connection to your database. Click
*server connections -> jdbc* and enter *inspire* (or an other
identifier) as connection id:

.Creating a JDBC connection
image::console_featurestore_mapping1.jpg[Creating a JDBC connection,scaledwidth=50.0%]

Afterwards, click *Create new* and enter the connection details to your
database:

.Creating a JDBC connection
image::console_featurestore_mapping2.jpg[Creating a JDBC connection,scaledwidth=50.0%]

By clicking *Test connection*, you can ensure that deegree can connect
to your database:

.Testing the JDBC connection
image::console_featurestore_mapping3.jpg[Testing the JDBC connection,scaledwidth=50.0%]

If everything works, click *Create* to finish the creation of your JDBC
resource:

.Testing the JDBC connection
image::console_featurestore_mapping4.jpg[Testing the JDBC connection,scaledwidth=50.0%]

Now, change to *data stores -> feature*. We will have to delete the
existing (memory-based) feature store first. Click *Delete*:

.Deleting the memory-based feature store
image::console_featurestore_mapping5.jpg[Deleting the memory-based feature store,scaledwidth=50.0%]

Enter "inspire" as name for the new feature store, select "SQL" from the
drop-down box and click *Create new*:

.Creating a new SQL feature store resource
image::console_featurestore_mapping6.jpg[Creating a new SQL feature store resource,scaledwidth=50.0%]

Select "Create tables from GML application schema" and click *Next*:

.Mapping a new SQL feature store configuration
image::console_featurestore_mapping7.jpg[Mapping a new SQL feature store configuration,scaledwidth=50.0%]

You can now select the GML application schema files to be used. For this
walkthrough, tick _Addresses.xsd_, _AdministrativeUnits.xsd_ and
_CadastralParcels.xsd_ (if you select all schema files, hundreds of
feature types from INPIRE Annex I will be mapped):

.Selecting the GML schema files to be considered
image::console_featurestore_mapping8.jpg[Selecting the GML schema files to be considered,scaledwidth=50.0%]

NOTE: This view presents any .xsd files that are located below the
*appschemas/* directory of your deegree workspace. If you want to map
any other GML application schema (such as GeoSciML or CityGML), place a
copy of the application schema files into the *appschemas/* directory
(using your favorite method, e.g. a file browser) and click *Rescan*.
You should now have the option to select the files of this application
schema in the services console view.

.Selecting the GML schema files to be considered
image::console_featurestore_mapping9.jpg[Selecting the GML schema files to be considered,scaledwidth=50.0%]

Scroll down and click *Next*.

.Selecting mapping type and storage CRS
image::console_featurestore_mapping10.jpg[Selecting mapping type and storage CRS,scaledwidth=50.0%]

You will be presented with a rough analysis of the feature types
contained in the selected GML application schema files. Select
"Relational" (you may also select BLOB if your prefer this kind of
storage) and enter "EPSG:4258" as storage CRS (this is the code for
ETRS89, the recommmended CRS for harmonized INSPIRE datasets). After
clicking *Next*, an SQL feature store configuration will be
automatically derived from the application schema:

.The auto-generated SQL feature store configuration
image::console_featurestore_mapping11.jpg[The auto-generated SQL feature store configuration,scaledwidth=50.0%]

Click *Save* to store this configuration:

.Auto-generated SQL statements for creating tables
image::console_featurestore_mapping12.jpg[Auto-generated SQL statements for creating tables,scaledwidth=50.0%]

Now, click *Create DB tables*. You will be presented with an
auto-generated SQL script for creating the required tables in the
database:

.Auto-generated SQL statements for creating tables
image::console_featurestore_mapping13.jpg[Auto-generated SQL statements for creating tables,scaledwidth=50.0%]

Click *Execute*. The SQL statements will now be executed against your
database and the tables will be created:

.Mapping finished
image::console_featurestore_mapping15.jpg[Mapping finished,scaledwidth=50.0%]

Click *Start feature store*:

.Finished
image::console_featurestore_mapping17.jpg[Finished,scaledwidth=50.0%]

Click *Reload* to force a reinitialization of the other workspace
resources. We're finished. Features access of the WFS and WMS uses your
database now. However, as your database is empty, the WMS will not
render anything and the WFS will not return any features when queried.
In order to insert some harmonized INSPIRE features, click *send
requests* and select one of the insert requests:

Use the third drop-down menu to select an example request. Entries
"Insert_200.xml" or "Insert_110.xml" can be used to insert a small
number of INSPIRE Address features using WFS-T insert requests:

.WFS-T example requests
image::console_workspace_inspire3.png[WFS-T example requests,scaledwidth=50.0%]

Click *Send* to execute the request. After successful insertion, the
database contains a few addresses, and you may want to move back to the
layer overview (*see layers*). If you activate the AD.Address layer, the
newly inserted features will be rendered by the deegree WMS (look for
them in the area of Enkhuizen):

.Ad.Address layer after insertion of example Address features
image::console_workspace_inspire4.png[Ad.Address layer after insertion of example Address features,scaledwidth=50.0%]

Of course, you can also perform WFS queries against the database
backend, such as requesting of INSPIRE Addresses by street name:

.More WFS examples
image::console_workspace_inspire5.png[More WFS examples,scaledwidth=50.0%]

Besides WFS-T requests, there's another handy option for inserting
GML-encoded features. Click *data stores -> feature* to access the
feature store view again:

.Accessing the feature store loader
image::console_featurestore_mapping18.jpg[Accessing the feature store loader,scaledwidth=50.0%]

After clicking *Loader*, you will be presented with a simple view where
you can insert a URL of a valid GML dataset:

.The feature store loader
image::console_featurestore_mapping19.jpg[The feature store loader,scaledwidth=50.0%]

Basically, you can use this view to insert any valid, GML-encoded
dataset, as long as it conforms to the application schema. The INSPIRE
workspace contains some suitable example datasets, so you may use a
file-URL like:

* file:/home/kelvin/.deegree/deegree-workspace-inspire/data/au-provincies.gml
* file:/home/kelvin/.deegree/deegree-workspace-inspire/data/au-gemeenten.gml
* file:/home/kelvin/.deegree/deegree-workspace-inspire/data/au-land.gml
* file:/home/kelvin/.deegree/deegree-workspace-inspire/data/cadastralparcels-limburg.xml
* file:/home/kelvin/.deegree/deegree-workspace-inspire/data/cadastralparcels-northholland.xml

TIP: The above URLs are for a UNIX system with a user named "kelvin". You
will need to adapt the URLs to match the location of your workspace
directory.

After entering the URL, click *Import*:

.Imported INSPIRE datasets via the Loader
image::console_featurestore_mapping20.jpg[Imported INSPIRE datasets via the Loader,scaledwidth=50.0%]

==== Spatial extent of FeatureTypes

The spatial extent of all feature types defined in all SQLFeatureStore configurations are cached in a file named _bbox_cache.properties_. The file is created when the workspace is initialised.
Expand Down

0 comments on commit d8902a1

Please sign in to comment.