Lab: Ontop - Revision history

Sinoa: /* If you really have a lot of time */

2025-02-25T18:09:23Z

If you really have a lot of time

← Older revision		Revision as of 18:09, 25 February 2025
Line 232:		Line 232:
	'''Task:''' Add the new virtual repository to the FedX federated repository you created before. You can now write a SPARQL query that combines triples from all the three repositories it contains, e.g.,		'''Task:''' Add the new virtual repository to the FedX federated repository you created before. You can now write a SPARQL query that combines triples from all the three repositories it contains, e.g.,

	* For all investigations, list its name, who was president, that president's Wikidata URL, and a brief description in English.		* For all investigations, list its name, who was president, that president's Wikidata URL, and a brief description of the president in English.

Sinoa: /* If you have more time */

2025-02-25T18:07:53Z

If you have more time

← Older revision		Revision as of 18:07, 25 February 2025
Line 193:		Line 193:
	'''Task:''' Use ''FILTER'' and the ''LANG'' function to only list president descriptions with language ''"en"''.		'''Task:''' Use ''FILTER'' and the ''LANG'' function to only list president descriptions with language ''"en"''.

	''...from here we could have gone on to add the Ontop repository too as a local repository, but Ontop does not support SQLite3 fully, unfortunately. Using H2 instead of SQLite3 is an option, we outline below.''		''...from here we could have gone on to add the Ontop repository too as a local repository, but Ontop does not support SQLite3 fully, unfortunately. Using H2 instead of SQLite3 is an option, as we outline below.''

	''Tip:'' Here are some useful PREFIXES:		''Tip:'' Here are some useful PREFIXES:

Sinoa: /* Installing and running SQLite3 */

2025-02-25T18:05:18Z

Installing and running SQLite3

← Older revision		Revision as of 18:05, 25 February 2025
Line 18:		Line 18:
	'''Task:''' Download and install a relational database system on your computer. If you are not experienced in databases, we recommend the very light-weight and easy-to-install [https://sqlite.org/download.html SQLite3].		'''Task:''' Download and install a relational database system on your computer. If you are not experienced in databases, we recommend the very light-weight and easy-to-install [https://sqlite.org/download.html SQLite3].

	(~~However~~, SQLite3 is not fully integrated with Ontop. A more powerful alternative is H2. At the end of this exercise you will find guidelines for doing the exercise with H2 instead.)		(Unfortunately, SQLite3 is not fully integrated with Ontop. A more powerful alternative is H2. At the end of this exercise you will find guidelines for doing the exercise with H2 instead.)

	'''Task:''' Create a folder on your local computer for this exercise. Go to the new folder and open a Console or Terminal window. Start SQLite3 with the name of a new database, for example 'MyTestDB.sqlite'.		'''Task:''' Create a folder on your local computer for this exercise. Go to the new folder and open a Console or Terminal window. Start SQLite3 with the name of a new database, for example 'MyTestDB.sqlite'.

Sinoa: /* Installing and running SQLite3 */

2025-02-25T18:01:43Z

Installing and running SQLite3

← Older revision		Revision as of 18:01, 25 February 2025
Line 16:		Line 16:

	===Installing and running SQLite3===		===Installing and running SQLite3===
	'''Task:''' Download and install a relational database system on your computer. We recommend the very light-weight and easy-to-install [https://sqlite.org/download.html SQLite3].		'''Task:''' Download and install a relational database system on your computer. If you are not experienced in databases, we recommend the very light-weight and easy-to-install [https://sqlite.org/download.html SQLite3].

			(However, SQLite3 is not fully integrated with Ontop. A more powerful alternative is H2. At the end of this exercise you will find guidelines for doing the exercise with H2 instead.)

	'''Task:''' Create a folder on your local computer for this exercise. Go to the new folder and open a Console or Terminal window. Start SQLite3 with the name of a new database, for example 'MyTestDB.sqlite'.		'''Task:''' Create a folder on your local computer for this exercise. Go to the new folder and open a Console or Terminal window. Start SQLite3 with the name of a new database, for example 'MyTestDB.sqlite'.

Sinoa: /* If you really have a lot of time */

2025-02-25T17:59:28Z

If you really have a lot of time

Sinoa: /* If you really have a lot of time */

2025-02-25T17:55:36Z

If you really have a lot of time

← Older revision		Revision as of 17:55, 25 February 2025
Line 211:		Line 211:
	'''Task:''' [https://www.h2database.com/html/download.html Download] and [https://www.h2database.com/html/installation.html install H2].		'''Task:''' [https://www.h2database.com/html/download.html Download] and [https://www.h2database.com/html/installation.html install H2].

	Again, ~~you need~~ an additional ''.jar''-file - a JDBC driver - for connecting to your H2 database. You will find this file in the folder where you installed H2. It is called something like ''h2-2.3.232.jar''. Place a copy of it in ''lib/''-subfolder of your GraphDB installation folder (same place where you put the ''sqlite-jdbc-3.49.0.0.jar''-file). Restart Ontop.		Again, Ontop needs an additional ''.jar''-file - a JDBC driver - for connecting to your H2 database. You will find this file in the folder where you installed H2. It is called something like ''h2-2.3.232.jar''. Place a copy of it in ''lib/''-subfolder of your GraphDB installation folder (same place where you put the ''sqlite-jdbc-3.49.0.0.jar''-file). Restart Ontop.

	'''Task:''' Start H2. Normally it opens a browser window on port ''8082'', but there is a console (command line) option too.		'''Task:''' Start H2. Normally it opens a browser window on port ''8082'', but there is a console (command line) option too.

Sinoa: /* If you really have a lot of time */

2025-02-25T17:54:59Z

If you really have a lot of time

← Older revision		Revision as of 17:54, 25 February 2025
Line 205:		Line 205:
	* a virtual repository (like the ''Investigations'' and ''InvestigatedPeople'' repository) as long as it runs on top of ''H2'' instead of ''SQLite3''.		* a virtual repository (like the ''Investigations'' and ''InvestigatedPeople'' repository) as long as it runs on top of ''H2'' instead of ''SQLite3''.

	'''Task:''' Install a recent Java on your computer if you do not have already, for example [~~for example from~~ https://adoptium.net/en-GB/download/ the open version from Adoptium].		'''Task:''' Install a recent Java on your computer if you do not have already, for example [https://adoptium.net/en-GB/download/ the open version from Adoptium].
	java -version		java -version
	must show version 11 or later, and Java cannot be blocked by your firewall.		must show version 11 or later, and Java cannot be blocked by your firewall.

Sinoa: /* If you really have a lot of time */

2025-02-25T17:52:51Z

If you really have a lot of time

@@ Line 211: / Line 211: @@
 '''Task:''' [https://www.h2database.com/html/download.html Download] and [https://www.h2database.com/html/installation.html install H2].
-Updated Java, for example from https://adoptium.net/en-GB/download/
+You may need to disconnect from H2 before you connect from GraphDB/Ontop.
-download h2 jdbc driver for Ontop
+* For all investigations, list its name, who was president, that president's Wikidata URL, and a brief description in English.

Sinoa: /* If you really have a lot of time */

2025-02-25T17:32:05Z

If you really have a lot of time

@@ Line 203: / Line 203: @@
 * a local repository (like the ''Presidents'' repository),
 * a remote repository (for example ''https://query.wikidata.org/sparql'' as a ''Generic SPARQL endpoint''), ''and''
-* a virtual repository (for example ``MyTestSchema'') as long as it runs on top of ''H2'' instead of ''SQLite3''.
+* a virtual repository (like the ''Investigations'' and ''InvestigatedPeople'' repository) as long as it runs on top of ''H2'' instead of ''SQLite3''.

Sinoa: /* If you have more time */

2025-02-25T17:26:38Z

If you have more time

← Older revision		Revision as of 17:26, 25 February 2025
Line 191:		Line 191:
	'''Task:''' Use ''FILTER'' and the ''LANG'' function to only list president descriptions with language ''"en"''.		'''Task:''' Use ''FILTER'' and the ''LANG'' function to only list president descriptions with language ''"en"''.

	''...from here we could have gone on to add the Ontop repository too as a local repository, but ~~the current version of~~ Ontop ~~for GraphDB~~ does not ~~seem to~~ support ~~Federated repositories that include Ontop repositories~~, unfortunately. ~~We have asked the Ontop developers for guidance~~, ~~and will update the exercise if they offer a solution~~.''		''...from here we could have gone on to add the Ontop repository too as a local repository, but Ontop does not support SQLite3 fully, unfortunately. Using H2 instead of SQLite3 is an option, we outline below.''

	''Tip:'' Here are some useful PREFIXES:		''Tip:'' Here are some useful PREFIXES:
Line 198:		Line 198:
	PREFIX owl: <http://www.w3.org/2002/07/owl#>		PREFIX owl: <http://www.w3.org/2002/07/owl#>
	PREFIX schema: <http://schema.org/>		PREFIX schema: <http://schema.org/>

			==If you ''really'' have a lot of time==
			It is possible to create a ''Federated'' (FedX Virtual SPARQL) repository that federates
			* a local repository (like the ''Presidents'' repository),
			* a remote repository (for example ''https://query.wikidata.org/sparql'' as a ''Generic SPARQL endpoint''), ''and''
			* a virtual repository (for example ``MyTestSchema'') as long as it runs on top of ''H2'' instead of ''SQLite3''.

← Older revision		Revision as of 17:59, 25 February 2025
Line 215:		Line 215:
	'''Task:''' Start H2. Normally it opens a browser window on port ''8082'', but there is a console (command line) option too.		'''Task:''' Start H2. Normally it opens a browser window on port ''8082'', but there is a console (command line) option too.
	* Default driver class: ''org.h2.Driver'' - this one is fine		* Default driver class: ''org.h2.Driver'' - this one is fine
	* Default JDBC URL: ''jdbc:h2:~/test'' - you can replace the ~ with a full path and perhaps give the DB a better name too		* Default JDBC URL: ''jdbc:h2:~/test'' - if you want, you can replace the ''~'' with a full path and perhaps give the DB a better name than ''test'' too
	You do not need a password or to change the user name.		You do not need a password or to change the user name.

	'''Task:''' Click connect and define your database. You can cut-and-paste the same SQL commands as before to create the ''Investigations'' and ''InvestigatedPeople'' databases. But ~~there are~~ a few critical differences:		'''Task:''' Click connect and define your database. You can cut-and-paste the same SQL commands as before to create the ''Investigations'' and ''InvestigatedPeople'' databases. But be aware of a few critical differences:
	* H2 uses ''schemas'' to organise tables. So first you need to create one for your two tables, for example:		* H2 uses ''schemas'' to organise tables. So first you need to create one for your two tables, for example:
	CREATE SCHEMA MyTestSchema ;		CREATE SCHEMA MyTestSchema ;
	* After that, you must always ~~write~~ ''MyTestSchema.Investigations'' and ''MyTestSchema.InvestigatedPeople'', both in your SQL commands and in your OBDA mappings.		* After that, you must always prefix your table names with the schema: ''MyTestSchema.Investigations'' and ''MyTestSchema.InvestigatedPeople'' - both in your SQL commands and in your OBDA mappings.
	* If you do not write schema, table, and column names inside "quotation marks", H2 will convert them to UPPER CASE.		* If you do not write schema, table, and column names inside "quotation marks", H2 will automatically convert them to UPPER CASE.

	'''Task:''' Create a new virtual repository like you did before, but using the driver class and JDBC URL for H2 instead. You need to enter a H2 user name too. This virtual repository is more stable than the one that wraps SQLite3.		'''Task:''' Create a new virtual repository like you did before, but using the driver class and JDBC URL for H2 instead. You need to enter a H2 user name too. This virtual repository is more stable than the one that wraps SQLite3.

	You may need to disconnect from H2 before you connect ~~from~~ GraphDB/Ontop.		You may need to disconnect from H2 in the browser window before you connect to it via GraphDB and Ontop.

	'''Task:''' Add the new virtual repository to the FedX federated repository you created before. You can now write a SPARQL query that combines triples from all the three repositories it contains, e.g.,		'''Task:''' Add the new virtual repository to the FedX federated repository you created before. You can now write a SPARQL query that combines triples from all the three repositories it contains, e.g.,

	* For all investigations, list its name, who was president, that president's Wikidata URL, and a brief description in English.		* For all investigations, list its name, who was president, that president's Wikidata URL, and a brief description in English.