<div class="brand">eccenca<small>mastering complexity</small></div> # SPARQL Basics eccenca Corporate Memory training · Module 2

## Why a query language for RDF? - How is information retrieved from an RDF graph? - How are search patterns expressed? - How is reasoning put to use? <span class="ask">Which German-language strings are used?</span> <span class="ask">Which relations hold between two entities?</span> <span class="ask">Which pairs of persons share a parent?</span>

## SPARQL - **S**PARQL **P**rotocol **a**nd **R**DF **Q**uery **L**anguage - W3C Recommendation: 1.0 in 2008, **1.1 in 2013** - A **query language** for RDF graphs – and an **update language** - A **protocol** for SPARQL endpoints over HTTP - Result formats in JSON, XML, CSV and TSV In Corporate Memory: the **Queries** module, the query catalog, cmemc `query` commands, and the SPARQL endpoint.

## Structure of a query ```sparql PREFIX rov: <http://www.w3.org/ns/regorg#> # 1. prefixes SELECT ?name # 2. query form and variables WHERE { # 3. graph pattern ?x rov:legalName ?name . } ``` - Variables start with `?` (or `$`) – in subject, predicate or object position - The `WHERE` block is a **triple pattern**: Turtle with variables <span class="ask">Give me the names of all entities that have a legal name.</span>

## Basic graph patterns ```turtle @prefix ex: <http://example.org/> . ex:E63_AMG ex:producedBy <http://mercedes-benz.de/> ; ex:model "E63 AMG" ; ex:class "E-Klasse" . ``` <!-- .element: class="data" --> ```sparql PREFIX ex: <http://example.org/> SELECT ?model ?class WHERE { ?car ex:producedBy <http://mercedes-benz.de/> . ?car ex:model ?model . ?car ex:class ?class . } ``` | model | class | | --- | --- | | E63 AMG | E-Klasse |

## Matching literals ```turtle ex:bsp1 ex:p "test" . ex:bsp2 ex:p "test"^^xsd:string . ex:bsp3 ex:p "test"@de . ex:bsp4 ex:p "42"^^xsd:integer . ``` <!-- .element: class="data" --> ```sparql SELECT ?subject WHERE { ?subject ex:p "test" . } ``` | subject | | --- | | ex:bsp1 | | ex:bsp2 | <p class="small muted">RDF 1.1: <code>"test"</code> and <code>"test"^^xsd:string</code> are the same term. <code>"test"@de</code> is not.</p>

## Complex patterns - Group patterns with `{ }` - `OPTIONAL { }` – parts that may be missing; their variables stay **unbound** - `{ } UNION { }` – alternatives; the result is the union of both <span class="ask">Show the names of all customers and, *if available*, phone and email.</span> <span class="ask">Show all contact data – phone *or* email – with the customer name.</span>

## OPTIONAL ```sparql SELECT ?customer ?phone ?email WHERE { ?customerID :hasName ?customer . OPTIONAL { ?customerID :hasPhoneNumber ?phone . } OPTIONAL { ?customerID :hasEmail ?email . } } ``` | customer | phone | email | | --- | --- | --- | | "Mueller" | "03414483711" | "mueller@example.org" | | "Meyer" | | "meyer@example.org" | | "Schulze" | "0308473625" | | | "Schmidt" | | |

## UNION ```sparql SELECT ?contactData ?customer WHERE { ?customerID :hasName ?customer . { ?customerID :hasPhoneNumber ?contactData . } UNION { ?customerID :hasEmail ?contactData . } } ``` | contactData | customer | | --- | --- | | "03414483711" | "Mueller" | | "mueller@example.org" | "Mueller" | | "meyer@example.org" | "Meyer" | | "0308473625" | "Schulze" |

## Combining OPTIONAL and UNION `OPTIONAL` and `UNION` bind equally strong and are **left-associative**. ```sparql { {s1 p1 o1} OPTIONAL {s2 p2 o2} UNION {s3 p3 o3} OPTIONAL {s4 p4 o4} } ``` is read as ```sparql { { { {s1 p1 o1} OPTIONAL {s2 p2 o2} } UNION {s3 p3 o3} } OPTIONAL {s4 p4 o4} } ``` When in doubt: add braces.

## FILTER <span class="ask">Which participants are younger than 30? Whose last name contains a hyphen?</span> `FILTER ( expression )` removes every solution for which the expression is **false** or an **error**. ```sparql PREFIX ex: <http://example.org/> SELECT ?car WHERE { ?car ex:producedBy <http://mercedes-benz.de/> . ?car ex:price ?price . FILTER (?price < 70000) } ```

## Operators <div class="columns"> <div> **Comparison** `<` `=` `>` `<=` `>=` `!=`<br> numbers, dates, strings, booleans **Arithmetic** `+` `-` `*` `/` ```sparql FILTER (?weight / (?length * ?height) >= 25) ``` </div> <div> **Logic** `&&` `||` `!` | A | B | A \|\| B | A && B | | --- | --- | --- | --- | | T | E | T | E | | F | E | E | F | | E | E | E | E | Error counts as false in the filter. </div> </div>

## Functions for RDF terms | Function | Returns | | --- | --- | | `BOUND(?x)` | true if `?x` has a value | | `isIRI(?x)`, `isBLANK(?x)`, `isLITERAL(?x)` | the kind of term | | `STR(?x)` | the lexical form as a string | | `LANG(?x)`, `DATATYPE(?x)` | language tag, datatype IRI | | `langMatches(LANG(?x), "de")` | language tag matches a range | | `REGEX(?x, "pattern", "i")` | text matches a regular expression | | `CONTAINS`, `STRSTARTS`, `LCASE`, … | string functions (SPARQL 1.1) |

## Negation: FILTER NOT EXISTS <span class="ask">Show all cars produced by Mercedes-Benz that have no price.</span> ```sparql PREFIX ex: <http://example.org/> SELECT ?car WHERE { ?car ex:producedBy <http://mercedes-benz.de/> . FILTER NOT EXISTS { ?car ex:price ?price . } } ``` Also: `MINUS { … }` removes matching solutions.

## Solution modifiers ```sparql SELECT DISTINCT ?book ?price WHERE { ?book ex:price ?price . } ORDER BY DESC(?price) ?book LIMIT 5 OFFSET 25 ``` | Modifier | Effect | | --- | --- | | `ORDER BY` | sort; `ASC()` is the default, `DESC()` reverses | | `DISTINCT` | remove duplicate rows | | `OFFSET` | skip the first n rows | | `LIMIT` | return at most n rows |

## Query forms | Form | Returns | | --- | --- | | `SELECT` | a table of variable bindings | | `ASK` | `true` or `false` – does the pattern match? | | `DESCRIBE` | an RDF graph about the resources found | | `CONSTRUCT` | a new RDF graph, built from a template |

## ASK and DESCRIBE ```turtle comp:A rov:legalName "Niké" ; org:hasRegisteredSite site:1234 . comp:B rov:legalName "BARCO" . ``` <!-- .element: class="data" --> <div class="columns"> <div> ```sparql ASK { comp:B org:hasRegisteredSite ?site . } ``` → `false` </div> <div> ```sparql DESCRIBE ?org WHERE { ?org org:hasRegisteredSite site:1234 . } ``` → the triples about `comp:A` </div> </div>

## CONSTRUCT ```sparql PREFIX rov: <http://www.w3.org/ns/regorg#> PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> CONSTRUCT { ?comp rdfs:label ?name . } WHERE { ?comp rov:legalName ?name . } ``` ```turtle comp:A rdfs:label "Niké" . comp:B rdfs:label "BARCO" . ``` <!-- .element: class="result" --> A new graph from existing data – the basis of many workflows.

## SPARQL 1.1: aggregation and more ```sparql PREFIX prod: <http://ld.company.org/prod-vocab/> SELECT ?manager (COUNT(?product) AS ?products) WHERE { ?product prod:hasProductManager ?manager . } GROUP BY ?manager ORDER BY DESC(?products) ``` - Aggregates: `COUNT`, `SUM`, `MIN`, `MAX`, `AVG`, `SAMPLE`, `GROUP_CONCAT` - `BIND`, `VALUES`, subqueries, property paths (`prod:memberOf/rdfs:label`)

## SPARQL Update ```sparql PREFIX prod: <http://ld.company.org/prod-vocab/> PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> INSERT DATA { GRAPH <http://ld.company.org/training/> { <http://ld.company.org/prod-instances/empl-Jane.Doe%40company.org> a prod:Employee ; rdfs:label "Jane Doe" . } } ``` Also: `DELETE DATA`, `DELETE { … } INSERT { … } WHERE { … }`, `CLEAR GRAPH`

## Summary <div class="columns"> <div> **Structure** `PREFIX`, `WHERE { }` **Query forms** `SELECT`, `ASK`, `DESCRIBE`, `CONSTRUCT` **Patterns** `{ }`, `OPTIONAL`, `UNION`, `FILTER NOT EXISTS` </div> <div> **Modifiers** `ORDER BY`, `LIMIT`, `OFFSET`, `DISTINCT` **Filters** `BOUND`, `isIRI`, `STR`, `LANG`, `DATATYPE`, `langMatches`, `REGEX` **1.1** aggregates, `GROUP BY`, `BIND`, update </div> </div>

<div class="brand">eccenca<small>mastering complexity</small></div> # Hands-on Query the Product Data Demo

## Exercises 1. Find all products – mind the class hierarchy: `a/rdfs:subClassOf*` 2. Find hardware lighter than 10 g – `FILTER (?weight < 10)` 3. Find products whose product manager is not in the organization map – `FILTER NOT EXISTS` 4. Find products without a responsible department. 5. Find alternative experts for a product – shared `prod:areaOfExpertise` 6. Count the products per manager and department – save it in the query catalog. 7. Add an employee with `INSERT DATA`. <p class="small muted">Solutions are below the slides.</p>