Let's say I have a triple store full of video metadata. A video can have an arbitrary selection of pre-defined tags, like this:
v1 ex:hasTag A.
v2 ex:hasTag B.
v3 ex:hasTag A;
ex:hasTag B.
So for the sake of this example, there are only the two pre-defined tags A and B.
Now I would like to get an overview of which video has which tags attached, in a matrix like this (or similar):
A B
v1 true false
v2 false true
v3 true true
However, I do not know how to achieve this. First of all, I would need a boolean test, e.g. something like isTrue(?video ex:hasTag A) or whatever it looks like. Even then, I do not know where to put it. If I put the previous statement in the WHERE clause, the query result only contains the videos having the tag A, of course.
Coming from SQL, I can imagine I need to use subqueries in some way, but these seem not to be standardized at the moment. I also saw the FILTER keyword, but I don't feel that is what I want.
I'm currently at a loss but always willing to learn. Any help is appreciated.
Subqueries are in SPARQL 1.1 and that is frozen now. There several complete implementations already.
As well as nested-SELECT subqueries, SPARQL 1.1 has the EXISTS and NOT EXISTS functions for testing whether a pattern matches or not. It#'s a form of subquery. You can use it in a FILTER or if you really want to return true/false:
BIND(EXISTS{?video ex:hasTag "A"} AS ?a)
or even:
SELECT ?video (EXISTS{?video ex:hasTag "A"} AS ?a) (EXISTS{?video ex:hasTag "B"} AS ?b)
{
?video a ex:Video .
}
Try starting with something like this:
SELECT ?video ?tagA ?tagB {
?video a ex:Video .
OPTIONAL { ?video ex:hasTag ?tagA . FILTER (?tagA = "A") }
OPTIONAL { ?video ex:hasTab ?tagB . FILTER (?tagB = "B") }
}
That will you roughly what you want. If you really must have boolean values rather than simple checking for unbound vals use:
SELECT ?video (bound(?tagA) as ?a) (bound(?tagB) as ?b) { ... }
P.S. Subqueries are (very nearly, proposed recommendation at time of writing) standardised in SPARQL 1.1.
Related
I have two arrays of strings and I want to check if a string of array a matches a string from array b. Those strings are phone numbers that might come in different formats. For example:
Array a might have a phone number with prefix like so +44123123123 or 0044123123123
Array b have a standard format without prefixes like so 123123123
So I'm looking for a regex that can match a part of a string like +44123123123 with 123123123
Btw I'm using Swift but I don't think there's a native way to do it (at least a more straightforward solution)
EDIT
I decided to reactivate the question after experimenting with the library #Larme mentioned because of inconsistent results.
I'd prefer a simper solution as I've stated earlier.
SOLUTION
Thanks guys for the responses. I saw many comments saying that Regex is not the right solution for this problem. And this is partly true. It could be true (or false) depending on my current setup/architecture ( which thinking about it now I realise that I should've explained better).
So I ended up using the native solution (hasSuffix/contains) but to do that I had to do some refactoring on the way the entire flow was structured. In the end I think it was the least complicated solution and more performant of the two. I'll give the bounty to #Alexey Inkin for being the first to mention the native solution and the right answer to #Ωmega for providing a more complete solution.
I believe regex is not the right approach for this task.
Instead, you should do something like this:
var c : [String] = b.filter ({ (short : String) -> Bool in
var result = false
for full in a {
result = result || full.hasSuffix(short)
}
return result
})
Check this demo.
...or similar solution like this:
var c : [String] = b.filter ({ (short : String) -> Bool in
for full in a {
if full.hasSuffix(short) { return true }
}
return false
})
Check this demo.
As you do not mention requirements to prefixes, the simplest solution is to check if string in a ends with a string in b. For this, take a look at https://developer.apple.com/documentation/swift/string/1541149-hassuffix
Then, if you have to check if the prefix belongs to a country, you may replace ^00 with + and then run a whitelist check against known prefixes. And the prefix itself can be obtained as a substring by cutting b's length of characters. Not really a regex's job.
I agree with Alexey Inkin that this can also nicely be solved without regex. If you really want a regex, you can try something like the following:
(?:(\+|00)(93|355|213|1684|376))?(\d+)
^^^^^^^^^^^^^^^^^^^^^ Add here all your expected country prefixes (see below)
^^^ ^^ Match a country prefix if it exists but don't give it a group number
^^^^^^^ Match the "prefix-prefix" (+ or 00)
^^^^ Match the local phone number
Unfortunatly with this regex, you have to provide all the expected country prefixes. But you can surely get this list online, e.g. here: https://www.countrycode.org
With this regex above you will get the local phone number in matching group 3 (and the "prefix-prefix" in group 1 and the country code in group 2).
Postgresql 10+
Example from the documentation...
jsonb_set('[{"f1":1,"f2":null},2,null,3]', '{0,f1}','[2,3,4]', false)
results in...
[{"f1":[2,3,4],"f2":null},2,null,3]
Fair enough. But I need to find my target node by attribute value, not index. For the life of me, I cannot figure out how do something like...
jsonb_set('[{"f1":1,"f2":null},2,null,3]', '{(where f1 = 1),f1}','[2,3,4]', false)
Any advice on how to accomplish this?
Thanks!
You can split the steps into two jobs:
Split in elements (jsonb_arral_elements)
Indentify wich elements must change (case when...)
Update that element (jsonb_set)
Join all together (jsonb_agg)
solution
select jsonb_agg(case when element->>'f1'='1' then jsonb_set(element, '{f1}', '[2,3,4]') else element end)
from jsonb_array_elements('[{"f1":1,"f2":null},2,null,3,{"f1":3},{"f1":1,"f2":2}]'::jsonb) element
note
I changed the input adding two more elements with "f1" key
I'm trying to get only a list of URIs in RDF instead of a list of triples:
PREFIX gr: <http://purl.org/goodrelations/v1#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
DESCRIBE ?product
WHERE
{
?product rdfs:subClassOf gr:ProductOrService .
}
Using SELECT, instead of DESCRIBE I receive only the subject (which I want), but not as an RDF but like a SPARQL Result with binds, vars, etc.
Using CONSTRUCT, I can't specify only the ?product, as above, so the closest I can get is:
PREFIX gr: <http://purl.org/goodrelations/v1#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
CONSTRUCT
WHERE
{
?product rdfs:subClassOf gr:ProductOrService .
}
Which returns an RDF with triples of different products, but the same properties and objects.
It looks like you should read over the SPARQL specification, which tells you that this is exactly as expected. To get the single column you wish, you must use SELECT.
SPARQL has four query forms. These query forms use the solutions from pattern matching to form result sets or RDF graphs. The query forms are:
SELECT
Returns all, or a subset of, the variables bound in a query pattern match.
CONSTRUCT
Returns an RDF graph constructed by substituting variables in a set of triple templates.
ASK
Returns a boolean indicating whether a query pattern matches or not.
DESCRIBE
Returns an RDF graph that describes the resources found.
I found a solution. I've used SELECT, but instead of binds and vars, as output received a "Comma-Separated Values (with fields in N-Triples syntax)" CSV file:
<http://www.productontology.org/id/Real_estate>
<http://www.productontology.org/id/Automobile>
<http://www.productontology.org/id/Auction>
<http://www.productontology.org/id/Video_game>
<http://www.productontology.org/id/Campsite>
<http://www.productontology.org/id/Car>
<http://www.productontology.org/id/Audiobook>
<http://www.productontology.org/id/Browser_game>
...
Representing the result of a SPARQL SELECT query as an RDF List is a tooling issue - it is not something that can be solved in SPARQL in general.
Some SPARQL tools may have ways to support rendering the query result as an RDF list. But it's not something you can fix by just formulating your query differently, you'll need to use a tool to (programmatically) format the result.
In Java, using Eclipse RDF4J, you can do it as follows (untested so you may need to tweak it to work properly, but it should give you a general idea):
String query = "SELECT ?product WHERE { ?product rdfs:subClassOf gr:ProductOrService . }";
// do the query on repo and convert to a Java list of URI objects
List<URI> results = Repositories.tupleQuery(
repo,
query,
r -> QueryResults.stream(r).map(bs -> (URI)bs.getValue("product")).collect(Collectors.toList()
);
// create a resource (bnode or URI) for the start of the rdf:List
Resource head = SimpleValueFactory.getInstance().createBNode();
// convert the Java list of results to an rdf:list and add it
// to a newly-created RDF Model
Model m = RDFCollections.asRDF(results, head, new LinkedHashModel());
Once you have your result as an RDF model you can use any of the existing RDF4J APIs to write it to file or to store it in a triplestore.
Now, I am not claiming that any of this is a good idea - I have never seen a use case for something like this. But this is how I would do it if it were necessary.
The Overpass API language guide does allow for logical operators when matching a tag value... for example:["name"~"holtorf|Gielgen"] will return whatever object has either name=holtorf or name=Gielgen.
You can also combine conditions and they will become an AND... so for example:
["name"]["name"="holtorf"]. Means to search for things that have the tag "name" and that the tag name is equal to "holtorf".
But what I want is an OR operator... something like:
["name"="holtorf"]|["name:eng"holtorf"]
In my specific application, I just want to know if there is ANY tag that start with "name"... so what I would like to do is put this into the API: ["^name"] (cause in this API "^" means "starts with"). But of course it searches for literal "^name" and returns nothing.
Is there some workaround?
There is no OR operation, but you can use UNION
(
way["name"="holtorf"];
way["name:eng"=holtorf"]
);
There is also a DIFFERENCE and negotiation http://wiki.openstreetmap.org/wiki/Overpass_API/Overpass_QL#Difference
And in your particular case, you could use key-value regexpressions matching. http://wiki.openstreetmap.org/wiki/Overpass_API/Overpass_QL#Key.2Fvalue_matches_regular_expression_.28.7E.22key_regex.22.7E.22value_regex.22.29
[~"^name.*$"~"^holtorf$"];
//or only for key
[~"^name.*$"="Holtorf"];
I've been using filters in Sinatra the way it has been declared in the documentation: with no match string, with a match string or with a match regexp. It has been working fine til now. Now, I have a particular use case. Let's say I have this route:
/resources/1
According to REST, and depending of the request method, this can either be a GET method, PUT method or DELETE method. First question is: How to write filters that are only called when it is a GET request? (currently I'm letting all of them get filtered and only then I test the method. It works, but I don't like it). Second question, and more important: let's say a PUT request like this is triggered:
/resources/
This is of course wrong, because the PUT request has no resource id associated. I would like to know if there is something in Sinatra that enables me to do something like this:
before "/resources/", :method => :put do
error_message
end
just this possibility does not exist (before accepts only one argument). How could I achieve this result at best?
Actually, filters do take conditions. You don't have to use a condition though, you could use a conditional within the filter:
before "/path/" do
if request.request_method == "PUT"
# do something
end
end
If you want to use a condition, I think you'll need to write one, perhaps something like this:
set(:accepted_verbs) {|*verbs|
condition {
verbs.any?{|v| v == request.request_method }
}
}
before "/path/", :accepted_verbs => ["GET","POST"] do
# do something
end
before "/path/", :accepted_verbs => ["PUT"] do
# do something else
end
See conditions for more.