I've been looking into GraphQL more recently, so it's a possibility. When I originally asked this question we had other systems that were using Lucene so we were looking for a quick way to federate a lucene query to multiple systems and aggregate the results, which that magic property worked perfectly well for.
As for now, it was mostly just curiosity about how it worked. Thanks for the feedback. On Wed, Aug 4, 2021, 9:15 PM Holger Knublauch <hol...@topquadrant.com> wrote: > Matt, > > I have checked and we consider this API to be rather internal and it's > undocumented for a reason :) > > Would you be able to use GraphQL for what you want to achieve? What was > your use case that you need those facets? > > Holger > > > On 2021-08-04 1:48 pm, Holger Knublauch wrote: > > Hi Matt, > > I need to ask a colleague and we'll get back to you. Meanwhile, looking at > the source code, it seems that ?facetFilters can be a resource that is the > subject of > > http://topbraid.org/saf/appliedfacet > > and you can see low-level details of how those values are constructed from > saf.ui.ttlx -> saf:buildAppliedFacetList > > Holger > > > On 2021-08-03 10:34 pm, Matt Goldberg wrote: > > That works, thanks. > > What is the correct syntax for the facet filters? I haven't been able to > find any examples anywhere. > > On Mon, Aug 2, 2021, 8:23 PM Holger Knublauch <hol...@topquadrant.com> > wrote: > >> I suspect you need to now provide a forth variable on the right hand >> side, e.g. >> >> ("term" ?facetFilters 0 100 false) smf:luceneQuery (?result ?score >> ?total ?homeGraph) >> >> Holger >> >> >> On 2021-08-03 2:14 am, Matt Goldberg wrote: >> >> Hello- >> >> I had been successfully using smf:luceneQuery before upgrading to EDG 7. >> If I run the example query you provided earlier in this thread >> >> SELECT ?result ?score ?total >> WHERE { >> ("term" ?facetFilters 0 100 false) smf:luceneQuery (?result ?score >> ?total) >> } >> >> in EDG or TBC 6.3.2, it works fine. If I run the same query in EDG/TBC >> 7.0.3, I get the following error: >> Failed to execute SPARQL request: java.lang.IndexOutOfBoundsException: >> Index 3 out of bounds for length 3 >> >> Any ideas why this might be happening? >> Thanks. >> >> On Tuesday, November 10, 2020 at 7:20:12 PM UTC-5 Matt Goldberg wrote: >> >>> There's a few use cases we're looking at. At the moment, the primary use >>> case involves a proof of concept regarding a global search that federates a >>> free text query. We have some other systems that also support Lucene, so >>> the idea is that at least all the basic operators would work (at least *, >>> ?, OR, AND, NOT) to provide moderately consistent performance across >>> systems. I wrote a SWP service that uses smf:luceneQuery, which does work >>> nicely in terms of the query itself, but it does run into the challenges >>> you mentioned above. >>> >>> On Tuesday, November 10, 2020 at 5:24:56 PM UTC-5 Holger Knublauch wrote: >>> >>>> I guess that's true. Due to the nature of the "normal" requirements >>>> that this feature is used for, it will take the input string and >>>> post-process it beyond recognition, for example to inject the names of the >>>> target graphs, to split multiple words into AND and to wrap each term with >>>> * ... *. However, you can use OR in between words. >>>> >>>> What operators in particular do you need? >>>> >>>> Holger >>>> >>>> >>>> On 11/11/2020 3:27 AM, Matt Goldberg wrote: >>>> >>>> Hello- >>>> >>>> Is it true that textindex:query magic property cannot support the >>>> Lucene operators? >>>> >>>> Thanks. >>>> >>>> On Friday, November 6, 2020 at 1:06:59 PM UTC-5 Matt Goldberg wrote: >>>> >>>>> Thanks for the heads up. I'll continue experimenting. >>>>> >>>>> On Thursday, November 5, 2020 at 6:49:55 PM UTC-5 Holger Knublauch >>>>> wrote: >>>>> >>>>>> Hi Matt, >>>>>> smf:luceneQuery is in fact the magic property that Search the EDG is >>>>>> using, but without knowing more about your specific use case I would say >>>>>> 'use with caution' as the magic property is really meant for use with in >>>>>> SWP and has some quirks. >>>>>> >>>>>> For example, the user must define a limit, facets are kind of a pain >>>>>> to pass through, target graphs must opt in, working copies are out of >>>>>> scope, and lastly all graphs in the index are queried - the only way to >>>>>> narrow the scope is by providing a graph facet. >>>>>> >>>>>> But generally speaking this should work >>>>>> (params are: search term, facets, offset, limit, advanced syntax >>>>>> (when true the magic property will use the input verbatim, essentially >>>>>> allowing lucene operands to be utilized) >>>>>> >>>>>> SELECT ?result ?score ?total >>>>>> WHERE { >>>>>> ("term" ?facetFilters 0 100 false) smf:luceneQuery (?result ?score >>>>>> ?total) >>>>>> } >>>>>> >>>>>> If the user needs facets, we'd have to work up a more thorough >>>>>> example, which would involve temp graphs. >>>>>> >>>>>> So the alternative would be to use the textindex:query magic property >>>>>> while iterating over the properties of interest. >>>>>> >>>>>> Holger >>>>>> >>>>>> >>>>>> On 11/6/2020 12:12 AM, Matt Goldberg wrote: >>>>>> >>>>>> That magic property requires a specific property to search values of, >>>>>> which is not quite what I want at the moment. After digging some more, I >>>>>> found smf:luceneQuery which appears to return similar results to Search >>>>>> the >>>>>> EDG. Is there any caveats with using this magic property? >>>>>> >>>>>> On Wednesday, November 4, 2020 at 6:04:54 PM UTC-5 Holger Knublauch >>>>>> wrote: >>>>>> >>>>>>> Hi Matt, >>>>>>> >>>>>>> yes there are functions in the textindex namespace ( >>>>>>> http://topbraid.org/textindex#) >>>>>>> >>>>>>> Open the file \server.topbraidlive.org\web\2018\textindex.ui.ttlx >>>>>>> in TBC to see their declarations. >>>>>>> >>>>>>> The name of the index for EDG vocabularies is "teamwork" - see the >>>>>>> Text Indices admin page. >>>>>>> >>>>>>> An example call would be >>>>>>> >>>>>>> ("teamwork" rdfs:comment "hello") textindex:query (?subject >>>>>>> ?score ?literal ?graph) >>>>>>> >>>>>>> If you have follow-up questions, please ask here. >>>>>>> >>>>>>> Holger >>>>>>> >>>>>>> >>>>>>> On 11/5/2020 8:17 AM, Matt Goldberg wrote: >>>>>>> >>>>>>> Hello- >>>>>>> >>>>>>> Is there a mechanism like a magic property to query the freetext >>>>>>> index from SPARQL? I didn't see one mentioned in the TopBraid >>>>>>> documentation. I've used some other systems that have this feature. >>>>>>> >>>>>>> Thanks. >>>>>>> >>>>>>> -- >>>>>>> You received this message because you are subscribed to the Google >>>>>>> Groups "TopBraid Suite Users" group. >>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>> send an email to topbraid-user...@googlegroups.com. >>>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/d/msgid/topbraid-users/ac42f5bc-323f-4254-8027-20e166d9e264n%40googlegroups.com >>>>>>> <https://groups.google.com/d/msgid/topbraid-users/ac42f5bc-323f-4254-8027-20e166d9e264n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>>> . >>>>>>> >>>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "TopBraid Suite Users" group. >>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>> send an email to topbraid-user...@googlegroups.com. >>>>>> >>>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/topbraid-users/59e998be-8c2c-4e47-82e2-e9948acaa846n%40googlegroups.com >>>>>> <https://groups.google.com/d/msgid/topbraid-users/59e998be-8c2c-4e47-82e2-e9948acaa846n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>> . >>>>>> >>>>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "TopBraid Suite Users" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to topbraid-user...@googlegroups.com. >>>> >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/topbraid-users/e8a7ccb2-bf2f-4d16-a550-dcd6bb65c27fn%40googlegroups.com >>>> <https://groups.google.com/d/msgid/topbraid-users/e8a7ccb2-bf2f-4d16-a550-dcd6bb65c27fn%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>>> -- >> You received this message because you are subscribed to the Google Groups >> "TopBraid Suite Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to topbraid-users+unsubscr...@googlegroups.com. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/topbraid-users/b1757d75-60c2-4b8f-b74d-d40baffa249cn%40googlegroups.com >> <https://groups.google.com/d/msgid/topbraid-users/b1757d75-60c2-4b8f-b74d-d40baffa249cn%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> -- >> You received this message because you are subscribed to the Google Groups >> "TopBraid Suite Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to topbraid-users+unsubscr...@googlegroups.com. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/topbraid-users/fcebb9cb-e3bd-0355-e5b1-7f56f60e254f%40topquadrant.com >> <https://groups.google.com/d/msgid/topbraid-users/fcebb9cb-e3bd-0355-e5b1-7f56f60e254f%40topquadrant.com?utm_medium=email&utm_source=footer> >> . >> > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/CAGyojU2oPOiHobCR7%2BH8vOvstA4SVz1gq9xsCAZfN3JWSgnX%3DQ%40mail.gmail.com > <https://groups.google.com/d/msgid/topbraid-users/CAGyojU2oPOiHobCR7%2BH8vOvstA4SVz1gq9xsCAZfN3JWSgnX%3DQ%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to a topic in the > Google Groups "TopBraid Suite Users" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/topbraid-users/T6-v3dwOk6M/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/437b44c8-61c3-e0f6-707c-b20adbad35ac%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/437b44c8-61c3-e0f6-707c-b20adbad35ac%40topquadrant.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to topbraid-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/topbraid-users/CAGyojU0D7HXk8YzP1Fr5hZtUgFm%3DmB1qQAORiRYPzEBaj1RSDQ%40mail.gmail.com.
