My MarkLogic version is 9. I have 2 queries returning expected results when run separately but when I try to combine them I get no result.
My data looks like below, I need to match only ABC, not ABC/* (or ABC/D if this is the searched criteria and in this case, not ABC/D/*)
<root xmlns:ns1="http://ns1">
<ns1:security>
<ns1:elem>ABC</ns1:elem>
<ns1:elem>ABC/D</ns1:elem>
<ns1:elem>ABC/D/E</ns1:elem>
</ns1:security>
</root>
Below code returns 4 results
xquery version "1.0-ml";
import module namespace search = "http://marklogic.com/appservices/search" at "/MarkLogic/appservices/search/search.xqy";
search:resolve(
<cts:path-range-query operator="=" xmlns:ns1="http://ns1">
<cts:path-expression>//ns1:security/ns1:elem</cts:path-expression>
<cts:value xsi:type="xs:string" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">ABC</cts:value>
</cts:path-range-query>
)
and this one only 3 results
xquery version "1.0-ml";
import module namespace search = "http://marklogic.com/appservices/search" at "/MarkLogic/appservices/search/search.xqy";
search:resolve(
<cts:element-query>
<cts:element xmlns:ns1="http://ns1">ns1:security</cts:element>
<cts:element-value-query>
<cts:element xmlns:ns1="http://ns1">ns1:elem</cts:element>
<cts:text xml:lang="en">ABC/*</cts:text>
<cts:option>wildcarded</cts:option>
</cts:element-value-query>
</cts:element-query>
)
So I was expecting to get 1 result when running this
xquery version "1.0-ml";
import module namespace search = "http://marklogic.com/appservices/search" at "/MarkLogic/appservices/search/search.xqy";
search:resolve(
<cts:and-not-query>
<cts:positive>
<cts:path-range-query operator="=" xmlns:ns1="http://ns1">
<cts:path-expression>//ns1:security/ns1:elem</cts:path-expression>
<cts:value xsi:type="xs:string" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">ABC</cts:value>
</cts:path-range-query>
</cts:positive>
<cts:negative>
<cts:element-query>
<cts:element xmlns:ns1="http://ns1">ns1:security</cts:element>
<cts:element-value-query>
<cts:element xmlns:ns1="http://ns1">ns1:elem</cts:element>
<cts:text xml:lang="en">ABC/*</cts:text>
<cts:option>wildcarded</cts:option>
</cts:element-value-query>
</cts:element-query>
</cts:negative>
</cts:and-not-query>
)
I tried also this but same result
xquery version "1.0-ml";
import module namespace search = "http://marklogic.com/appservices/search" at "/MarkLogic/appservices/search/search.xqy";
search:resolve(
<cts:and-query>
<cts:path-range-query operator="=" xmlns:ns1="http://ns1">
<cts:path-expression>//ns1:security/ns1:elem</cts:path-expression>
<cts:value xsi:type="xs:string" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">ABC</cts:value>
</cts:path-range-query>
<cts:not-query>
<cts:element-query>
<cts:element xmlns:ns1="http://ns1">ns1:security</cts:element>
<cts:element-value-query>
<cts:element xmlns:ns1="http://ns1">ns1:elem</cts:element>
<cts:text xml:lang="en">ABC/*</cts:text>
<cts:option>wildcarded</cts:option>
</cts:element-value-query>
</cts:element-query>
</cts:not-query>
</cts:and-query>
)
Is this a known bug ? Am I doing something wrong here ? Any help appreciated :)
Search plan
<search:response snippet-format="snippet" total="0" start="1" page-length="10" xmlns:search="http://marklogic.com/appservices/search">
<search:plan>
<qry:query-plan xmlns:qry="http://marklogic.com/cts/query">
<qry:expr-trace>impl:apply-search(map:map(<map:map xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" .../>), "xdmp:plan", fn:false())</qry:expr-trace>
<qry:info-trace>Analyzing path for search: fn:collection()</qry:info-trace>
<qry:info-trace>Step 1 is searchable: fn:collection()</qry:info-trace>
<qry:info-trace>Path is fully searchable.</qry:info-trace>
<qry:info-trace>Gathering constraints.</qry:info-trace>
<qry:info-trace>Comparison contributed string range value constraint: //ns1:security/ns1:elem = "CTPA"</qry:info-trace>
<qry:partial-plan>
<qry:range-query weight="0" min-occurs="1" max-occurs="4294967295" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<qry:key>12785637774270294680</qry:key>
<qry:annotation>path(//ns1:security/ns1:elem)</qry:annotation>
<qry:lower-bound xsi:type="xs:string" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CTPA</qry:lower-bound>
<qry:upper-bound xsi:type="xs:string" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CTPA</qry:upper-bound>
</qry:range-query>
</qry:partial-plan>
<qry:elem-word-trace text="CTPA" elem-name="elem" elem-uri="http://ns1">
<qry:key>6185531260368494803</qry:key>
</qry:elem-word-trace>
<qry:info-trace>Search query contributed 1 constraint: cts:and-query((cts:path-range-query("//ns1:security/ns1:elem", "=", "CTPA", ("collation=http://marklogic.com/collation/"), 1), cts:not-query(cts:element-query(xs:QName("ns1:security"), cts:element-value-query(xs:QName("ns1:elem"), "CTPA/*", ("wildcarded","lang=en"), 1), ()), 1)), ())</qry:info-trace>
<qry:partial-plan>
<qry:and-not-two-queries>
<qry:range-query weight="0" min-occurs="1" max-occurs="4294967295" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<qry:key>12785637774270294680</qry:key>
<qry:annotation>path(//ns1:security/ns1:elem)</qry:annotation>
<qry:lower-bound xsi:type="xs:string" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CTPA</qry:lower-bound>
<qry:upper-bound xsi:type="xs:string" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CTPA</qry:upper-bound>
</qry:range-query>
<qry:and-two-queries>
<qry:or-two-queries>
<qry:term-query weight="0">
<qry:key>17253116673510471442</qry:key>
<qry:annotation>element(ns1:security)</qry:annotation>
</qry:term-query>
<qry:term-query weight="0">
<qry:key>12929598538251878498</qry:key>
<qry:annotation>link-child(descendant(element(ns1:security)))</qry:annotation>
</qry:term-query>
</qry:or-two-queries>
<qry:term-query weight="1">
<qry:key>6185531260368494803</qry:key>
<qry:annotation>element(http://one.oecd.org/ns1:elem,word("CTPA"))</qry:annotation>
</qry:term-query>
</qry:and-two-queries>
</qry:and-not-two-queries>
</qry:partial-plan>
<qry:info-trace>Executing search.</qry:info-trace>
<qry:ordering/>
<qry:final-plan>
<qry:and-query>
<qry:and-not-two-queries>
<qry:range-query weight="0" min-occurs="1" max-occurs="4294967295" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<qry:key>12785637774270294680</qry:key>
<qry:annotation>path(//ns1:security/ns1:elem)</qry:annotation>
<qry:lower-bound xsi:type="xs:string" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CTPA</qry:lower-bound>
<qry:upper-bound xsi:type="xs:string" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">CTPA</qry:upper-bound>
</qry:range-query>
<qry:and-two-queries>
<qry:or-two-queries>
<qry:term-query weight="0">
<qry:key>17253116673510471442</qry:key>
<qry:annotation>element(ns1:security)</qry:annotation>
</qry:term-query>
<qry:term-query weight="0">
<qry:key>12929598538251878498</qry:key>
<qry:annotation>link-child(descendant(element(ns1:security)))</qry:annotation>
</qry:term-query>
</qry:or-two-queries>
<qry:term-query weight="1">
<qry:key>6185531260368494803</qry:key>
<qry:annotation>element(http://one.oecd.org/ns1:elem,word("CTPA"))</qry:annotation>
</qry:term-query>
</qry:and-two-queries>
</qry:and-not-two-queries>
</qry:and-query>
</qry:final-plan>
<qry:info-trace>Selected 0 fragments to filter</qry:info-trace>
<qry:result estimate="0"/>
</qry:query-plan>
</search:plan>
<search:metrics>
<search:query-resolution-time>PT0.001512S</search:query-resolution-time>
<search:total-time>PT0.0024561S</search:total-time>
</search:metrics>
</search:response>
So, I believe the issue is that the negative query, looking to exclude the
ns1:elem
elements that have values starting withCTPA/
is producing a term-query for the wordCTPA
. From the plan, we see:inside of the not-query:
Now, if that were a positive query, you would scoop up the desired items as well as some false positives, and then they could be filtered. But when you negate it, those false positives (the ones that are only the value
CTPA
) are included in the set to exclude, so everything is eliminated.There are database options such as trailing wildcard searches and three character searches that can help with query resolution, as well as some other wildcard indexing options that can help with wildcarded queries. Read more in the documentation for Understanding and Using Wildcard Searches.
I enabled both of those options mentioned above, and changed the word-query to a value-query with the
punctuation-sensitive
andwildcarded
options applied:and then it produced this query in my plan:
And is now producing the expected results, finding those with
ABC
and not those that haveABC/*
.