Thunderstone Support Forums

My query is "Minn* West High schools".

I am trying to locate 'Minnesota high school', 'Minnetonka high school' or 'Minneapolis high school'.

Instead, "Minn* West High schools" is locating extra words after Minn* rooted words. ('Minnesota East West High School' for example). It seems like within double quotes, "Minn* High" behaves like 'Minn*High' (without double quotes). Is this an expected behavior for wildcard searches within double quotes?

I have <apicp exactphrase "on">
and <apicp minwordlen 5>

Yes, this is expected (historically defined) behavior: a wildcard inside a phrase (other than the last word) can span words, up to 80 characters away. (I presume it's a typo that `west' is missing from your match examples; otherwise they would not match anyway.)

You could split the query like this:

Minn* "west high school"

and use LIKEP with likeporder and likepproximity turned up and the other likep* settings down; there's no guarantee they'll be adjacent (as with your original query) but at least closer ones will be returned first.

Another more verbose but more accurate solution is to use a paren list that contains phrases:

(Minnesota West High School,
Minnetonka West High School,
Minneapolis West High School)

(no newlines). You lose the wildcard, but gain the adjacency of phrase words.

Correct me if I am wrong but isn't [Minn* "west high school"] equivalent to [Minn*] and ["west high school"]? If it is then it won't find the entire phrase in a sequence.

According to the manual (http://www.thunderstone.com/site/texisman/node204.html), prop* will match property, proper,propane etc. and 'Pres*Busto' will match 'President Jorge Busto'. It doesn't mention the expected behavior for a wildcard within double quotes. Is there a way to do an exactphrase search for ["Minn* High School"] and retain the sequence?

* will span up to 80 characters whether word boundaries or not. It is not limited to a "word". So probably not the way you want.