Bug: 4.2 beta 7: Bible datatype searching not behaving correctly

Searching for <bible = John 3:1-36> ought to return results for a simple John 3, but it doesn't, as this search demonstrates: <bible = John 3> NOTEQUALS <bible = John 3:1-36> (there should be no results).

Even more worryingly, <bible ~ John 3:1-36> doesn't pick up any references for John 3, either. superset does work as expected though, as does subset.

It's true for any chapter in the Bible, not just John 3.

Find more posts tagged with

Comments

Bradley Grainger (Logos)

There are enough edge cases (versification differences, Psalm titles, introductory material, etc.) that we decided that (in the general case) "chapter X" is not exactly the same as "X:1-end"; the system considers a reference to the entire chapter to be a slightly larger superset of verses 1 to the end. (To make matters worse, some old books (particularly LLS era) encode "John 3" internally as "John 3:1-36", while new books encode it as written.

Thus an exact Bible search for John 3 or John 3:1-36 will not find matches for the other. Unfortunately, we don't have a convenient syntax for specifying "find references to John 3 or close enough". Obviously, you can use "<= Jn 3>, <= Jn 3:1-36>" to find the union of both queries, but it's a little awkward. Furthermore, for the Psalms, this wouldn't always work, because neither <=Ps 3> nor <= Ps 3:1-8> would match references to Ps 3:title-8 (aka BHS Ps 3:1-9).

You could run something like "<bible subset Ps 3> ANDEQUALS <bible superset Ps 3:1-8>" to find all possible ways to express a reference to this chapter; unfortunately, the implementation of that search is somewhat inefficient (on my system it takes about 100x longer to run than just <= Ps 3> or <= Ps 3:1-8>).

It seems that (just like for versification differences) we should just assume we know more about the internal encoding of resources than you do [:)] and automatically expand the search as necessary (probably using your default Bible's versemap) so that you don't need to know that Jn 3 and Jn 3:1-36 are treated differently internally. I'll file a case.