[sword-devel] Search bounds, one more time

[Karl Kleinpaste]

"Troy A. Griffitts" <scribe at crosswire.org> writes:
> mkfastmod NETfree
> ./search NETfree "meeting" lk,act
> Acts 5:12
> Acts 13:43
> Acts 20:8
> Acts 24:23
> This seems to be correct.  Can you try examples/cmdline/search and see 
> what it returns for you?

Yes, search at the command line like that works the same way for me.

So I have to wonder how GS' main_do_dialog_search() routine can get
differing results, when its handling of range choice is completely
insensitive to search type, multi-word -vs- optimized.  That is, I get a
wrong result in the optimized case inside GS, when the only difference
from a correctly functioning search is the change from multi-word to
optimized.