[sword-devel] Searching for entity attributes

[David "Judah's Shadow" Blue]

On Monday, May 6, 2024 12:17:38 PM EDT you wrote:
> On Friday, May 3, 2024 6:00:59 PM EDT Troy A. Griffitts wrote:
> > Have a look in sword/examples/cmdline/search.cpp  We have comments near
> > the top how to do a strongs search with entry attributes, and you can
> > experiment easily in there.
> > 
> > https://crosswire.org/svn/sword/trunk/examples/cmdline/search.cpp
> 
> Ok, I've got it working...sort of. So here's where we run into the 0 padding
> issue from the OSHB thread, when I search for, say, alpha in the NASGreek
> lexicon one of the results is 05598 (omega, which matches because it has
> the word alphabet in the entry). If I then search for G05598 in the NASB I
> get no results. However, if I search for G5598 I get results. So what
> appears to be happening is that the NASGreek module is returning a padded 0
> (either encoded that way of via the SWLD::strongsPad method I'm not sure
> which) but the NASB module doesn't have the padding in its attributes. And
> from what came up in the OSHB thread, there isn't a consistent way of
> encoding the numbers in the bible modules, and SWLD::strongsPad isn't being
> invoked in the search on the lemmas.
> 
> I only knew to strip the padded 0 off because I used diatheke to see how the
> numbers were encoded in the module. So, there isn't a way for my users to
> know if they need to include the leading 0 from the lexicon module or not
> in their search. Unless I'm missing something.

Based on further testing and using diatheke for comparison, the KJV is 
confusingly 0 padded for Hebrew words in the Old Testament but not 0 padded 
for Greek words in the New Testament. So I'm not certain at all on how to tell 
users to search for entries.