How are characters being counted for NEAR/WITHIN 48 CHARS search?

Kiyah

I'm trying to figure out why the following search didn't return this entry (θάλεια) in LSJ:

θάλεια NEAR banquet (which is the same as the search θάλεια WITHIN 48 CHARS banquet)

The terms seem to be within 48 characters (see screenshot below) of each other. I had MS Word count them and counted them myself. Logos seems to be counting the characters differently.

The search doesn't work until I type the following:

θάλεια WITHIN 50 CHARS banquet 

Or

θάλεια WITHIN 9 WORDS banquet 

Can someone explain to me how Logos is counting characters and why my original search didn't work? Does it count from/to the beginning of each word, the middle, or the end? No matter how I tried to count them I could never get 50 characters. The terms are so close in the lexicon I would have thought a NEAR search would have sufficed.

Dave Hooton

Can someone explain to me how Logos is counting characters and why my original search didn't work?

The minimum space between 2 words is 2 CHARS  (one space and the first letter)

• θάλεια to θᾰ is correctly 4 CHARS  as the comma and [ characters are counted
• θᾰ to  ἡ  is 4 CHARS as  ] and the comma are counted
• θάλεια to ἡ is unexpectedly 10 CHARS (not 9), and then
• θάλεια to rich is 14 CHARS  (not 13)

And i have no explanation for the discrepancy.

Andrew Batishko

If I had to guess it would be something related to unicode normalization, but I don't know for certain why the range is not as expected. Ultimately though, I'm not sure the answer matters. Even if this particular instance didn't have a couple character discrepancy, there could easily be another case that is just a little bigger than any range you set. With this approach you are always going to need to go bigger than you want and then manually filter out the unneeded ones.

Depending on what you are trying to accomplish, perhaps these might be better search patterns for your lexicon searches:

• headword:θάλεια INTERSECTS banquet
• θάλεια banquet