Long-standing bug: Inconsistent search counts

Page 1 of 1 (9 items)
This post has 8 Replies | 2 Followers

Posts 13360
Forum MVP
Mark Barnes | Forum Activity | Posted: Fri, Mar 25 2011 3:14 AM | Locked

I know this has been an issue for a long time, but today I was bitten again by this bug. I find it extremely confusing, and I hate to think what a Logos newbie would make of it.

Basically the problem is that using ANDEQUALS doubles the search count. So if an ANDEQUALS search found 25 results in 20 verses, it would report it as 50 results in 40 verses.

It bit me today because I was trying to get some ratios of how often πνεῦμα referred to God, and how often to something else. Roughly speaking, I knew I could match case and do

lemma:πνεῦμα ANDEQUALS Spirit

and then lemma:πνεῦμα ANDEQUALS spirit

After initially forgetting about the bug, I eventually remembered to divide all the results in half. But then I realised my second search term was inadequate, and it should be lemma:πνεῦμα NOTEQUALS Spirit. So again, I compiled by statistics for different verse ranges, dutifully dividing my results in half - before realising that you don't need to do this with NOTEQUALS, only with ANDEQUALS! And goodness only knows what I'd do if I ended up with an ANDEQUALS and a NOTEQUALS in the same search!

So, please, in 4.3, fix this! Thanks.

Posts 726
Theolobias | Forum Activity | Replied: Fri, Mar 25 2011 3:25 AM | Locked

Mark Barnes:
So, please, in 4.3, fix this! Thanks.

Yes, please!

MS Windows 7 Home Premium 32-bit SP1 | Intel Core 2 Duo CPU T5250 @ 1.5GHz, 2.0GB RAM, ATI Radeon XPress 1250 | theolobias.de

Posts 912
David Knoll | Forum Activity | Replied: Fri, Mar 25 2011 8:26 AM | Locked

Yes

Posts 3663
BillS | Forum Activity | Replied: Fri, Mar 25 2011 9:37 AM | Locked

Mark Barnes:

So, please, in 4.3, fix this! Thanks.

Yes If there's anything we HAVE to be able to trust, it's search results...

Grace & Peace,
Bill


Asus GF63 8RD, I-7 8850H, 32GB RAM, 1TB SSD, 2TB HDD, NVIDIA GTX 1050Max
Samsung S9+, 64GB
Fire 10HD 64GB 7th Gen

Posts 8022
LogosEmployee
Bradley Grainger (Faithlife) | Forum Activity | Replied: Mon, Mar 28 2011 11:40 AM | Locked

Mark Barnes:

Basically the problem is that using ANDEQUALS doubles the search count. So if an ANDEQUALS search found 25 results in 20 verses, it would report it as 50 results in 40 verses.

Did you mean "50 results in 20 verses", or are you actually seeing the verse count double?

Posts 24959
Forum MVP
Dave Hooton | Forum Activity | Replied: Mon, Mar 28 2011 4:40 PM | Locked

Bradley Grainger:
Did you mean "50 results in 20 verses"

20 verses was meant.

But the result depends on the number of matching interlinear lines. So in LGNTI  <Lemma = lbs/el/σιγάω> ANDEQUALS silent  produces 25 results in 9 verses rather than 9 results in 9 verses!

Dave
===

Windows & Android

Posts 1471
Angela Murashov | Forum Activity | Replied: Wed, Aug 14 2013 4:57 PM | Locked

Mark Barnes:
Basically the problem is that using ANDEQUALS doubles the search count. So if an ANDEQUALS search found 25 results in 20 verses, it would report it as 50 results in 40 verses.

This will be fixed in 5.1a Beta 3. 

Posts 24959
Forum MVP
Dave Hooton | Forum Activity | Replied: Wed, Aug 14 2013 5:03 PM | Locked

Angela Murashov:

Mark Barnes:
Basically the problem is that using ANDEQUALS doubles the search count. So if an ANDEQUALS search found 25 results in 20 verses, it would report it as 50 results in 40 verses.

This will be fixed in 5.1a Beta 3. 

As long as the fix doesn't affect the verse count!

Dave
===

Windows & Android

Posts 24959
Forum MVP
Dave Hooton | Forum Activity | Replied: Fri, Aug 16 2013 3:36 PM | Locked

Angela Murashov:

Mark Barnes:
Basically the problem is that using ANDEQUALS doubles the search count...

This will be fixed in 5.1a Beta 3. 

Indeed, it is finally consistent with NOTEQUALS.

Dave
===

Windows & Android

Page 1 of 1 (9 items) | RSS