Unable to verify results - merge passage list bug

MJ. Smith

1. Do a Bible Fact search on Mary, mother of Jesus - 154 occurrences if you add them up

2. Do a Bible Search on Mary, mother of Jesus - 133 occurrences

3. Convert each to a Passage List

4. Merge into a Symmetric Difference Passage list

5. Discover merge failed .. so much for this approach to verification No, I didn't expect the results to be equal, I expected to be able to see what was gained by the tagging.

 

Dave Hooton

MJ. Smith said:

1. Do a Bible Fact search on Mary, mother of Jesus - 154 occurrences if you add them up

2. Do a Bible Search on Mary, mother of Jesus - 133 occurrences

1. Fact  "search" from what?

2. Do you mean Clause Search as this gives 133 results?

It raises an interesting question as Clause Search does not allow results to be saved as a Passage List. I had to Export to clipboard and then input that into a Passage List.

MJ. Smith said:

4. Merge into a Symmetric Difference Passage list

5. Discover merge failed .. so much for this approach to verification No, I didn't expect the results to be equal, I expected to be able to see what was gained by the tagging.

To see what one has that the other does not you need a Difference merge. In my case I did a Bible Search for Mary in the NT and saved to a PL. Then I did the Difference merge from the larger Clause Search PL. It doesn't matter that the Bible Search has different persons named "Mary", the difference still shows the advantages gained by tagging for Mary the Mother - an extra 54 verses compared to the ESV.

EDIT: if you do a Difference the other way (from the Bible Search PL) you get all the verses that explicitly mention "Mary".

MJ. Smith

https://community.logos.com/discussion/comment/406926#Comment_406926

Dave Hooton said:

1. Fact  "search" from what?

I'm referring to the Passage List on the Biblical Fact entry for Mary, Mother of Jesus

Dave Hooton said:

2. Do you mean Clause Search as this gives 133 results?

Yes, I messed up on terminology - a clausal search using the search string person:Mary, Mother of Jesus gives 133 results

I think Bradley has already noted the missing "Save to Passage List" but I'm not positive.

Dave Hooton said:

To see what one has that the other does not you need a Difference merge.

But what I wanted was all passages unique to either list rather than assuming that the shorter list was wholly contained in the longer list. That assumption should work but I was trying to prove that it was a reasonable assumption. Unfortunately, what I found was that the merge did not work accurately. I suspect it is related to the fact that one element contains a sub-verse 'a'.

Because I have found both errors and linguistic theory differences in the Biblical People reference data, I was trying to find a logically complete method of finding and identifying the hand-coded entries that aren't clear deixis. See the Barnabas/Silas issue I found today that may be either an error in the display or an error in the coding or another case where Vincent will try to convince me of a linguistic theory which I understand but do not agree with. I suspect that it will cause confusion when L5 goes live ... but will be delighted if I am proven wrong.

Dave Hooton

https://community.logos.com/discussion/comment/406936#Comment_406936

MJ. Smith said:

But what I wanted was all passages unique to either list rather than assuming that the shorter list was wholly contained in the longer list.

If the shorter list is wholly contained in the longer list one of the Difference merges will be zero. But the main idea is to get all the verses unique to the longer list to see the advantages of tagging as I did above. My shorter Bible Search PL was not wholly contained within the Clause PL as it included passages that mentioned the "other" Mary.

MJ. Smith said:

Unfortunately, what I found was that the merge did not work accurately. I suspect it is related to the fact that one element contains a sub-verse 'a'.

Where did that come from? Even when I inserted Mt 1.20a in a PL it produced the difference I expected!

I suspect your Mary Person PL is tripping you up. If it is supposed to represents Bible Facts it is missing a significant number of passages (I did it manually). The difference (symmetric or otherwise) between Facts and Clause is only Lk 1:36 - interestingly missed from the Clause Search.

MJ. Smith

https://community.logos.com/discussion/comment/406958#Comment_406958

Dave Hooton said:

If the shorter list is wholly contained in the longer list one of the Difference merges will be zero.

Depending upon which set you use as base I hope. Otherwise we are developing our own Logos set theory.[;)]

I redid it first thing today after applying an update and removing the old work files. My counts were different today but the symmetric differences still showed the error.

1. Clausal search person:Mary, mother of Jesus

2. Export to clipboard as basis for passage list

3. Biblical facts: Mary, mother of Jesus

4. Select Referred to As, copy, build from clipboard (for some reason I couldn't get build from selection to work)

5. Create Symmetric difference passage list - same result that should have been dropped.

The same error occurs in the Difference when Bible Facts is the base but not if Clausal Search is the base. I'm absolutely sure that this is a bug.

Dave Hooton

https://community.logos.com/discussion/comment/407037#Comment_407037

MJ. Smith said:

Dave Hooton said:

If the shorter list is wholly contained in the longer list one of the Difference merges will be zero.

Depending upon which set you use as base I hope.

Of course!

MJ. Smith said:

The same error occurs in the Difference when Bible Facts is the base but not if Clausal Search is the base. I'm absolutely sure that this is a bug.

I can reproduce the error now I know how to get Facts verses onto the clipboard** (I built it manually at first). Not sure how 1.20a gets there, but it should be deleted from the PL.

** What a kludge, though. Copy is faint whilst Select All is allowed.

MJ. Smith

https://community.logos.com/discussion/comment/407063#Comment_407063

Dave Hooton said:

** What a kludge, though.

[Y][B]