Multiple Individual records for same indiv. created from a bulk upload

In which Wildbook did the issue occur? ACW

What operating system were you using? Win 10

What web browser were you using? latest Chrome

What is your role on the site? admin

What happened?
Processed a bulk upload that was a replacement for a previous bulk upload. It appears that the same problem that happened last time, happened again this time - there are 4 individuals in the upload spreadsheet and 3 images for each individual (left side, right side and hind view). The import log record looks correct (import ID: 5bef2851), although the column called “w/Indiv” is confusing to me.
But I now have 7 copies of each individual from that spreadsheet:
7 individuals called Milcah - 6 of these have 1 encounter, 1 has 3 encounters
7 individuals called Rundu - 6 of these have 1 encounter, 1 has 3 encounters
7 individuals called Sepupa - 6 of these have 1 encounter, 1 has 3 encounters
7 individuals called Zillah - 6 of these have 1 encounter, 1 has 3 encounters
This is more copies of each than the last time this happened - I believe there were only 4 copies of each last time.
Steps I took: I logged in as admin. I uploaded the jpg images. I then selected the file and uploaded. I then noted that a formula I’d used to extract the indivID string was causing the system pains so I hit the Back button in my browser to go back to the images uploaded confirmation page, then went into my spreadsheet to paste values where there were formulas, re-saved the file with a new name, then uploaded it. This time the upload was all green but I noticed a gap in the “Socialunitname” column for a few of the records so I followed the same steps as above - clicked the browser back button, corrected my spreadsheet again (changing “UO” to “UO_Unknown Origin”), re-saved (with same file name not a new one), then re-uploaded the spreadsheet. It was all green again and this time I hit “commit these results”.
Interestingly, looking at Milcah’s individual record that has 3 encounters associated with it, the change I made to the Social Unit Name, from “UO” to “UO_Unknown Origin”, is there. But in the other records for that dog, which all only have a single encounter associated with them, some show “UO” only and others show “UO_Unknown Origin”.

What did you expect to happen?
1 individual record for each dog with 3 encounter records associated with each. Correct pack name from the file that I “commit(ted) these results” for.

What are some steps we could take to reproduce the issue?
I have forwarded the 2 different spreadsheets I used for the process described above to the email address below. I cc’d @jason because he dealt with this issue when it came up a while back when I first uploaded these same individuals and got duplicates that time. He worked on the issue and thought it was fixed but maybe that was only part of it? Or a different issue?

If this is a bulk import report, send the spreadsheet to services@wildme.org with the email subject line matching your bug report

Assigned ticket WB-664. Testing fix.

I’m going through the process of deleting each of these records so that I can try a re-upload to test the fixes. In that process, I have to remove the individual from its social group. I just found one record that has 2 social groups listed, not start & end dates. See this link: https://africancarnivore.wildbook.org/individuals.jsp?id=de0f64e3-6dd8-48b8-978e-ac6cb21297f1

I will not continue deleting this record until after you’ve had a chance to review - screenshot below. Hopefully this is covered by your fixes but thought you should see it.

I finally completed deleting the 40 or so replicated individual records and did 1 upload of 1 dog + 3 pics (Milcah). It worked as expected!!! So I uploaded the other 3 missing individuals (Rundu, Sepupa and Zillah) and this upload also created only 1 Marked Individual for each individual and assigned the 3 photos as encounters, also as expected.
So I think the latest fix is solid.

Just that issue I reported earlier today in this same stream about the weird appearance of 2 different social group names, both without start and end dates, in one of the replicated records from yesterday’s upload. But I had to move ahead and delete those records to be able to get this done. Finally!!! I love these dogs but I really hope to never have to deal with these 4 dogs (or at least their records) again…

Thanks Jason!