Offline/XLS Upload Issues

[kolatubosun]

Currently, when a lexicographer decides to work on entries offline using our Excel sheet upload functionality, the names he/she uploads replaces whatever we have in the dictionary (even if the current dictionary entries are much fuller/better entries).

I suggest two ways to deal with this problem:

1. Make it such that when an upload is done, only entries that are NOT in the dictionary are uploaded, and a message given to the user that so-and-so number of entries were not uploaded because they already exist.
2. Follow-up to Add test coverage #1 above, let all the uploads done via Excel NOT go directly into the public index, but directly to the "unpublished" entries list, where they can be vetted again before being published.

I made the priority for this high because I haven't been able to use the offline upload functionality because of this, and all other ways of adding entries are long and tedious. As soon as this is solved, we can decentralise the lexicography work and more volunteers can do their work offline, increasing efficiencey.

[kolatubosun]

Fixing this issue will accelerate lexicographic work, since we now have thousands of names in excel sheets waiting to be uploaded, vetted, and published. Help, someone!

[kolatubosun]

Hi @dadepo and @temiyemi How long will it take to fix this particular issue? It could impact progress pretty significantly.

[dadepo]

@kolatubosun I cant tell as I do not have plans to look into this issue anytime soon.

But if I remember correctly the behaviour you described i.e. "the names he/she uploads replaces whatever we have in the dictionary (even if the current dictionary entries are much fuller/better entries)" was not what was implemented.

The implementation was that when a name is uploaded, and an entry with that name already exists in the dictionary, then it appears as a duplicate. It is then up to the lexicographers to view all the duplicates and manually update the existing name with fields from the one that was just uploaded, or to delete the duplicate if no update is necessary.

If this is not the current behaviour then it might be a bug that needs fixing.

[kolatubosun]

"The implementation was that when a name is uploaded, and an entry with that name already exists in the dictionary, then it appears as a duplicate. It is then up to the lexicographers to view all the duplicates and manually update the existing name with fields from the one that was just uploaded, or to delete the duplicate if no update is necessary."

Oh okay. I didn't remember that this was the current behaviour. But even then, how would one know which ones were duplicated and which were not? Is there a list generated that one can use as a guide to find which ones need deletion?

Actually, if I could just get the uploaded names to go into the queued list rather than a published list, I think I'll be fine. It will be easy to find duplicate entries there since they won't publish.

[dadepo]

"how would one know which ones were duplicated and which were not?"
For this I would have to check, I know for sure the old dashboard had this. And on every entry, there was a section that lists the duplicates. Can't remember if we ported this functionality to the new dashboard. Maybe @temiyemi has a better memory of this.

[temiyemi]

Not sure I recall. I don't think we did.