Overly sensitive timestamps create duplicate entries

[MercedesThieme]

Description

In Fieldbook, correcting a value results in a new entry when syncing instead of updating the existing one. This often creates duplicate records only seconds apart.

Expected: Allow defining a time window (e.g., per day) or treat quick edits as updates rather than new entries, keeping only the latest value.

Expected behavior

No response

Field Book version

No response

[trife]

I'm not entirely understanding the issue. Can you upload the debug zip file in an installation that's having this issue? Settings > System > Debug Application

[MercedesThieme]

debug_2026-03-31-04-02-33.zip

Additionally to the reported issue, this trial also seems to have a syncing issue. As fieldbook reported syncing 7 observations, but only 4 were received in the breedbase instance. However, this is the first time that I have observed that not all measurements were transferred.

20260327_test345_phenotypes(8).csv

[trife]

The missing values looks like it might be an issue with breedbase. In your export, different observations from the repeated values are being assigned the same observationDbId which is something that happens server side. @lukasmueller

But for your main issue, I'm still not sure I understand the problem. Two observations for plot 1 were exported (1 and 3) because two observations for plot 1 were collected (13s apart). What are you expecting to happen?

[MercedesThieme]

Thanks for your feedback. I wasn’t able to reproduce the issue on my end.

Initially, I suspected that the duplicated entries might have been caused by editing an existing value. However, it’s more likely that a new value was entered instead of modifying the original one.

Since timestamps were not yet displayed in Fieldbook (as described in the fixed issue #1454), this may have led to some confusion, causing users to create a new entry rather than overwrite the existing one.

[trife]

Sounds good. I'm going to go ahead and close this for now. I think @lukasmueller will need to look into the observationDbId issues a bit more on his end.