You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Parameter to capture processing start datetime #37
Hello,
I am currently working on the project Reference System for ESA.
I would like to know whether it would be possible to introduce an additional parameter to capture the processing:start_datetime.
I understand that processing:datetime corresponds to the end of the processing.
Thank you.
A dedicated processing:start_datetime feels like a fairly niche addition. Could you describe a concrete use case for this information, particularly in a search context? Understanding how users would filter or query on processing start time (vs. end time) would help evaluate whether this belongs in the extension.
The existing processing:datetime field does not correspond to a strictly defined moment within the processing pipeline. It is primarily used to filter items based on when they were processed (broadly). It is not explicitly the end of processing, so clarifying that distinction would also be valuable in any proposal.
If there is a strong use case, we would welcome a PR that includes a clear field description, the rationale, and ideally a reference to how it is used in the ESA Reference System. That will help the community evaluate the proposal concretely.
We would like to extract the production associated with a processing run.
My initial idea was to retrieve the production associated with the derived-from inputs for a given collection. However, since the same input can be processed multiple times, we need a way to distinguish between different processing runs. At first, I considered using the processing start parameter (as mentioned in my previous message). After further cross-checking, I believe the processing-execution field would be a more efficient way to uniquely identify each run.
One final remark: it could be useful to clarify whether processing:datetime refers to the start time or the end time of the processing. That said, having both timestamps would mainly support performance analysis, which is not necessarily the primary objective of this extension.
Activity
emmanuelmathot commentedon Apr 29, 2026
Thank you for raising this @pcuq-ads.
A few thoughts on the proposal:
A dedicated
processing:start_datetimefeels like a fairly niche addition. Could you describe a concrete use case for this information, particularly in a search context? Understanding how users would filter or query on processing start time (vs. end time) would help evaluate whether this belongs in the extension.The existing
processing:datetimefield does not correspond to a strictly defined moment within the processing pipeline. It is primarily used to filter items based on when they were processed (broadly). It is not explicitly the end of processing, so clarifying that distinction would also be valuable in any proposal.If there is a strong use case, we would welcome a PR that includes a clear field description, the rationale, and ideally a reference to how it is used in the ESA Reference System. That will help the community evaluate the proposal concretely.
pcuq-ads commentedon May 7, 2026
Hello,
We would like to extract the production associated with a processing run.
My initial idea was to retrieve the production associated with the
derived-frominputs for a given collection. However, since the same input can be processed multiple times, we need a way to distinguish between different processing runs. At first, I considered using the processing start parameter (as mentioned in my previous message). After further cross-checking, I believe theprocessing-executionfield would be a more efficient way to uniquely identify each run.One final remark: it could be useful to clarify whether
processing:datetimerefers to the start time or the end time of the processing. That said, having both timestamps would mainly support performance analysis, which is not necessarily the primary objective of this extension.Regards