{246481476177: [1],
246473461035: [2],
246504296705: [3, 357],
246479326503: [4],
246504254547: [5, 9527, 52134],
246487529880: [6],
246509881747: [7],
....}
173 - 246585781150
386 - 246585781150
8666 - 246585781150
9056 - 246585781150
9832 - 246585781150
9839 - 246585781150
(1-2 min break)
22 - 246585850365
24 - 246585850365
954 - 246585850365
1636 - 246585850365
3495 - 246585850365
8465 - 246585850365
9057 - 246585850365
9060 - 246585850365
64018 - 246585850365
initial_version_number = get_max_version_number_from("trains?")
wait(X min) -- during this time it will gather data for initial_version_number (I think this is limited to 2500 object?, I can also have a function to check this that it doesnt exceed e.g. 2000 object)
# After waiting, I can query for the data using initial_version_number
data = get_data_using("trains?version=initial_version_number")
After getting the data, I can compare initial_version_number and version numbers came with data object. E.g. Lets say initial_version_number was 15 and I received [1, 2, 5, 6, 7, 12, 15, 17, 18] version numbers with data object. I can find the next version number after 15 which is 17 (this should also contain further data after this), and do the similar stuff again (wait a bit to gather next data for that version number (17 this time), fetch it and go with the next version number).
Finally, after gathering all the data (there will be duplicates for the trains), I can use max version number for each train to get latest data for that train.
Hi,
I drew this simple diagram for you. Hope it helps.
When using ?version parameter, you should never get entries with older version number than given.