"live" time series query
I'm not currently using TanStack Query, but I would like to! and want to know if it could fit my current use case. I'll try to be as concise as possible.
I have a database table called
which is unique on timestamp+case_id
And another table called
which is unique on case_id only
A user on the frontend (vite react) selects a specific case, and updates its
Another server sees the
I would like the frontend to show the case's data as soon as it's available, instead of pinwheeling for ~10 mins; so meanwhile the frontend incrementally queries the
{
case_id: number,
starting_data_index: number,
num_results: number
}
And the backend's response is like:
{
results: [ data, data, ...],
actual_num_results: number
}
The response may not include the total
Can something similar be achieved in TanStack query? I can change the backend functionality, but I can't make the "other server" insert records faster, and I'd like to keep the incremental data loading if possible.
Sorry for being long winded, and thanks!
I have a database table called
case_data with columns like sowhich is unique on timestamp+case_id
And another table called
cases with columns like sowhich is unique on case_id only
A user on the frontend (vite react) selects a specific case, and updates its
status to indicate a request to be processed.Another server sees the
status update, and begins appending records to case_data starting with an earliest timestamp and marching forward monotonically. The process takes ~10 mins let's say, and results in a total number expected_num_data_records of records being appended to case_data.I would like the frontend to show the case's data as soon as it's available, instead of pinwheeling for ~10 mins; so meanwhile the frontend incrementally queries the
case_data table in chunks, starting from an earliest expected timestamp. Currently the structure of the requests are:{
case_id: number,
starting_data_index: number,
num_results: number
}
And the backend's response is like:
{
results: [ data, data, ...],
actual_num_results: number
}
The response may not include the total
num_results as the other server may not have inserted those yet. However the results are guaranteed "coherent" between starting_data_index and going for actual_num_results; the frontend appends the results to a buffer, and repeats, setting a new starting index to starting_data_index + actual_num_results, continuing until expected_num_data_records have been received.Can something similar be achieved in TanStack query? I can change the backend functionality, but I can't make the "other server" insert records faster, and I'd like to keep the incremental data loading if possible.
Sorry for being long winded, and thanks!
