David Lutterkort
David Lutterkort
I wonder if the current or the new behavior is really what users want. IMHO, names are mostly annoying, and I think it would be nicer/easier if `graphman deploy `...
The fix might be simple: the whole of copying private data sources runs outside of a transaction. We should just wrap all of it in a txn.
Copying private data sources should be very quick; at the very worst, it's a table with 100k rows, but usually much less. The actual data copying, which can take days,...
True; I just looked through the code again, and it does one `insert` statement per row, which will be very slow for these numbers of data sources. We'll also need...
Can you explain why this is needed, what it's supposed to achieve, and how we would use this workflow?
It defaults to `2*reorg_threshold`. Is `ETHEREUM_REORG_THRESHOLD` set?
I noticed a few things in the GraphQL schema (which is well documented): * the schema has a `QueryRoot` and `MutationRoot`. IIRC, at least for queries, that must be called...
For the general approach how to expose the same business logic in the CLI and API, if you have a function `biz` that implements the business logic (I'll write as...
Interesting. This used to fail CI; do you have any idea where the reorg threshold is set to 0 when we run tests?
Digging into one such example on Avalanche, this seems to be caused by blocks having the same timestamps. For example, on Avalanche blocks 51482738 and 51482739 both have the timestamp...