Updates for 20.0.0 release #209
Conversation
There was a problem hiding this comment.
Changes look good. Feel free to merge when ready.
While the raw data structure is named ttl, we might want to consider naming marts tables something a little more descriptive than ttl. 🤷♀️ that or we'll need to have robust docs so that end users know exactly what that means
| @@ -13,7 +13,7 @@ COPY . . | |||
| RUN go build -v -o /usr/local/bin ./... | |||
|
|
|||
| # stage 2: runtime enviroment | |||
| FROM stellar/stellar-core:20.0.0-1504.rc2.22088c1f2.focal | |||
| FROM chowbao/stellar-core:20.0.0-1615.617729910.focal | |||
There was a problem hiding this comment.
ah, still no public docker image?
There was a problem hiding this comment.
There is stellar/unsafe-stellar-core:20.0.0-1615.617729910.focal that horizon is using but the unsafe-stellar-core part is new. Not sure if they are going to release a normal stellar-core
| case 2: | ||
| return l.MustV2().TxProcessing |
There was a problem hiding this comment.
did transactionmetav2 get deleted? Confirming this was intentional.
| details["type"] = "extend_footprint_ttl" | ||
| details["extend_to"] = op.ExtendTo |
There was a problem hiding this comment.
Since the keys in the object were renamed, how do you think making the updates to the details object in BQ is going to go? Curious if we'll be able to drop the old keys, or if we will just have to append the new fields
There was a problem hiding this comment.
It just appends new fields. Bigquery doesn't seem to support dropping columns within a record. One hacky is to
- Create new record column (details_new)
- Copy old record data into new record column without the expire column (details --> details_new)
- Delete old record column (details)
- Create a record column (empty details)
- Copy data into new record column (details_new --> details)
There was a problem hiding this comment.
meh, I don't think it's worth the pain to do the drop and recopy (plus that would be pretty expensive) I already don't love the history_operations schema, and would prefer if we're making updates to the column to just convert to a json blob instead.
I agree with renaming it downstream. We could rename it here in stellar-etl but not sure if we would want cause it is how the raw data is named. |
Updates for futurenet release of 20.0.0
Expirationis renamed tottl