-
Notifications
You must be signed in to change notification settings - Fork 1.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Spanner gapic migration #3130
Spanner gapic migration #3130
Conversation
Fixes googleapis#3015. Looks like the previous version wasn't annotated properly or something.
… store it in a Table (googleapis#2974)
We create fewer blobs since the basic case is already tested by other ITs. Fixes googleapis#3019, hopefully.
There has been no way to use it since Aug 2017 and removing it will make moving the GAPIC stub easier.
We create fewer blobs since the basic case is already tested by other ITs. Also slightly extend the timeout. Fixes googleapis#3019, hopefully.
…leapis#3046) This is in line with documentation at https://cloud.google.com/storage/docs/json_api/v1/how-tos/batch.
We had two constants called PAGE_SIZE. This PR collapses them into one and tests building requests.
Regenerate Bigtable GAPIC stubs to enable resource headers
This is for forward-compatibility concern. If PubSub creates a new format for subscriptions, we won't be able to parse the name into ProjectSubscriptionName.
This is a bit ugly: the autogenerate bigtable data api doesn't reference instances, only tables. So the resource name for instances is never generated. Up until this point I've been borrowing resource name from admin, which is really ugly. This PR simply copies the admin InstanceName to data's models package. At some point in the future when GAPIC implements common resource names, this can be cleaned up
* add a smoke-test mvn profile * remove excludes in pom
* let surefire not run smoke tests * fix comment
So that external systems that need to ensure that all messages have been published can clear a batch.
The prefetcher can load data on another thread in between calls, and it can intelligently use its existing buffer to deal with small seeks without having to fetch again.
Previously we send ack/modack requests on the StreamingPull streams. However, if pull responses are buffered, gRPC cannot promptly tell us that the stream is broken, so we'll keep sending requests on the dead stream. This commit implements a temporary solution. We send requests by unary RPCs instead and only use streaming for receiving messages. Load test shows no regression. However, modacks seem to take longer to take effect. The timeout for nack test had to be increased.
We also add JobId.Builder since the static constructors are getting unwieldy. The property isn't threaded through in the right places yet, so we keep it package-private for now.
This updates maven enforcer to the latest stable (and recommended) release. It fixes a compilation failure in java 9.
So there's good news and bad news. 👍 The good news is that everyone that needs to sign a CLA (the pull request submitter and all commit authors) have done so. Everything is all good there. 😕 The bad news is that it appears that one or more commits were authored or co-authored by someone other than the pull request submitter. We need to confirm that all authors are ok with their commits being contributed to this project. Please have them confirm that here in the pull request. Note to project maintainer: This is a terminal state, meaning the |
Merge master into spanner-gapic-migration