qe: sync @default(now()) and @updatedAt within a request #3200
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
tomhoule
force-pushed
the
qe/per-request-timestamps
branch
from
0b79960
to
46e1d7b
Compare
tomhoule
commented
Sep 20, 2022
tomhoule
commented
Sep 20, 2022
tomhoule
force-pushed
the
qe/per-request-timestamps
branch
from
46e1d7b
to
3045ebf
Compare
There are two points of difficulty here: - Using the same PrismaValue::DateTime as the value of `now()` and `@updatedAt` for the duration of a request. We achieve that through a task local. Alternatives: passing the context through explicitly. I tried, it's a massive refactoring. - The loss of precision due to the `From<PrismaValue>` impl for `QueryValue`. This is fixed by adding a `QueryValue::DateTime` variant that preserves the original precision.
tomhoule
force-pushed
the
qe/per-request-timestamps
branch
from
3045ebf
to
5918f1e
Compare
pimeys
approved these changes
Sep 20, 2022
query-engine/connector-test-kit-rs/query-engine-tests/tests/new/regressions/prisma_12572.rs
Show resolved
Hide resolved
| trace_id, | ||
| self.force_transactions, | ||
| ) | ||
| super::with_request_now(async move { |
There was a problem hiding this comment.
I know this is the best way for now, I'm a bit worried though somebody adds a new operation, forgets this and we find it out way too late.
There was a problem hiding this comment.
As noted above, it will panic very early and unambiguously. One test with an @default(now()) that uses the new method is coverage enough. We both agree it's not the ideal solution, but I'm not too worried about this specific problem.
| /// A timestamp that should be the `NOW()` value for the whole duration of a request. So all | ||
| /// `@default(now())` and `@updatedAt` should use it. | ||
| /// | ||
| /// That panics if REQUEST_NOW has not been set with with_request_now(). |
There was a problem hiding this comment.
Oh, this is cool actually. Better to panic than write a wrong date in a new operation!
There was a problem hiding this comment.
The only problem I see is somebody adds a new operation, does not test it with now() or updatedAt and the panic is triggered by a user.
| @@ -233,7 +228,7 @@ impl QueryDocumentParser { | |||
| } | |||
|
|
|||
| /// Attempts to parse given query value into a concrete PrismaValue based on given scalar type. | |||
| pub fn parse_scalar( | |||
next PR! |
|
closes #3200 |
|
Does |
SevInf
added a commit
that referenced
this pull request
Jan 16, 2024
We implemented introduced consitent `request_now` value in #3200, however, we've opted out of thread local value if there is no active tokio handle. This is problem for WASM build, since it does not have use an actual tokio runtime. Replacing handle check with `LocalKey::try_with` allows us to use task local variable without the runtime, while still preserving fallback for immediate value if task local is not set. Fix prisma/team-orm#645
SevInf
added a commit
that referenced
this pull request
Jan 16, 2024
We implemented introduced consitent `request_now` value in #3200, however, we've opted out of thread local value if there is no active tokio handle. This is problem for WASM build, since it does not have use an actual tokio runtime. Replacing handle check with `LocalKey::try_with` allows us to use task local variable without the runtime, while still preserving fallback for immediate value if task local is not set. Fix prisma/team-orm#645
SevInf
added a commit
that referenced
this pull request
Jan 17, 2024
* qe-wasm: Fix consistent `now()` value We implemented introduced consitent `request_now` value in #3200, however, we've opted out of thread local value if there is no active tokio handle. This is problem for WASM build, since it does not have use an actual tokio runtime. Replacing handle check with `LocalKey::try_with` allows us to use task local variable without the runtime, while still preserving fallback for immediate value if task local is not set. Fix prisma/team-orm#645 * Fix formatting * Reword comment
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Alternatives: passing the context through explicitly. I tried, it's a massive refactoring.
There are two points of difficulty here:
now()and@updatedAtfor the duration of a request. We achieve that through a task local.From<PrismaValue>impl forQueryValue. This is fixed by adding aQueryValue::DateTimevariant that preserves the original precision.