refactor(iota-transactional-test-runner): document run-graphql syntax
#4699
Conversation
iota-ci
added
infrastructure
samuel-rufi
changed the base branch from
develop
to
sc-platform/transactional-tests-syntax
|
Deployment failed with the following error: |
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 4 Skipped Deployments
|
|
|
||
| 1. **Object Placeholders** | ||
| - **Syntax**: `@{obj_x_y}` or `@{obj_x_y_opt}` | ||
| - Here, `(x, y)` corresponds to the transaction index and the creation index of the object within that transaction. The placeholder will be replaced with the object ID as a string (like `0xABCD...`). |
There was a problem hiding this comment.
What is the transaction index? I know it captures the order of creation within the test, but I don't recall if it is sequence number of the command that included the transaction (e.g. init) or the absolute sequence number of the transaction. Could elaborate/disambiguate?
There was a problem hiding this comment.
Good catch, it's the task index, therefore it does not necessarily align with the absolute sequence number of the transaction. When the IotaTestAdapter gets initialized, we set the next_fake to (0,0).
For each objectID we create in the same task, we use the very same task index:
fn enumerate_fake(&mut self, id: ObjectID) -> FakeID {
if let Some(fake) = self.object_enumeration.get_by_left(&id) {
return *fake;
}
let (task, i) = self.next_fake;
let fake_id = FakeID::Enumerated(task, i);
self.object_enumeration.insert(id, fake_id);
self.next_fake = (task, i + 1);
fake_id
}the next_fake is only ever increased when we switch to the next task:
fn next_task(&mut self) {
self.next_fake = (self.next_fake.0 + 1, 0)
}There was a problem hiding this comment.
Refactored with fe77000
Description of change
Documents the
run-graphqlsyntax of theiota-transactional-test-runnerframework.Links to any relevant issues
Fixes #4201
Type of change