Revision-safe interests; CI wiring; remove pg_worker; create dest dir#323
Merged
Revision-safe interests; CI wiring; remove pg_worker; create dest dir#323
Conversation
…tegy Add a detailed 616-line execution plan for roadmap item 3.5.4 defining and implementing a revision-safe update strategy for user interests. This new document outlines the purpose, constraints, tolerances, risks, progress stages, surprises, decision log, outcomes, plan of work, interfaces, error mapping, and team ownership related to replacing the legacy last-write-wins interests update with an optimistic concurrency model aligned with the existing user preferences aggregate. No implementation is included yet; this draft locks the architectural and concurrency decisions to guide subsequent development and testing. Co-authored-by: devboxerhub[bot] <devboxerhub[bot]@users.noreply.github.com>
Reviewer's GuideAdds an ExecPlan markdown document defining the revision-safe update strategy for PUT /api/v1/users/me/interests, including goals, constraints, risks, progress tracking, and a staged implementation plan, without changing any production code. Sequence diagram for revision-safe PUT /api/v1/users/me/interestssequenceDiagram
actor Client
participant HttpUsersHandler
participant UserInterestsCommand
participant DieselUserInterestsCommand
participant UserPreferencesRepository
participant Postgres
Client->>HttpUsersHandler: PUT /api/v1/users/me/interests
activate HttpUsersHandler
HttpUsersHandler->>HttpUsersHandler: parse_InterestsRequest_with_expectedRevision
HttpUsersHandler->>UserInterestsCommand: update(UpdateUserInterestsRequest)
activate UserInterestsCommand
UserInterestsCommand->>DieselUserInterestsCommand: delegate_update(UpdateUserInterestsRequest)
activate DieselUserInterestsCommand
DieselUserInterestsCommand->>UserPreferencesRepository: find_by_user_id(UserId)
activate UserPreferencesRepository
UserPreferencesRepository->>Postgres: SELECT_from_user_preferences
Postgres-->>UserPreferencesRepository: row_or_none_with_revision
UserPreferencesRepository-->>DieselUserInterestsCommand: UserPreferencesOption
deactivate UserPreferencesRepository
alt no_existing_row_and_expected_revision_none
DieselUserInterestsCommand->>UserPreferencesRepository: save(new_row_revision_1)
UserPreferencesRepository->>Postgres: INSERT_user_preferences_rev_1
Postgres-->>UserPreferencesRepository: insert_ok
UserPreferencesRepository-->>DieselUserInterestsCommand: saved_preferences_rev_1
DieselUserInterestsCommand-->>UserInterestsCommand: Ok(UserInterests_rev_1)
else no_existing_row_and_expected_revision_some
DieselUserInterestsCommand-->>UserInterestsCommand: Err(conflict_revision_mismatch_actual_0)
else existing_row_and_expected_revision_none
DieselUserInterestsCommand-->>UserInterestsCommand: Err(conflict_missing_expected_revision)
else existing_row_and_revision_mismatch
DieselUserInterestsCommand-->>UserInterestsCommand: Err(conflict_revision_mismatch_actual_current)
else existing_row_and_revision_matches
DieselUserInterestsCommand->>UserPreferencesRepository: save(updated_row_revision_plus_1)
UserPreferencesRepository->>Postgres: UPDATE_user_preferences_set_interests_increment_revision
Postgres-->>UserPreferencesRepository: update_ok
UserPreferencesRepository-->>DieselUserInterestsCommand: saved_preferences_new_revision
DieselUserInterestsCommand-->>UserInterestsCommand: Ok(UserInterests_new_revision)
end
deactivate DieselUserInterestsCommand
UserInterestsCommand-->>HttpUsersHandler: UserInterests_or_conflict_error
deactivate UserInterestsCommand
alt success
HttpUsersHandler-->>Client: 200_OK_JSON_with_interestThemeIds_and_revision
else stale_write_conflict
HttpUsersHandler-->>Client: 409_Conflict_JSON_code_conflict_details_revision_mismatch
end
deactivate HttpUsersHandler
Class diagram for revision-safe interests domain and adapter contractsclassDiagram
class UpdateUserInterestsRequest {
+UserId user_id
+Vec~InterestThemeId~ interest_theme_ids
+Option~u32~ expected_revision
}
class UserInterests {
+UserId user_id
+Vec~InterestThemeId~ interest_theme_ids
+u32 revision
}
class Error {
+code: String
+details: String
}
class UserInterestsCommand {
<<interface>>
+update(request UpdateUserInterestsRequest) Result~UserInterests, Error~
}
class UserPreferencesRepository {
<<interface>>
+find_by_user_id(user_id UserId) Result~Option~UserPreferences~, UserPreferencesRepositoryError~
+save(preferences UserPreferences, expected_revision Option~u32~) Result~UserPreferences, UserPreferencesRepositoryError~
}
class UserPreferencesRepositoryError {
+kind: String
+message: String
}
class UserPreferences {
+UserId user_id
+Vec~InterestThemeId~ interest_theme_ids
+Vec~SafetyToggleId~ safety_toggle_ids
+UnitSystem unit_system
+u32 revision
}
class DieselUserInterestsCommand {
+repository: UserPreferencesRepository
+update(request UpdateUserInterestsRequest) Result~UserInterests, Error~
-handle_insert_case(request UpdateUserInterestsRequest) Result~UserInterests, Error~
-handle_update_case(request UpdateUserInterestsRequest, preferences UserPreferences) Result~UserInterests, Error~
-map_repository_error(error UserPreferencesRepositoryError) Error
}
class HttpUsersHandler {
+update_interests(request_body InterestsRequest, user_id UserId) HttpResponse
}
class InterestsRequest {
+Vec~InterestThemeId~ interest_theme_ids
+Option~u32~ expected_revision
}
class HttpResponse {
+status: u16
+body: String
}
UserInterestsCommand <|.. DieselUserInterestsCommand
DieselUserInterestsCommand --> UserPreferencesRepository
UserPreferencesRepository --> UserPreferences
HttpUsersHandler --> UserInterestsCommand
HttpUsersHandler --> InterestsRequest
HttpUsersHandler --> HttpResponse
UpdateUserInterestsRequest --> UserInterests
UserPreferences --> UserInterests
UserPreferencesRepositoryError --> Error
Flow diagram for revision-safe interests persistence decision matrixflowchart TD
A[Start_update_interests] --> B[Load_user_preferences_by_user_id]
B --> C{Preferences_row_exists}
C -->|no| D{expected_revision_is_some}
D -->|no| E[Insert_new_preferences_row_with_interests_revision_1]
E --> F[Return_UserInterests_revision_1]
D -->|yes| G[Return_conflict_error_actualRevision_0_expectedRevision_supplied]
C -->|yes| H{expected_revision_is_some}
H -->|no| I[Return_conflict_error_missing_expectedRevision_for_existing_row]
H -->|yes| J{expected_revision_equals_current_revision}
J -->|no| K[Return_conflict_error_revision_mismatch]
J -->|yes| L[Update_existing_row_set_interests_preserve_safety_toggle_ids_and_unit_system_increment_revision]
L --> M[Return_UserInterests_with_new_revision]
F --> N[End]
G --> N
I --> N
K --> N
M --> N
File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
Contributor
Summary by CodeRabbitRelease Notes
WalkthroughThis PR introduces revision-safe user interests updates with optimistic concurrency control throughout the backend stack. The pg_worker binary is removed from Cargo and refactored to a Makefile-based build approach. Comprehensive BDD test coverage validates revision conflict scenarios against a real database. Changes
sequenceDiagram
actor Client
participant HTTP as HTTP Handler
participant Domain as Domain<br/>(UpdateUserInterestsRequest)
participant Persist as Persistence<br/>(DieselUserInterestsCommand)
participant DB as Database<br/>(user_preferences)
Client->>HTTP: PUT /interests<br/>(expected_revision: 2)
HTTP->>HTTP: Parse expected_revision<br/>into UpdateUserInterestsRequest
HTTP->>Persist: set_interests(request)
Persist->>DB: SELECT user_preferences<br/>WHERE user_id = ?
DB-->>Persist: existing revision: 1
alt Revision Mismatch
Persist->>Persist: Build conflict error<br/>(expected: 2, actual: 1)
Persist-->>HTTP: Err(Conflict {<br/>expectedRevision: 2,<br/>actualRevision: 1 })
HTTP-->>Client: 409 Conflict
else Revision Match
Persist->>Persist: Calculate next<br/>revision: 2 + 1 = 3
Persist->>DB: UPDATE user_preferences<br/>SET interest_theme_ids = ?,<br/>revision = 3<br/>WHERE user_id = ? AND revision = 2
DB-->>Persist: OK
Persist-->>HTTP: UserInterests<br/>{revision: 3}
HTTP-->>Client: 200 OK<br/>{revision: 3}
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~55 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
📝 Coding Plan
Comment |
![codescene-delta-analysis[bot]](https://avatars.githubusercontent.com/in/53921?s=60&v=4){kind=small}
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
…s update - Introduce UpdateUserInterestsRequest to carry expectedRevision for concurrency control. - UserInterests now includes a revision field to track aggregate version. - Convert DieselUserInterestsCommand to enforce revision checks and reject stale writes. - Expose revision-safe update and conflict semantics in HTTP API for PUT /users/me/interests. - Remove implicit retries on concurrency conflicts; surface 409 Conflict with revision details. - Add extensive unit and integration tests covering revision conflicts and concurrency handling. - Update documentation to reflect that interests share aggregate and revision with user preferences, with optimistic locking. This change ensures safe concurrent updates on user interests by leveraging the aggregate revision from user preferences, providing explicit conflict feedback to clients when revisions are stale or missing after the first write. Co-authored-by: devboxerhub[bot] <devboxerhub[bot]@users.noreply.github.com>
leynos
changed the title
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
…errors and failures Add a comprehensive 500-line investigation document describing three transient failures observed in the backend validation process. This report includes failure reproductions, environment snapshots, failure diagnoses, relevant code paths, and recommended next steps. The failures are analyzed in detail, identifying causes such as a broken /dev/null device and transient toolchain instabilities, with evidence ruling out persistent code defects. This addition aids in debugging and future failure mitigation related to the Rust toolchain and container environment. Co-authored-by: devboxerhub[bot] <devboxerhub[bot]@users.noreply.github.com>
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: Excess Number of Function Argumentsbackend/tests/user_interests_revision_conflicts_bdd/db_support.rs: seed_preferences What lead to degradation?seed_preferences has 6 arguments, max arguments = 4 Why does this problem occur?Functions with many arguments indicate either a) low cohesion where the function has too many responsibilities, or b) a missing abstraction that encapsulates those arguments. How to fix it?Start by investigating the responsibilities of the function. Make sure it doesn't do too many things, in which case it should be split into smaller and more cohesive functions. Consider the refactoring INTRODUCE PARAMETER OBJECT to encapsulate arguments that refer to the same logical concept. |
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: Code Duplicationbackend/tests/user_interests_revision_conflicts_bdd/flow_support.rs: What lead to degradation?The module contains 5 functions with similar structure: preferences_snapshot,run_first_write,run_missing_revision_conflict,run_stale_revision_conflict and 1 more functions Why does this problem occur?Duplicated code often leads to code that's harder to change since the same logical change has to be done in multiple functions. More duplication gives lower code health. How to fix it?A certain degree of duplicated code might be acceptable. The problems start when it is the same behavior that is duplicated across the functions in the module, ie. a violation of the Don't Repeat Yourself (DRY) principle. DRY violations lead to code that is changed together in predictable patterns, which is both expensive and risky. DRY violations can be identified using CodeScene's X-Ray analysis to detect clusters of change coupled functions with high code similarity. Read More |
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: Code Duplicationbackend/tests/user_interests_revision_conflicts_bdd.rs: What lead to degradation?The module contains 6 functions with similar structure: existing_preferences_revision_1_with_preserved_safety_and_unit_settings,existing_preferences_revision_2,the_first_interests_response_includes_revision_1,the_response_is_a_conflict_with_expected_revision_1_and_actual_revision_2 and 2 more functions Why does this problem occur?Duplicated code often leads to code that's harder to change since the same logical change has to be done in multiple functions. More duplication gives lower code health. How to fix it?A certain degree of duplicated code might be acceptable. The problems start when it is the same behavior that is duplicated across the functions in the module, ie. a violation of the Don't Repeat Yourself (DRY) principle. DRY violations lead to code that is changed together in predictable patterns, which is both expensive and risky. DRY violations can be identified using CodeScene's X-Ray analysis to detect clusters of change coupled functions with high code similarity. Read More |
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: Code Duplicationbackend/tests/user_interests_revision_conflicts_bdd/flow_support.rs: What lead to degradation?The module contains 5 functions with similar structure: preferences_snapshot,run_first_write,run_missing_revision_conflict,run_stale_revision_conflict and 1 more functions Why does this problem occur?Duplicated code often leads to code that's harder to change since the same logical change has to be done in multiple functions. More duplication gives lower code health. How to fix it?A certain degree of duplicated code might be acceptable. The problems start when it is the same behavior that is duplicated across the functions in the module, ie. a violation of the Don't Repeat Yourself (DRY) principle. DRY violations lead to code that is changed together in predictable patterns, which is both expensive and risky. DRY violations can be identified using CodeScene's X-Ray analysis to detect clusters of change coupled functions with high code similarity. Read More |
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This change removes the local `backend/src/bin/pg_worker.rs` binary and the corresponding entry in `Cargo.toml`. Instead, the Makefile is updated to use the `pg_worker` binary provided by the `pg-embed-setup-unpriv` crate. - Makefile updated to check for an existing `pg_worker` on PATH and install it if missing. - Introduced `PG_WORKER_INSTALL_ROOT` and `PG_EMBED_SETUP_UNPRIV_VERSION` variables in Makefile. - Documentation updated to describe the new usage pattern for the embedded Postgres worker helper binary. This simplifies maintenance by relying on an external, versioned binary rather than one built and maintained in this repository. Co-authored-by: devboxerhub[bot] <devboxerhub[bot]@users.noreply.github.com>
leynos
changed the title
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
…flow and DB support for user interests revision - Extracted common test logic into helper functions (e.g., with_world, seed_prefs) to reduce code duplication and improve readability. - Modified seed_preferences DB support function to accept user_id as explicit parameter enhancing clarity. - Simplified test step implementations by wrapping them with helper functions managing skip checks. - Refined flow_support functions to unify update operations and introduce reusable session handling (with_app_session), improving async control and reducing boilerplate. - Cleaned up conditional checks scattered in tests and standardized update call patterns. These changes improve maintainability and clarity of BDD test code for user preferences revision conflict scenarios. Co-authored-by: devboxerhub[bot] <devboxerhub[bot]@users.noreply.github.com>
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: backend/tests/user_interests_revision_conflicts_bdd.rs Comment on file fn the_first_interests_response_includes_revision_1(world: &mut World) {
with_world(world, |world| {
assert_interests_snapshot(
world.first_update.as_ref().expect("first update response"),
&[FIRST_THEME_ID],
1,
);
});
}❌ New issue: Code Duplication |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: backend/tests/user_interests_revision_conflicts_bdd.rs Comment on file fn seed_prefs(
world: &mut World,
interests: &[&str],
safety: &[&str],
unit_system: &str,
revision: i32,
) {
let db = world.db.as_ref().expect("db context");
let user_id = Uuid::parse_str(FIXTURE_AUTH_ID).expect("valid fixture UUID");
seed_preferences(
db.database_url.as_str(),
user_id,
SeedPreferences {
interest_ids: interests,
safety_ids: safety,
unit_system,
revision,
},
)
.expect("seed user preferences");
}❌ New issue: Excess Number of Function Arguments |
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: backend/tests/user_interests_revision_conflicts_bdd/flow_support.rs Comment on file async fn update_interests_snapshot<S>(
app: &S,
cookie: Cookie<'static>,
payload: InterestsRequest,
) -> Snapshot
where
S: actix_web::dev::Service<
actix_http::Request,
Response = actix_web::dev::ServiceResponse<actix_web::body::BoxBody>,
Error = actix_web::Error,
>,
{
let req = actix_test::TestRequest::put()
.uri("/api/v1/users/me/interests")
.cookie(cookie)
.set_json(payload)
.to_request();
capture_snapshot(actix_test::call_service(app, req).await, false).await
}❌ New issue: Code Duplication |
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 7d9d7706b4
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Contributor
There was a problem hiding this comment.
Actionable comments posted: 7
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
backend/src/inbound/http/schemas.rs (1)
200-210: 🧹 Nitpick | 🔵 TrivialExtend schema test to verify the new
revisionfield.The test asserts
interestThemeIdsis present but does not check for the newly addedrevisionfield.♻️ Proposed fix
fn user_interests_schema_has_expected_name() { let schema_json = schema_to_json::<UserInterestsSchema>(); let name = UserInterestsSchema::name(); // utoipa replaces :: with . in schema names assert_eq!(name, "crate.domain.UserInterests"); assert!( schema_json.contains("interestThemeIds"), "schema should contain interestThemeIds field" ); + assert!( + schema_json.contains("revision"), + "schema should contain revision field" + ); }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/src/inbound/http/schemas.rs` around lines 200 - 210, The test user_interests_schema_has_expected_name currently checks schema_json for "interestThemeIds" but misses the newly added "revision" field; update this test (in the function user_interests_schema_has_expected_name) to also assert that schema_to_json::<UserInterestsSchema>() contains "revision" (or the exact JSON key used for the revision field) so the test verifies both interestThemeIds and revision are present in UserInterestsSchema's JSON schema.backend/tests/diesel_profile_interests_adapters.rs (1)
167-184:⚠️ Potential issue | 🟡 MinorAdd
skip_serializing_ifto properly test field omission.The
expected_revisionfield must be marked with#[serde(skip_serializing_if = "Option::is_none")]in theInterestsRequeststruct. Without this, tests serialiseNoneas"expectedRevision": nullinstead of omitting the field entirely, so the omission path remains unexercised.Alternatively, build test payloads as raw JSON strings when asserting the omission behaviour.
Also applies to test cases at lines 302-305, 349-352.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/tests/diesel_profile_interests_adapters.rs` around lines 167 - 184, The test serialises InterestsRequest with expected_revision = None but currently serialises as "expectedRevision": null; update the InterestsRequest definition to annotate the expected_revision field with #[serde(skip_serializing_if = "Option::is_none")] so that None values are omitted (or, if you prefer not to change the struct, adjust the tests that call do_update_interests to send raw JSON payloads that omit expectedRevision for the omission path); target the InterestsRequest struct and its expected_revision field referenced by the tests (also update the related test cases exercising omission).
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@backend/src/domain/user_interests.rs`:
- Around line 20-21: Add short Rustdoc examples for the public API: include
example code for UserInterests::new that constructs a UserInterests (provide
sample UserId and InterestThemeId values) and asserts the revision returned by
UserInterests::revision(); also add a separate doc example on the revision()
method showing it returns the expected u32. Place these examples in the doc
comments for the new function and the revision() method so they demonstrate
construction and assert the revision value.
In `@backend/src/inbound/http/users/tests.rs`:
- Around line 358-370: The test is asserting HTTP camelCase on the domain DTO
(UpdateUserInterestsRequest) instead of the HTTP DTO; change the test to
serialize the inbound HTTP type InterestsRequest (the DTO consumed by
update_interests) and assert that its JSON uses "expectedRevision" with value 3,
removing the camelCase assertion from any domain types (e.g.,
UpdateUserInterestsRequest / crate::domain::ports) so the wire contract is
tested at the HTTP boundary handled by update_interests.
In `@docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md`:
- Around line 223-230: The documentation bullet describing revision-mismatch
responses is inconsistent with the implemented contract; update the sentence
under the PUT /api/v1/users/me/interests / DieselUserInterestsCommand notes to
state the conflict envelope shape used by the implementation: top-level code:
"conflict" with nested revision details (e.g. expectedRevision and
actualRevision) rather than top-level code: "revision_mismatch"; make this
wording match the validation section already in the doc so the ExecPlan and the
implementation are consistent.
In `@docs/investigations/2026-03-13-toolchain-errors.md`:
- Around line 286-287: Replace first-person narration in the specified prose
(e.g., phrases like "I re-ran…", "I also re-ran…", "I therefore assign…") with
neutral technical voice; update the sentences in the ranges referenced (lines
around 286-287, 311-313, 408-409, 466-467) to passive or third‑person
formulations such as "The probe was re-run…" or "The result is therefore
assigned…" so the document follows the guideline avoiding 1st/2nd person
pronouns.
- Line 1: The document uses title case for multiple headings (e.g., "# Toolchain
Error Investigation: 2026-03-13", "## Environment Snapshot", "## Relevant Code
Paths"); update each heading to sentence case across the file (including the
listed occurrences at lines referenced in the review) so they read like normal
sentences (for example, change "Toolchain Error Investigation: 2026-03-13" to
"Toolchain error investigation: 2026-03-13" and similarly convert "Environment
Snapshot" and "Relevant Code Paths" to "Environment snapshot" and "Relevant code
paths"); scan the whole file and replace all other headings mentioned in the
comment with their sentence-case equivalents to comply with the guideline.
In `@docs/wildside-backend-architecture.md`:
- Around line 475-482: Update the paragraph describing the 409 response for PUT
/api/v1/users/me/interests to reflect the actual envelope shape used by the
code/tests: top-level code must be "conflict" and the nested details must
include code: "revision_mismatch" plus the expectedRevision and actualRevision
fields; adjust the sentence that currently says code: "revision_mismatch" at top
level to instead describe details.code and include that the response contains
those revision fields to match the persistence regression tests and client
contract.
In `@docs/wildside-pwa-data-model.md`:
- Around line 384-401: Update the document metadata header in
docs/wildside-pwa-data-model.md by changing the "Last updated: 14 December 2025"
timestamp to the current date for this change; locate the top-of-file header
line that starts with "Last updated:" and replace the old date so the document
update that modifies the PUT /api/v1/users/me/interests contract is reflected in
the file metadata.
---
Outside diff comments:
In `@backend/src/inbound/http/schemas.rs`:
- Around line 200-210: The test user_interests_schema_has_expected_name
currently checks schema_json for "interestThemeIds" but misses the newly added
"revision" field; update this test (in the function
user_interests_schema_has_expected_name) to also assert that
schema_to_json::<UserInterestsSchema>() contains "revision" (or the exact JSON
key used for the revision field) so the test verifies both interestThemeIds and
revision are present in UserInterestsSchema's JSON schema.
In `@backend/tests/diesel_profile_interests_adapters.rs`:
- Around line 167-184: The test serialises InterestsRequest with
expected_revision = None but currently serialises as "expectedRevision": null;
update the InterestsRequest definition to annotate the expected_revision field
with #[serde(skip_serializing_if = "Option::is_none")] so that None values are
omitted (or, if you prefer not to change the struct, adjust the tests that call
do_update_interests to send raw JSON payloads that omit expectedRevision for the
omission path); target the InterestsRequest struct and its expected_revision
field referenced by the tests (also update the related test cases exercising
omission).
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
Run ID: a464fa6b-1587-418b-9ad1-8fe5a9376f3e
⛔ Files ignored due to path filters (1)
pnpm-lock.yamlis excluded by!**/pnpm-lock.yaml
📒 Files selected for processing (38)
.github/workflows/ci.ymlMakefilebackend/Cargo.tomlbackend/src/bin/pg_worker.rsbackend/src/domain/ports/mod.rsbackend/src/domain/ports/user_interests_command.rsbackend/src/domain/user_interests.rsbackend/src/inbound/http/schemas.rsbackend/src/inbound/http/users.rsbackend/src/inbound/http/users/tests.rsbackend/src/outbound/persistence/diesel_user_interests_command.rsbackend/src/outbound/persistence/diesel_user_interests_command/tests.rsbackend/src/outbound/persistence/diesel_user_interests_command/tests/fixture_guards.rsbackend/src/outbound/persistence/diesel_user_interests_command/tests/mapping.rsbackend/src/outbound/persistence/diesel_user_interests_command/tests/retry.rsbackend/src/outbound/persistence/diesel_user_interests_command/tests/support.rsbackend/tests/adapter_guardrails/doubles_users.rsbackend/tests/adapter_guardrails/harness_defaults.rsbackend/tests/admin_enrichment_reporting_bdd.rsbackend/tests/catalogue_endpoints_bdd.rsbackend/tests/diesel_profile_interests_adapters.rsbackend/tests/features/user_interests_revision_conflicts.featurebackend/tests/offline_walk_endpoints_bdd.rsbackend/tests/pwa_annotations_bdd.rsbackend/tests/pwa_preferences_bdd.rsbackend/tests/user_interests_revision_conflicts_bdd.rsbackend/tests/user_interests_revision_conflicts_bdd/db_support.rsbackend/tests/user_interests_revision_conflicts_bdd/flow_support.rsbackend/tests/user_session_bdd.rsbackend/tests/user_state_profile_interests_startup_modes_bdd.rsbackend/tests/user_state_profile_interests_startup_modes_bdd/flow_support.rsdocs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.mddocs/investigations/2026-03-13-toolchain-errors.mddocs/wildside-backend-architecture.mddocs/wildside-pwa-data-model.mddocs/wildside-testing-guide.mdfrontend-pwa/scripts/run-audit.test.mjspackage.json
💤 Files with no reviewable changes (9)
- backend/tests/admin_enrichment_reporting_bdd.rs
- backend/src/outbound/persistence/diesel_user_interests_command/tests/fixture_guards.rs
- backend/tests/pwa_annotations_bdd.rs
- backend/tests/catalogue_endpoints_bdd.rs
- backend/tests/offline_walk_endpoints_bdd.rs
- backend/tests/pwa_preferences_bdd.rs
- backend/Cargo.toml
- backend/src/outbound/persistence/diesel_user_interests_command/tests.rs
- backend/src/bin/pg_worker.rs
When pg_worker is found on PATH, the prepare-pg-worker target copies it to target/pg_worker without first creating the parent directory. On a clean checkout or after 'make clean', this fails with 'No such file or directory', blocking 'make test' in environments with a global pg_worker. This commit adds 'mkdir -p $(dirname $(PG_WORKER_PATH))' before the install command in both branches of the conditional to ensure the target directory always exists. Also removes invalid 'noImportCycles' Biome rule that was causing check-fmt to fail, and applies auto-formatting to test files. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
leynos
changed the title
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@Makefile`:
- Around line 168-181: The prepare-pg-worker target duplicates mkdir -p
"$$(dirname "$(PG_WORKER_PATH)")" in both branches; move that directory creation
above the if to run unconditionally. Update the prepare-pg-worker recipe so it
first runs mkdir -p "$$(dirname "$(PG_WORKER_PATH)")" once, then performs the if
command -v pg_worker ... else ... install/cargo steps and the final install of
"$(PG_WORKER_INSTALL_ROOT)/bin/pg_worker" to "$(PG_WORKER_PATH)"; ensure you
keep the same shell escaping and variables (PG_WORKER_PATH,
PG_WORKER_INSTALL_ROOT, PG_EMBED_SETUP_UNPRIV_VERSION) intact.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
Run ID: e27851e7-0c66-4693-ac37-30b410f8fd17
📒 Files selected for processing (2)
Makefilebiome.jsonc
💤 Files with no reviewable changes (1)
- biome.jsonc
…date This commit addresses inline and outside-diff code review findings: Code changes: - Add Rustdoc examples to UserInterests::new and revision() methods demonstrating construction and revision assertion - Fix test in users/tests.rs to assert on HTTP DTO (InterestsRequest) instead of domain DTO for proper boundary testing - Add revision field assertion to user_interests_schema test - Add skip_serializing_if to InterestsRequest.expected_revision to omit None values from JSON serialization - Deduplicate mkdir command in prepare-pg-worker Makefile target by moving it before the conditional Documentation changes: - Update execplan, backend architecture, and PWA data model docs to correctly describe conflict response envelope shape: top-level code: "conflict" with nested details containing code: "revision_mismatch", expectedRevision, and actualRevision - Fix first-person narration in toolchain errors investigation doc, replacing "I re-ran" with passive constructions - Convert all headings to sentence case in toolchain errors doc per style guidelines - Update PWA data model doc timestamp to current date Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
leynos
changed the title
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (3)
docs/investigations/2026-03-13-toolchain-errors.md (1)
348-352:⚠️ Potential issue | 🟡 MinorRemove first-person phrasing in the diagnosis section.
Rewrite the heading at Line 348 to neutral technical voice to match repository documentation style.
♻️ Proposed fix
-What I cannot prove from current evidence: +What cannot be proven from current evidence:As per coding guidelines, "Avoid 2nd person or 1st person pronouns ("I", "you", "we")."
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/investigations/2026-03-13-toolchain-errors.md` around lines 348 - 352, The diagnosis heading "What I cannot prove from current evidence:" uses first-person phrasing; change it to a neutral technical voice (e.g., "Unproven causes based on current evidence" or "Potential root causes not proven by current evidence") and scan the surrounding diagnosis section for any other first- or second-person pronouns (e.g., "I", "we", "you") and replace them with neutral phrasing so the entire diagnosis reads in objective, third-person technical style.backend/src/inbound/http/users/tests.rs (1)
359-359: 🧹 Nitpick | 🔵 TrivialRename the test to match the DTO under test.
Rename the function so it reflects
InterestsRequest, notUpdateUserInterestsRequest.♻️ Proposed fix
-fn update_user_interests_request_serializes_expected_revision_in_camel_case() { +fn interests_request_serializes_expected_revision_in_camel_case() {🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@backend/src/inbound/http/users/tests.rs` at line 359, The test function name update_user_interests_request_serializes_expected_revision_in_camel_case is misleading; rename it to reflect the DTO under test (InterestsRequest) — e.g., interests_request_serializes_expected_revision_in_camel_case — by updating the function declaration and any references to that test so its name matches the tested type (InterestsRequest) and keeps the same test body and attributes.docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md (1)
205-207:⚠️ Potential issue | 🟡 MinorCorrect the conflict-envelope wording in the decision log.
Align this entry with the implemented envelope: top-level
code: "conflict"and nested detailscode: "revision_mismatch".🛠️ Proposed fix
-- Decision: align stale-write error mapping with the existing preferences and - annotations conflict envelope using `code: revision_mismatch`, - `expectedRevision`, and `actualRevision`. +- Decision: align stale-write error mapping with the existing preferences and + annotations conflict envelope using top-level `code: "conflict"` and nested + details containing `code: "revision_mismatch"`, `expectedRevision`, and + `actualRevision`.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md` around lines 205 - 207, The decision log currently states the stale-write error mapping uses top-level `code: revision_mismatch`; update the wording to match the implemented envelope: top-level `code: "conflict"` with nested details containing `code: "revision_mismatch"` and the `expectedRevision` / `actualRevision` fields; edit the sentence in the decision entry so it explicitly describes the top-level `conflict` code and nested `revision_mismatch` details (and keep the mention of `expectedRevision` and `actualRevision` in the nested details).
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md`:
- Around line 420-493: The ordered execution steps use repeated "1." markers;
replace each repeated "1." with ascending numbers (1., 2., 3., ...) so the steps
are sequential and deterministic, keeping the existing code blocks and examples
intact and ensuring any nested lists remain properly indented; search for the
repeated "1." list items (the top-level execution steps that contain the cargo
test and make commands and the "Expected ..." blocks) and renumber them in
order, and if your repository guideline requires first-level bullets be hyphens,
convert the very top-level list marker to "-" and then use sequential numeric
markers for the ordered sub-steps.
In `@docs/investigations/2026-03-13-toolchain-errors.md`:
- Around line 311-313: The markdown paragraph containing the sentence starting
with "The command `cargo clippy --workspace --all-targets --all-features -- -D
warnings`..." exceeds 80 columns and triggers MD013; reflow that sentence into
multiple lines so each line is ≤80 characters (for example split after the
clause "...-D warnings` was also re-run." and continue the rest on following
wrapped lines) ensuring the meaning stays unchanged and no backticks or inline
code are broken.
---
Duplicate comments:
In `@backend/src/inbound/http/users/tests.rs`:
- Line 359: The test function name
update_user_interests_request_serializes_expected_revision_in_camel_case is
misleading; rename it to reflect the DTO under test (InterestsRequest) — e.g.,
interests_request_serializes_expected_revision_in_camel_case — by updating the
function declaration and any references to that test so its name matches the
tested type (InterestsRequest) and keeps the same test body and attributes.
In `@docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md`:
- Around line 205-207: The decision log currently states the stale-write error
mapping uses top-level `code: revision_mismatch`; update the wording to match
the implemented envelope: top-level `code: "conflict"` with nested details
containing `code: "revision_mismatch"` and the `expectedRevision` /
`actualRevision` fields; edit the sentence in the decision entry so it
explicitly describes the top-level `conflict` code and nested
`revision_mismatch` details (and keep the mention of `expectedRevision` and
`actualRevision` in the nested details).
In `@docs/investigations/2026-03-13-toolchain-errors.md`:
- Around line 348-352: The diagnosis heading "What I cannot prove from current
evidence:" uses first-person phrasing; change it to a neutral technical voice
(e.g., "Unproven causes based on current evidence" or "Potential root causes not
proven by current evidence") and scan the surrounding diagnosis section for any
other first- or second-person pronouns (e.g., "I", "we", "you") and replace them
with neutral phrasing so the entire diagnosis reads in objective, third-person
technical style.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
Run ID: 4f67615a-a131-476c-b017-fd3264d3e8f1
📒 Files selected for processing (9)
Makefilebackend/src/domain/user_interests.rsbackend/src/inbound/http/schemas.rsbackend/src/inbound/http/users.rsbackend/src/inbound/http/users/tests.rsdocs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.mddocs/investigations/2026-03-13-toolchain-errors.mddocs/wildside-backend-architecture.mddocs/wildside-pwa-data-model.md
Owner
Author
|
@coderabbitai Please suggest a fix for this issue and supply a prompt for an AI coding agent to enable it to apply the fix: backend/src/inbound/http/schemas.rs Comment on file assert!(
schema_json.contains("revision"),
"schema should contain revision field"
);❌ New issue: Code Duplication |
Fix MD013/line-length violation at line 312 by wrapping the sentence to stay within the 80-character limit. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
This comment was marked as resolved.
This comment was marked as resolved.
Owner
Author
|
@coderabbitai have the following now been resolved? docs/investigations/2026-03-13-toolchain-errors.md (1)
As per coding guidelines, "Avoid 2nd person or 1st person pronouns ("I", "you", "we")." backend/src/inbound/http/users/tests.rs (1)
docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md (1)
|
Contributor
|
Tip For best results, initiate chat on the files or code changes. 🧩 Analysis chain🏁 Script executed: Length of output: 2687
1.
|
Replace repeated "1." markers with sequential numbering (1., 2., 3., 4., 5., 6.) in the concrete execution steps section. Indent all list item content (code blocks and text) to maintain proper list continuity and satisfy MD029/ol-prefix markdownlint rule with "ordered" style. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Introduce assert_schema_contains helper to replace repeated pattern across all five has_expected_name test functions. The helper takes the expected schema name and a slice of field names to check, eliminating ~40 lines of duplicated test code. Also remove unused UpdateUserInterestsRequest import from users/tests.rs that was left over from the previous refactoring. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
leynos
changed the title
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
Contributor
There was a problem hiding this comment.
Actionable comments posted: 3
♻️ Duplicate comments (1)
docs/investigations/2026-03-13-toolchain-errors.md (1)
348-348:⚠️ Potential issue | 🟡 MinorEliminate first-person phrasing in the heading text.
Replace
What I cannot prove from current evidence:with neutral wording, e.g.
What cannot be proved from current evidence:.As per coding guidelines: "Avoid 2nd person or 1st person pronouns ('I', 'you', 'we')".
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/investigations/2026-03-13-toolchain-errors.md` at line 348, Change the heading text to remove first-person phrasing: replace the line "What I cannot prove from current evidence:" with a neutral form such as "What cannot be proved from current evidence:" so it conforms to the guideline to avoid first‑person/second‑person pronouns.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@backend/src/inbound/http/users/tests.rs`:
- Around line 358-369: Rename the test function to reference the DTO (e.g.,
update_user_interests_request_serializes_expected_revision_in_camel_case ->
interests_request_serializes_expected_revision_in_camel_case) and harden it by
also asserting that the snake_case key "expected_revision" is not present in the
serialized JSON; locate the InterestsRequest creation and serde_json::to_value
call in the test and add an assertion that value.get("expected_revision") is
None while keeping the existing assertion that
value.get("expectedRevision").and_then(Value::as_u64) == Some(3).
In `@docs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.md`:
- Line 228: The phrase "an existing preferences row" has incorrect article/noun
agreement; locate that exact phrase in the document (search for "an existing
preferences row") and replace it with either "an existing preference row" for
singular or "existing preferences rows" for plural depending on the intended
meaning; ensure surrounding sentence grammar remains correct after the
substitution.
In `@docs/investigations/2026-03-13-toolchain-errors.md`:
- Around line 497-500: The long bullet lines under the "prepare-pg-worker `os
error 2`" section exceed 80 columns; split each long command bullet into a short
descriptive bullet (label) and a following fenced code block so lines wrap under
80 columns; specifically update the bullet containing the strace command
(`strace -f -o /tmp/pg-worker-build.strace cargo build -p backend --bin
pg_worker`) into a label like "Capture strace output:" followed by a fenced code
block with the command, and ensure the other bullets referencing `ls -l`, `ldd`,
`mount`, and `df` are similarly wrapped and line-wrapped to respect the
80-column Markdown rule.
---
Duplicate comments:
In `@docs/investigations/2026-03-13-toolchain-errors.md`:
- Line 348: Change the heading text to remove first-person phrasing: replace the
line "What I cannot prove from current evidence:" with a neutral form such as
"What cannot be proved from current evidence:" so it conforms to the guideline
to avoid first‑person/second‑person pronouns.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
Run ID: be5ba56c-940d-4a70-808d-9a28e10d3dcd
📒 Files selected for processing (4)
backend/src/inbound/http/schemas.rsbackend/src/inbound/http/users/tests.rsdocs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.mddocs/investigations/2026-03-13-toolchain-errors.md
Fix three remaining issues from code review: 1. Replace first-person phrasing "What I cannot prove" with neutral voice "What the available evidence does not establish" in toolchain errors investigation doc 2. Rename test function from update_user_interests_request_serializes_expected_revision_in_camel_case to interests_request_serializes_expected_revision_in_camel_case to match the actual DTO being tested (InterestsRequest) 3. Update execplan conflict-envelope description to accurately reflect the two-level structure: top-level code: "conflict" with nested details containing code: "revision_mismatch", expectedRevision, and actualRevision Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Fix three remaining issues from code review: 1. Harden interests_request serialization test by adding assertion that snake_case key "expected_revision" is not present in JSON, ensuring only camelCase "expectedRevision" is serialized 2. Fix article/noun agreement in execplan: change "an existing preferences row" to "an existing preference row" (singular) 3. Update toolchain investigation doc to use neutral voice: change "What the available evidence does not establish:" to "What cannot be proved from current evidence:" to avoid first-person phrasing Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (1)
docs/investigations/2026-03-13-toolchain-errors.md (1)
497-500:⚠️ Potential issue | 🟡 MinorReflow the long bullet lines to satisfy the 80-column rule.
Break the long command bullets at Line 498 and Line 499 into short labels plus
fenced blocks so every bullet line stays within 80 columns.Proposed fix
- If the `prepare-pg-worker` `os error 2` recurs, capture: - - `strace -f -o /tmp/pg-worker-build.strace cargo build -p backend --bin pg_worker` - - `ls -l` and `ldd` for the exact `rustc` path named in the failure + - strace output: + + ```text + strace -f -o /tmp/pg-worker-build.strace cargo build -p backend --bin pg_worker + ``` + + - `ls -l` and `ldd` output for the exact `rustc` path named in the + failure - `mount`, `df`, and any container lifecycle logsAs per coding guidelines: "Markdown paragraphs and bullet points must be wrapped
at 80 columns."🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/investigations/2026-03-13-toolchain-errors.md` around lines 497 - 500, Reflow the long bullet lines under the "prepare-pg-worker `os error 2`" note so each markdown paragraph and list item is wrapped to 80 columns: break the long `strace -f -o /tmp/pg-worker-build.strace cargo build -p backend --bin pg_worker` line into a short bullet label (e.g., "Capture a strace of the build:") followed by a fenced code block containing the full strace command, do the same for the `ls -l` and `ldd` instruction (e.g., "Collect `ls -l` and `ldd` output for the `rustc` path:" with a fenced block for commands), and ensure the final bullet for `mount`, `df`, and container lifecycle logs is wrapped to 80 columns by splitting or rephrasing into multiple wrapped lines; keep the content identical but formatted to meet the 80-column rule.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In `@docs/investigations/2026-03-13-toolchain-errors.md`:
- Around line 497-500: Reflow the long bullet lines under the "prepare-pg-worker
`os error 2`" note so each markdown paragraph and list item is wrapped to 80
columns: break the long `strace -f -o /tmp/pg-worker-build.strace cargo build -p
backend --bin pg_worker` line into a short bullet label (e.g., "Capture a strace
of the build:") followed by a fenced code block containing the full strace
command, do the same for the `ls -l` and `ldd` instruction (e.g., "Collect `ls
-l` and `ldd` output for the `rustc` path:" with a fenced block for commands),
and ensure the final bullet for `mount`, `df`, and container lifecycle logs is
wrapped to 80 columns by splitting or rephrasing into multiple wrapped lines;
keep the content identical but formatted to meet the 80-column rule.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
Run ID: be07cad3-8930-48ba-aaac-3dec07ff1097
📒 Files selected for processing (3)
backend/src/inbound/http/users/tests.rsdocs/execplans/backend-3-5-4-revision-safe-interests-update-strategy.mddocs/investigations/2026-03-13-toolchain-errors.md
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Summary
Implement revision-safe interests update strategy (3.5.4) and related backend wiring, aligning with existing revision-concurrency patterns across the API. This update also includes CI/test infra tweaks to reliably provision and copy the pg_worker binary during tests.
Create destination directory before copying pg_worker to ensure reliable placement of the binary in test/CI environments.
Remove legacy in-tree pg_worker binary (backend/src/bin/pg_worker.rs) as tests now rely on the embedded/installed worker provided by pg-embed-setup-unpriv.
Update CI workflow (.github/workflows/ci.yml) to invoke make prepare-pg-worker in place of direct cargo build for the worker, ensuring the new directory-prep step runs.
Updated: The Makefile to support a robust pg_worker setup:
Changes to tests and adapters to reflect the typed UpdateUserInterestsRequest and revision-based flow.
Documentation and execplan artifacts for 3.5.4 revision-safe interests, including updated architecture notes and a new execplan doc.
Updated: Several test doubles and harnesses to support the new typed request and revision-encoded flow (UpdateUserInterestsRequest in tests and mocks).
OpenAPI/Docs & Other Artifacts
Rationale
Validation & Acceptance Criteria
OpenAPI/Docs & Other Artifacts
Prior Art / References
Next Steps
📎 Task: https://www.devboxer.com/task/d81ac072-93a0-4fdd-ad2a-7a435c5516a7
📎 Task: https://www.devboxer.com/task/92fae8db-ed86-44a8-b7ea-f51aa3be4824