fix(graph): UNIQUE constraints + idempotent writes for node/edge dedup (#181, #195) by Jose-Gael-Cruz-Lopez · Pull Request #249 · SaplingLearn/Sapling

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)

cloudflare-workers-and-pages · 2026-06-22T04:02:57Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Updated (UTC)
❌ Deployment failed View logs	frontend	`260f952`	Jun 22 2026, 04:03 AM

coderabbitai · 2026-06-22T04:02:57Z

Warning

Review limit reached

@Jose-Gael-Cruz-Lopez, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 40 minutes and 36 seconds. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e10971e-3224-46bd-b2f9-8aab89a7491b

📥 Commits

Reviewing files that changed from the base of the PR and between 32409a8 and 46fb5ac.

📒 Files selected for processing (6)

backend/db/dedup_nodes.py
backend/db/migrations/0001_baseline_schema.sql
backend/db/migrations/0021_graph_dedup_constraints.sql
backend/services/graph_service.py
backend/tests/test_dedup_unique_migration.py
backend/tests/test_graph_dedup_constraints.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/graph-dedup-constraints

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

…whitespace) (#181)

…index build

cloudflare-workers-and-pages · 2026-06-24T04:49:11Z

Deploying with Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	frontend-staging	`46fb5ac`	Commit Preview URL Branch Preview URL	Jun 24 2026, 02:52 PM

…raints

main restructured DB into ordered migrations/ applied by migrate.py and removed the flat migration_dedup_unique.sql path. Relocate the node/edge dedup + UNIQUE index DDL to migrations/0021_graph_dedup_constraints.sql, preserving dedup-before-index ordering and the BEGIN/COMMIT wrapper. Fresh DBs already get the two indexes inline from 0001_baseline_schema.sql.

The old flat migration_dedup_unique.sql and supabase_schema.sql were removed by main's DB restructure. Point the drift guard at the files that exist: migrations/0021_graph_dedup_constraints.sql (back-fill) and migrations/0001_baseline_schema.sql (fresh-DB inline). Preserve original intent: normalized expression, NULLS NOT DISTINCT, dedup-before-index order, plus transaction-wrap and edge on_conflict-column checks.

AndresL230 · 2026-06-25T02:53:11Z

Superseded by the DB modular redesign (#279). graph_nodes/graph_edges UNIQUE-backed dedup (#181/#195) ships in 0023 and dedup_nodes.py is retired. Closing as obsolete. Reopen if not fully covered.

Jose-Gael-Cruz-Lopez · 2026-06-22T04:02:50Z

Graph-node and graph-edge dedup was best-effort in app code (a select-then-insert race in apply_graph_update) plus a manual db/dedup_nodes.py cleanup. Two concurrent updates (two tabs, chat+quiz) both miss the existence check and write duplicates. This makes duplicates impossible at the DB level and turns the writes idempotent.

DB — `backend/db/migration_dedup_unique.sql` (new)

Nodes (No UNIQUE constraint backs graph-node dedup — ships a manual dedup_nodes.py instead #181): collapse existing duplicates per (user_id, lower(concept_name), course_id) — keeping the strongest row and repointing/removing dependent edges, quiz_attempts, quiz_context (same policy as dedup_nodes.py) — then CREATE UNIQUE INDEX … (user_id, lower(concept_name), course_id) NULLS NOT DISTINCT. lower(concept_name) matches the app's _normalize_concept; NULLS NOT DISTINCT (PG15+) makes course-less duplicates collide too.
Edges (No UNIQUE constraint backs edge dedup — concurrent/duplicate edges #195): dedup (user_id, source_node_id, target_node_id) then CREATE UNIQUE INDEX idx_graph_edges_unique.
Both mirrored into supabase_schema.sql.

App — `backend/services/graph_service.py`

Node insert now recovers from the resulting 409 (concurrent insert won the race): it refetches and resolves the winning row so the rest of the batch still links. Non-409 errors still propagate.
Edge write switched from insert to upsert(on_conflict="user_id,source_node_id,target_node_id").
dedup_nodes.py docstring updated — it's now a one-time backfill, not an ongoing safeguard.

Tests

test_graph_dedup_constraints.py — 409-recovery resolves the winner; non-409 propagates; edges go through upsert with the right on_conflict.
test_dedup_unique_migration.py — drift guard on both indexes + dedup-before-index ordering.

Verification

ruff check . clean; gated suite green (+6 new). No live Postgres in CI, so the NULLS NOT DISTINCT build is left for a reviewer with DB access.

Closes#181, #195.

)

…des (#181)

…, course_id) (#181)

…rt (#181)

… race (#195)

…NIQUE index (#181)