[PR #379] [MERGED] Fix pricing calculations and preserve model field during preflight operations processing #685

New issue

Closed

opened 2026-03-13 21:03:43 +03:00 by kerem · 0 comments

kerem commented

2026-03-13 21:03:43 +03:00

Owner

📋 Pull Request Information

Original PR: https://github.com/AJaySi/ALwrity/pull/379
Author: @AJaySi
Created: 3/5/2026
Status: ✅ Merged
Merged: 3/5/2026
Merged by: @AJaySi

Base: main ← Head: codex/review-subscription-system-for-production-readiness

📝 Commits (5)

81052d0 Fix preflight model mapping when skipping invalid providers
45dbf09 Add Stripe webhook idempotency persistence guard
81f49f4 Add explicit usage summary uniqueness and billing indexes
7d530b3 Preserve full provider breakdown in billing UI coercion
01bf568 Fix unlimited video limit display in usage rings

📊 Changes

6 files changed (+246 additions, -136 deletions)

View changed files

📝 backend/api/subscription/routes/preflight.py (+14 -7)
📝 backend/models/subscription_models.py (+21 -1)
📝 backend/services/subscription/stripe_service.py (+88 -19)
📝 frontend/src/components/billing/CompactBillingDashboard/components/UsageLimitRings.tsx (+37 -14)
📝 frontend/src/services/billingService.ts (+85 -90)
📝 frontend/src/types/billing.ts (+1 -5)

📄 Description

Motivation

Prevent incorrect cost estimates by fixing token/audio pricing math which was previously scaled by 1/1000.
Ensure model lookups remain consistent when invalid providers are skipped by preserving the originating model field in the validated operations list.
Avoid index drift when iterating validated operations after filtering by iterating directly over the validated list.

Description

Add model to each entry in operations_to_validate and keep request fields together to ensure stable model lookup.
Change the loop to iterate over operations_to_validate directly and use op.get('model') when fetching pricing.
Adjust pricing calculations to treat cost_per_input_token as a cost-per-token unit and multiply by tokens_requested without dividing by 1000, and add clarifying comments.
Preserve special-case audio model pricing logic and apply rounding of per-operation costs to 4 decimal places, while keeping existing fallback default costs.

Testing

Ran the backend subscription/pricing unit tests via pytest (including preflight pricing checks) and the tests completed successfully.

Codex Task

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/AJaySi/ALwrity/pull/379 **Author:** [@AJaySi](https://github.com/AJaySi) **Created:** 3/5/2026 **Status:** ✅ Merged **Merged:** 3/5/2026 **Merged by:** [@AJaySi](https://github.com/AJaySi) **Base:** `main` ← **Head:** `codex/review-subscription-system-for-production-readiness` --- ### 📝 Commits (5) - [`81052d0`](https://github.com/AJaySi/ALwrity/commit/81052d06b430d7b9ca98ed9e5aab8ae14ec06280) Fix preflight model mapping when skipping invalid providers - [`45dbf09`](https://github.com/AJaySi/ALwrity/commit/45dbf095f6470958db14ca9b9c71b959089e9534) Add Stripe webhook idempotency persistence guard - [`81f49f4`](https://github.com/AJaySi/ALwrity/commit/81f49f4ebd204bef3b6f349b4e853bd218e315d6) Add explicit usage summary uniqueness and billing indexes - [`7d530b3`](https://github.com/AJaySi/ALwrity/commit/7d530b3220711282444b0ddb188a92aac3091197) Preserve full provider breakdown in billing UI coercion - [`01bf568`](https://github.com/AJaySi/ALwrity/commit/01bf56837fbacb37a2f85881a67e585a4daa81ea) Fix unlimited video limit display in usage rings ### 📊 Changes **6 files changed** (+246 additions, -136 deletions) <details> <summary>View changed files</summary> 📝 `backend/api/subscription/routes/preflight.py` (+14 -7) 📝 `backend/models/subscription_models.py` (+21 -1) 📝 `backend/services/subscription/stripe_service.py` (+88 -19) 📝 `frontend/src/components/billing/CompactBillingDashboard/components/UsageLimitRings.tsx` (+37 -14) 📝 `frontend/src/services/billingService.ts` (+85 -90) 📝 `frontend/src/types/billing.ts` (+1 -5) </details> ### 📄 Description ### Motivation - Prevent incorrect cost estimates by fixing token/audio pricing math which was previously scaled by 1/1000. - Ensure model lookups remain consistent when invalid providers are skipped by preserving the originating `model` field in the validated operations list. - Avoid index drift when iterating validated operations after filtering by iterating directly over the validated list. ### Description - Add `model` to each entry in `operations_to_validate` and keep request fields together to ensure stable model lookup. - Change the loop to iterate over `operations_to_validate` directly and use `op.get('model')` when fetching pricing. - Adjust pricing calculations to treat `cost_per_input_token` as a cost-per-token unit and multiply by `tokens_requested` without dividing by 1000, and add clarifying comments. - Preserve special-case audio model pricing logic and apply rounding of per-operation costs to 4 decimal places, while keeping existing fallback default costs. ### Testing - Ran the backend subscription/pricing unit tests via `pytest` (including preflight pricing checks) and the tests completed successfully. ------ [Codex Task](https://chatgpt.com/codex/tasks/task_e_69a844ade6848328ba473afa2f4dd0c1) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

kerem

2026-03-13 21:03:43 +03:00

closed this issue
added the
pull-request
label

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

starred/ALwrity#685

No description provided.

Rows
Columns