Exactly-once ingest with idempotent checkpointing

[Miracle656]

Background

On restart mid-batch the indexer can double-insert or skip events. Ingest should be exactly-once via idempotent writes + durable checkpoints.

What to build

Wrap each batch in a transaction that upserts events by unique eventId and advances a persisted cursor atomically, so a crash/restart resumes exactly where it left off.

Key files

• src/indexer.ts
• src/indexer/checkpoint.ts (new)

Acceptance criteria

• Kill mid-batch → no dupes, no gaps on resume
• Cursor advances atomically with the batch
• Test simulates crash between insert and commit

Drips Wave · Complexity: Advanced · 200 points

[cyberdocs120]

@cyberdocs120 has applied to work on this issue as part of the Stellar Wave Program's 5th wave.

I’d like to work on this issue because it contributes to improving the project's functionality, reliability, and overall user experience. I enjoy contributing to open-source projects and collaborating with maintainers to deliver clean, maintainable, and well-tested solutions.

I have experience working across frontend, backend, testing, API integration, debugging, and performance optimization tasks. I'm comfortable analyzing existing codebases, understanding requirements, implementing features or fixes, writing tests, and ensuring changes align with project standards and acceptance criteria.

My approach is to first understand the current implementation, identify the most maintainable solution, implement the required changes, add or update tests where necessary, and verify that everything works correctly before submitting a PR. I can communicate progress clearly and incorporate feedback quickly throughout the review process.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @cyberdocs120 to this issue.

[jerrymusaga]

@jerrymusaga has applied to work on this issue as part of the Stellar Wave Program's 5th wave.

Hello! Please I'll like to work on this issue

ℹ️ Repo Maintainers: To accept this application, review their application or assign @jerrymusaga to this issue.

[Zainab5970]

@Zainab5970 has applied to work on this issue as part of the Stellar Wave Program's 5th wave.

please can i work on the issue

ℹ️ Repo Maintainers: To accept this application, review their application or assign @Zainab5970 to this issue.

[Daveside9]

@Daveside9 has applied to work on this issue as part of the Stellar Wave Program's 5th wave.

can i do this? merge in 2hours.

ℹ️ Repo Maintainers: To accept this application, review their application or assign @Daveside9 to this issue.

[oncleweynom]

@oncleweynom has applied to work on this issue as part of the Stellar Wave Program's 5th wave.

goodmorning would want to work on this issue please

ℹ️ Repo Maintainers: To accept this application, review their application or assign @oncleweynom to this issue.