adding Marker to parse effectively pdf

This commit is contained in:
Adrien
2026-04-04 21:30:18 +02:00
parent b154e29f2d
commit ea1276dc2e
25 changed files with 2318 additions and 285 deletions
+6 -2
View File
@@ -1,10 +1,13 @@
# ai-teacher Development Guidelines
Auto-generated from all feature plans. Last updated: 2026-04-03
Auto-generated from all feature plans. Last updated: 2026-04-04
## Active Technologies
- Java 25 (backend), TypeScript / Node 20 (frontend) + Spring Boot 4.0.5, Spring AI 2.0.0-M4, OpenAI API (embeddings + chat), PDFBox (via Spring AI PDF reader dependency) (002-image-aware-embedding)
- PostgreSQL (JPA + Flyway), pgvector (Spring AI `VectorStore`), local file system (extracted images — `/uploads/figures/`) (002-image-aware-embedding)
- Java 25 (backend), TypeScript / Node 20 (frontend) + Spring Boot 4.0.5, Spring AI 2.0.0-M4, OpenAI API, PDFBox (rendering only), `com.google.cloud:google-cloud-documentai` (~2.40.x) (002-image-aware-embedding)
- PostgreSQL (JPA + Flyway), pgvector (Spring AI VectorStore), S3 / local filesystem (figure images) (002-image-aware-embedding)
- PostgreSQL (JPA + Flyway), pgvector (Spring AI `VectorStore`), S3-compatible (002-image-aware-embedding)
- Java 21 (backend), TypeScript / Node 20 (frontend) (001-neuro-rag-learning)
@@ -24,9 +27,10 @@ npm test && npm run lint
Java 21 (backend), TypeScript / Node 20 (frontend): Follow standard conventions
## Recent Changes
- 002-image-aware-embedding: Added Java 25 (backend), TypeScript / Node 20 (frontend) + Spring Boot 4.0.5, Spring AI 2.0.0-M4, OpenAI API (embeddings +
- 002-image-aware-embedding: Added Java 25 (backend), TypeScript / Node 20 (frontend) + Spring Boot 4.0.5, Spring AI 2.0.0-M4, OpenAI API, PDFBox (rendering only), `com.google.cloud:google-cloud-documentai` (~2.40.x)
- 002-image-aware-embedding: Added Java 25 (backend), TypeScript / Node 20 (frontend) + Spring Boot 4.0.5, Spring AI 2.0.0-M4, OpenAI API (embeddings + chat), PDFBox (via Spring AI PDF reader dependency)
- 001-neuro-rag-learning: Added Java 21 (backend), TypeScript / Node 20 (frontend)
<!-- MANUAL ADDITIONS START -->
<!-- MANUAL ADDITIONS END -->