feature/seed-data-and-dedup
- Create seed CSV with 520 supplement records covering 10 categories, 30 brands, 8 forms, 10 countries with realistic ingredient compositions - Change deduplication strategy from sourceUrl to name+brand combination - Implement incremental loading: update existing records, add new ones, never delete - Add SeedDataLoader (ApplicationRunner) that loads seed data on first startup when supplements table is empty - Add unit tests for deduplication, incremental loading, statistics logging - 24 tests pass, 0 failures
Description
Data collector service for supplement (BAD) catalog
85 KiB
Languages
Java
99.5%
Dockerfile
0.5%