examples.md

Bridge — where it lives outside this repo

Cited, real implementations to study (or point an agent at) when designing or reviewing bridge-shaped code.

Python standard library

• logging.Logger × logging.Handler. The logger is the abstraction callers hold; handlers are the interchangeable implementation hierarchy — one logger.info() call fans out to console, file, or syslog backends, and both sides grow independently. docs.python.org/3/library/logging.html

Major ecosystems

• Matplotlib figures over rendering backends. The Figure/Artist layer is one stable abstraction; Agg, SVG, PDF, and GUI canvases are swappable implementors selected at runtime — the canonical large-scale bridge. matplotlib.org/stable/users/explain/figure/backends.html
• Django ORM over database backends. One QuerySet abstraction compiles through per-database implementor packages (PostgreSQL, MySQL, SQLite…); application code never learns which. docs.djangoproject.com/en/stable/ref/databases/
• SQLAlchemy Engine over Dialect/DBAPI. The same split one level down: Core's execution abstraction bridges to per-driver dialects. docs.sqlalchemy.org/en/20/core/engines.html

What to notice across all of them

The implementor interface is always narrow and stable — Handler.emit, the backend canvas API, the dialect contract — while both sides multiply freely behind it. When reviewing bridge-shaped code, check which axis a new requirement lands on: if most changes touch both sides at once, the axes were drawn in the wrong place.