Skip to main content

Choosing a package

All three packages wrap the same Rust core, so they share semantics and the cursor wire format. Pick by the language you build in.

You are building in…InstallPackageGuide
Python (3.11+)pip install pypaginatepypaginatePython
Node / TypeScript (18+)npm i @cyblow/paginate@cyblow/paginateTypeScript
Rustcargo add paginate-corepaginate-coreRust core

:::tip Polyglot systems You don't have to choose just one. A common shape is a Python service that mints cursors and a TypeScript client that reads them — because the codec is shared, the cursor round-trips byte-for-byte. See parity. :::

Python — pypaginate​

The native engine ships inside the wheel (built with PyO3 + maturin), so there's no Rust toolchain to install. The core has zero runtime dependencies; Pydantic is needed only for the FastAPI extra.

pip install pypaginate                  # in-memory + native engine
pip install "pypaginate[sqlalchemy]"    # SQLAlchemy 2.0 (sync + async)
pip install "pypaginate[fastapi]"       # FastAPI dependencies (+ Pydantic)
pip install "pypaginate[django]"        # Django Q-object builders
pip install "pypaginate[all]"           # everything

→ Python installation

TypeScript / Node — @cyblow/paginate​

An idiomatic TypeScript adapter over the native addon @cyblow/paginate-core, which installs as a platform-specific dependency (prebuilt binaries via napi-rs).

npm i @cyblow/paginate
# or: bun add @cyblow/paginate

→ TypeScript installation

Rust — paginate-core​

The pure engine itself, with no host/binding dependencies — embed it directly in a Rust service, or read it to understand what the language packages delegate to.

[dependencies]
paginate-core = "0.1"

→ Rust core · docs.rs

What's the same everywhere​

  • The 20 filter operators, the stable null-aware sort, and search ranking.
  • The opaque, URL-safe cursor format.
  • The validation rules and limits (e.g. MAX_LIMIT, filter nesting depth).
  • The error taxonomy — each package maps the core's errors to host-native exceptions with the same hierarchy.

What differs (by design)​

  • Naming style. Python uses snake_case (has_next); TypeScript uses camelCase (hasNext).
  • Spec shapes. Python specs are dataclasses (FilterSpec(field=…, operator=…)); TypeScript specs are plain objects ({ field, operator }).
  • Integrations. Python ships SQLAlchemy / Django / FastAPI; TypeScript ships Express / Prisma / Drizzle.