sifr.* namespace. Every module is designed with Rust-backed safety guarantees — predictable ownership, explicit error types, and no hidden global mutation. Because Sifr compiles to Rust rather than running on the CPython interpreter, the standard library is its own surface; it is not a thin wrapper around CPython’s built-in modules.
For a compact inventory of every module currently shipped in stdlib/sifr, see the Module Index.
How to Import
Usefrom sifr.<module> import <symbol> for every standard library import:
import sifr.math (module-object form) is also unsupported in the current release. Always import the specific symbols you need using from sifr.<module> import <symbol>.math, json, or similar exists in your project, the normal top-level resolution imports it without a diagnostic. The SIFR-IMPORT-0008 error only fires when no real top-level module matches the bare name.
Available Modules
Math, Data, and Serialization
Math, Data, and Serialization
Collections
Collections
Filesystem and Environment
Filesystem and Environment
Text, Encoding, and Internationalization
Text, Encoding, and Internationalization
Concurrency and Parallelism
Concurrency and Parallelism
Concurrency has its own docs section. Start with the Concurrency Overview, then use the Concurrency API for module reference.
Networking
Networking
Process, Signal, and Time
Process, Signal, and Time
What Sifr Does Not Expose
Several CPython-shaped names are intentionally absent because they rely on global mutable state or interpreter-specific behavior that conflicts with Sifr’s safety model:asyncioevent-loop objects and loop-policy mutationthreading.Lock,queue.Queue, andmultiprocessing.*codecs.register,locale.setlocale, andgettext.installsignal.signalandset_wakeup_fdfor arbitrary handler registration
sifr.* namespace. See the individual module pages for details.