Skip to content

IRI/URL sanitization #60

Description

@thenonameguy

In case the Dataframe contains invalid IRI/URL data, currently this gets serialized without it being detected, nor at mapping/nor at serialization-time.

Leaving library users having to implement functions like:

 def sanitize_iris(df: pl.DataFrame, *cols: str) -> pl.DataFrame:
     """Sanitize IRI columns using native Polars expressions (columnar).
 
     1. Strips leading/trailing whitespace.
     2. Percent-encodes spaces (``%20``).
     3. Nulls out values that don't start with ``http``, contain non-ASCII
        characters, or contain characters that are illegal in an IRI
        (``" < > \\ ^ ` { | }`` and control chars) — broken source URLs that
        RDFox/rdflib reject.
     """

Having a more ergonomic/strict mode of serialization that actively catches these cases (like rdflib does at read-time) would be useful.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions