GeoJSON Text Sequences
Extracted elements (9)
GeoJSON text sequences exist to solve the problem of arbitrarily large geographic datasets (tens of thousands or millions of features) that are impractical to represent as a single GeoJSON FeatureCollection because they cannot be parsed incrementally.
Using the ASCII RS character (0x1e) as a record delimiter means producers and parsers need not enforce a canonical form of GeoJSON; any valid GeoJSON, pretty-printed or compact, can appear in a sequence.
GeoJSON text sequences containing compact GeoJSON texts with no internal newlines are more interoperable with existing non-standardized newline-delimited JSON parsers deployed on the internet.
In distributed systems where message ordering and exactly-once delivery are hard to guarantee, GeoJSON text sequences that do not rely on text ordering for extra semantics are more interoperable than those that do.
Each JSON text in a GeoJSON text sequence MUST contain a single GeoJSON object as defined in RFC 7946.
A GeoJSON text sequence is any number of GeoJSON texts, each encoded in UTF-8, preceded by one ASCII record separator (RS, 0x1e) character, and followed by a line feed (LF). It conforms to all rules of RFC 7464 (JSON text sequences).
Heterogeneous sequences containing a mix of GeoJSON Geometry, Feature, and FeatureCollection objects are permitted within a single GeoJSON text sequence.
IANA registered the MIME media type 'application/geo+json-seq' in the Media Types registry for use with GeoJSON text sequences, using the structured syntax suffix established in RFC 8091.
GeoJSON text sequences have no security considerations beyond those of JSON text sequences (RFC 7464) and the GeoJSON format (RFC 7946).