Module upload 

Rdio API upload queue processor.

Polls the streams table for completed voice recordings with upload_status = 'pending' and uploads them to an SDRTrunk-compatible Rdio Scanner API server using the POST /api/call-upload multipart/form-data protocol.

The protocol wire format — field names, the User-Agent: sdrtrunk header, the "Call imported successfully." success marker — lives in rdio. This module is the orchestrator: it drains the database queue, maps StreamRow fields onto the Rdio field set, calls rdio::upload_stream for each row, and transitions the row through the upload lifecycle.

Lifecycle transitions

For every pending row returned by crate::db::uploads::get_pending the processor:

• on success — calls crate::db::uploads::mark_uploaded.
• on failure with attempts below max_retries — calls crate::db::uploads::increment_attempts to bump the counter and leave the row in pending so the next tick retries.
• on failure with attempts at or above max_retries — calls crate::db::uploads::mark_failed to transition to the terminal failed state.

System id scheme

Rdio Scanner expects every system to be numeric, but D-STAR reflector names are alphabetic with a numeric suffix (REF030, DCS030, …). To disambiguate reflectors that share a numeric suffix across protocol families we prefix the numeric suffix with a protocol tag:

Protocol	Base	Example	Resulting id
dplus (REF)	10_000	REF030	10030
dextra (XLX)	20_000	XLX030	20030
dcs (DCS)	30_000	DCS030	30030
dextra (XRF)	40_000	XRF030	40030

Pi-Star labels XLX and XRF reflectors identically as dextra; we distinguish them by the 3-letter prefix on the reflector callsign.

Talkgroup scheme

Rdio Scanner talkgroups are numeric; D-STAR modules are letters A-Z. We map A → 1, B → 2, …, Z → 26. Unknown module letters fall back to 0.

Modules

rdio 🔒

SDRTrunk-compatible Rdio API multipart upload client.

Constants

BATCH_SIZE 🔒

Maximum number of pending rows to drain per tick.

Functions

compute_system_id 🔒

Returns the Rdio Scanner system id for a given D-STAR reflector.

handle_upload_failure 🔒

Handles an upload failure by deciding whether to retry or mark the row as permanently failed, based on the attempts-so-far counter.

make_audio_name 🔒

Builds the audio part’s filename.

make_talker_alias 🔒

Builds the talkerAlias field from a callsign and optional suffix.

module_to_talkgroup 🔒

Returns the Rdio Scanner talkgroup id for a D-STAR module letter.

numeric_suffix 🔒

Extracts the trailing numeric suffix of a reflector callsign.

process_one 🔒

Uploads a single stream and records the outcome.

process_pending 🔒

Drains one batch of pending uploads.

protocol_label 🔒

Returns the human-readable protocol label used in systemLabel.

reflector_prefix 🔒

Returns the first three ASCII-uppercase letters of a reflector name, or None if the prefix is shorter than three bytes or contains non-ASCII.

run 🔒

Runs the upload queue processor loop.