correspond.channels.ntfy

ntfy: push notifications to a phone or a desktop through an ntfy server (send only).

References

ntfy:<topic>, or ntfy: for the default topic, which is resolved in this order: $NTFY_TOPIC; the macOS Keychain (service topic_keychain_service in the [ntfy] config table, default correspond-ntfy-topic); an ssh host named by topic_remote, asked for the NTFY_TOPIC=... line of topic_remote_file. A dry run reads only the environment and the config file (no Keychain, no remote host: those are looked up when sending), and a plan shows a topic masked.

A topic is a bearer secret: anyone who knows an unauthenticated one can publish to it and read it. Prefer ntfy: to writing a topic into a reference that ends up in logs.

>>> Ntfy().parse_ref("").kind
'topic'

Module Attributes

PRIORITY

Draft priorities as ntfy's 1-5 scale.

Classes

Ntfy(*[, http, run])

An ntfy server: publish to a topic.

class correspond.channels.ntfy.Ntfy(*, http=<function urllib_http>, run=<function run>)

Bases: object

An ntfy server: publish to a topic.

audience(ref, *, draft=None)

Who reads a topic: anyone who knows its name, unless the config says the server denies anonymous reads.

Nothing is asked of the server. Its cache keeps each message for subscribers who connect later (cache_duration, 12h by default), and every subscriber’s device gets a copy.

Return type:

Audience

property capabilities: Capabilities

Send only, with priorities.

parse_ref(id)

A topic, or nothing for the default topic.

Return type:

ConversationRef

send(ref, draft, *, dry_run=False)

Publish draft.text (with its title and priority) to the topic.

Return type:

SendResult

correspond.channels.ntfy.PRIORITY = {'high': '4', 'low': '2', 'normal': '3', 'urgent': '5'}

Draft priorities as ntfy’s 1-5 scale.