v10.0:

Breaking changes:

• bluesky:

  • Drop AT_URI_PATTERN in favor of lexrpc.base.AT_URI_RE.
  • to_as1: for app.bsky.actor.profile records, use uri (if provided) as id instead of repo_did.

• nostr:

  • Switch default AS1 actor and object ids from bech32-encoded to hex. This avoids confusion when the same id is bech32-encoded with a different prefix, eg nevent vs note. The previous behavior, bech32-encoded ids, may still be generated by passing id_format='bech32' to nostr.to_as1.
  • from_as1: replace from_protocol kwarg with new proxy_tag kwarg.

• Source.postprocess_object:

  • Drop mentions kwarg in favor of new as1.expand_tags and as1.add_tags_for_html_content_links functions.

• pixelfed:

  • Switch AS1 actor and object ids from tag: URIs to the real ActivityPub ids.

Non-breaking changes:

• Add multiple kwarg to all to_as functions, defaulting to False. When True, returns a list of output objects instead of only a single object. Currently only has effect in bluesky.to_as1; details below.
• as1:
  • targets: exclude hashtags in tags.
  • Add new is_content_html, convert_html_content_to_text, expand_tags, and add_tags_for_html_content_links functions.
• as2:
  • to_as1: extract summary from preview when it's a Note.
• bluesky:
  • Support Web Monetization. Convert the AS2 monetization field in actors to/from community.lexicon.payments.webMonetization records. to_as1 only returns a single-element dict with key monetization, not a real AS1 object.
  • Support new website property in app.bsky.actor.profile, app.bsky.actor.defs#profileViewDetailed, etc.
  • Convert AS1 block of app.bsky.graph.list at:// URI to/from app.bsky.graph.listblock record.
  • from_as1:
    • When multiple=True, returns a list of output records.
    • When converting an actor to app.bsky.actor.profile with a monetization property, and multiple=True, include a community.lexicon.payments.webMonetization record in the list.
    • When converting an article and multiple=True or out_type='site.standard.document', include a site.standard.document record.
    • Add support for converting actors to site.standard.publication with out_type='site.standard.publication'.
    • Add dynamic_sensitive_labels kwarg for choosing label other than graphic-media based on keywords in summary.
    • For articles, convert HTML content, summary, and preview.content more consistently to plain text and facets.
    • For notes and articles, when a hashtag in content is dropped due to truncation, populate it into the standalone tags field.
  • to_as1:
    • Add new client kwarg, paralleling the existing one in from_as1.
• mastodon:
  • status_to_as1_object: bug fix for when mention.url is explicitly null.
• nostr:
  • Add support for d (id) tags with kind 30023 (article) events.
  • Add support for NIP-27 user mentions to both to_as1 and from_as1.
  • from_as1:
    • Bug fix for when inReplyTo is an object with an author field.
    • Bug fix for reposts (shares) with compacted object.
    • Support objectType: comment.
    • Always set created_at to the current time, which some relays require, instead of the input object's published.
    • Always include alt element, even if it's blank, in imeta tags for images and videos. (NIP-92 requires imeta tags to have at least one field besides url.)
    • Try harder to sniff whether content is HTML or plain text.
  • to_as1:
    • Add new id_format kwarg for choosing between hex and bech32-encoded ids.
    • Add new nostr_uri_ids boolean kwarg for whether to prefix ids with nostr:.
    • Add author when converting user relays events (kind 10002).
    • Populate url/urls field for more object types, using njump.me URLs.
    • Add content_is_html property.
    • Bug fix for NIP-92 imeta tags without m (MIME type) tag.
  • Rename user_url to object_url, expand to accept any NIP-19, NIP-05, or hex id. user_url is kept as an alias.
  • web_url_to_at_uri: add app.bsky.graph.list support.
  • bech32_decode, bech32_encode: add TLV support, error handling.
  • Bug fix for converting reposts, both directions.
• rss:
  • to_as1: bug fix for multiple categories.
  • Add CONTENT_TYPE_RDF.