npub1tkq8unhsd5jqx6ueex5lcpsgknrpquxuk44ftpjlpm3ulaake7xs76txrw/ngit-cli-mirror/src/lib/client.rs, branch v2.3.0

feat(cover-note): add kind-1624 cover notes for PRs, patches, and issues

2026-03-05T14:23:07+00:00

Implements experimental kind-1624 cover note events: - KIND_COVER_NOTE constant and process_cover_note() in git_events.rs; replaceable semantics (latest created_at, hex-id tiebreak), author or maintainer only - kind-1624 events fetched alongside labels in the fetch pipeline; cover_notes count added to FetchReport display - ngit pr/issue view: cover note displayed in place of description with a clear 'Cover Note:' header; maintainer-authored notes identify the author; original description shown only with --comments; cover_note object included in --json output - ngit pr set-cover-note / ngit issue set-cover-note: publish a kind-1624 event; nostr: mentions in --body converted to q/p tags via tags_from_content (same rules as issue --body) - Fix pre-existing clippy::too_many_lines on repo/mod.rs show_info

feat(labels): fetch and apply NIP-32 kind-1985 label events

2026-03-05T11:28:36+00:00

- Add KIND_LABEL (kind 1985) constant to git_events.rs - Add get_labels() merging inline t-tags with external kind-1985 events, gating each on author-or-maintainer permission - Extend get_fetch_filters() to request kind-1985 events for all known issue and proposal IDs - Track label event counts in FetchReport (field + Display + consolidation) - Update issue_list.rs and list.rs to fetch label events from cache and pass them through get_labels() instead of reading t-tags inline

feat: fetch and display NIP-22 comment counts on issues and proposals

2026-03-04T14:50:01+00:00

Download kind-1111 NIP-22 comments from relays and show a comment count in the detail view of `ngit list ` and `ngit issue list `. - git_events: add KIND_COMMENT constant (kind 1111) - client: import KIND_COMMENT; add `comments` field to FetchReport; route kind-1111 events into report.comments in process_fetched_events; consolidate comments across relay reports; display "N comment(s)" in FetchReport; add a #E-tagged kind-1111 filter in get_fetch_filters covering all known issue and proposal root IDs - issue_list: add get_comment_counts() to query the local cache for kind-1111 events by #E tag and count per issue; extend the filtered tuple with comment_count; show a CMTS column in the table, a "Comments: N" line in the detail view, and a "comments" field in JSON - list: add KIND_COMMENT import; add resolve_event_id() helper and get_comment_count_for_proposal() to look up the count for a single proposal from the local cache; pass comment_count into show_proposal_details(); display "Comments: N" in plain text and "comments" in JSON; align detail-view labels

feat: add ngit issue list command

2026-03-04T13:28:30+00:00

Non-interactive listing of NIP-34 issues (kind 1621) with status resolution, hashtag display, and detail view. - get_issues_from_cache: fetch GitIssue events from local cache by repo coordinate, mirroring get_proposals_and_revisions_from_cache - ngit issue list: table output of ID, status, title and hashtags at end of each row; status resolved via existing get_status() logic - --status: comma-separated filter (open,draft,closed,applied; default: open) - --hashtag: comma-separated label filter (case-insensitive, OR match) - --json: machine-readable output including hashtags and description - --offline: skip network fetch, use local cache only - : optional positional argument (hex event-id or nevent) to show full details of a specific issue including body content

feat: fetch and report NIP-34 issues and their statuses

2026-03-04T13:12:26+00:00

Download kind-1621 issues from relays into the local cache alongside patches and PRs. Issue IDs are tracked separately from proposal IDs throughout the fetch pipeline so they never appear in proposal lists. - get_fetch_filters: include Kind::GitIssue in the #a-tag filter and add a dedicated issue_ids parameter to fetch status events (1630-1633) tagged with known issue IDs - FetchRequest: add issue_ids field populated from the local cache - fetch_all_from_relay: track fresh_issue_roots independently of fresh_proposal_roots; loop continues until both are drained - process_fetched_events: route GitIssue events into fresh_issue_roots and report.issues; route status events to issue_statuses or statuses based on whether the root ID is a known issue or proposal - FetchReport: add issues and issue_statuses fields, consolidated across relays, and displayed as "N issue(s), M issue status(es)" in the report

fix: include KIND_PULL_REQUEST IDs in proposal_ids for fetch filters

2026-02-27T16:11:36+00:00

kind-5 deletion events that only #e-tag a PR Kind event (kind 1618) were never fetched because proposal_ids was built from Kind::GitPatch events only. The deleted PR event remained in the local cache and continued to appear as a remote ref (refs/heads/pr/).

fix: update fallback signer relays

2026-02-27T14:26:48+00:00

Replace nsec.app with bucket.coracle.social and nos.lol.

feat: track and display kind-5 deletion count in FetchReport

2026-02-26T21:13:29+00:00

Add a deletions counter to FetchReport, incremented in process_fetched_events for each kind-5 event received and accumulated across relays in consolidate_fetch_reports. The count is included in the fetch summary display (e.g. "1 deletion", "2 deletions").

feat: request kind-5 deletions for state/announcement events by #e

2026-02-26T21:12:27+00:00

Add a non_proposal_event_ids parameter to get_fetch_filters and a corresponding field on FetchRequest. On each fetch, the event IDs of cached repo announcements and the state event are collected and used to build a dedicated kind-5 filter keyed on #e tags, as specified by NIP-09. The existing #a-tagged filter already covers addressable-event deletions; this new filter catches deletions from clients that follow NIP-09 strictly and do not embed a repo coordinate in their deletion event.

fix: show detailed relay view when --verbose flag is set in send_events

2026-02-26T15:26:21+00:00

The use_concise flag in send_events was computed with || instead of &&, causing !is_test to short-circuit to true in production and making --verbose have no effect on published event display.