diff options
| author | Craig Jennings <c@cjennings.net> | 2026-06-06 23:53:52 -0500 |
|---|---|---|
| committer | Craig Jennings <c@cjennings.net> | 2026-06-06 23:53:52 -0500 |
| commit | fef03a1f4f19c45a31928989aecb4ff8a47cea4a (patch) | |
| tree | 88ad9fa1dad62f4d63f2f3ad0b65a14b39d9079f /tests/test-pearl-fields.el | |
| parent | 4a4b1bc974a3fdb4d88af3ee575786ca363696e2 (diff) | |
| download | pearl-fef03a1f4f19c45a31928989aecb4ff8a47cea4a.tar.gz pearl-fef03a1f4f19c45a31928989aecb4ff8a47cea4a.zip | |
feat(scripts): import an org backlog into the Pearl workspace
I had 44 open tasks in the org backlog and no way to get them into Linear short of typing each one in by hand. This adds a script that reads the level-2 TODO/DOING headings under the "Pearl Open Work" section and creates one issue per task.
The priority cookie maps to Linear priority, a [#D] "someday" task lands in Icebox and the rest in Backlog, tags become labels (missing ones created), and the heading body becomes the description. The discuss/next/cleanup/pearl workflow tags are dropped rather than turned into labels. It skips a task whose title already exists, so a re-run only fills in what's missing, and the two umbrella headings plus the Resolved section are left out.
Same shape as the seed script: a pure parser and mapper with no I/O, behind a thin GraphQL client that's the only network boundary, so a fake transport routing on the operation name covers the import end to end. 8 tests across parse, mapping, and the import flow.
Diffstat (limited to 'tests/test-pearl-fields.el')
0 files changed, 0 insertions, 0 deletions
