@sjawhar/pi-legion 10.26.0 → 10.27.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/legion.js CHANGED
@@ -16956,7 +16956,7 @@ async function bootLegionClaim(host2) {
16956
16956
  // package.json
16957
16957
  var package_default = {
16958
16958
  name: "@sjawhar/pi-legion",
16959
- version: "10.26.0",
16959
+ version: "10.27.1",
16960
16960
  license: "Apache-2.0",
16961
16961
  type: "module",
16962
16962
  omp: {
@@ -24,9 +24,16 @@ line.
24
24
  ## What a mention carries
25
25
 
26
26
  The delivery's `message` holds Slack's fields: `channel_id`, `user_id` (a Slack user id such as
27
- `U0123ABCD`), `text` (the mention itself appears in it as `<@U…>`, the app's own id), `ts` (the
28
- mention's own timestamp), `thread_ts` (the thread's root, only when the mention was posted inside a
29
- thread), and `bot_id` and `bot_name` when a bot posted it.
27
+ `U0123ABCD`), `text` (the message's own words, then the display text its blocks and then its
28
+ attachments carry, a line apiece; a line an object has already shown under another of its own
29
+ names is left out, and so is one equal to the message's own words or to a line of them, while two
30
+ objects saying the same thing both deliver it; an attachment's `fallback` comes last of that
31
+ attachment and only when the attachment did not say it already; a mention a person typed
32
+ in a Slack client carries its own words alone, since a `rich_text` block that only repeats those
33
+ words is not read, while an app's rich text, which says more, is delivered; the mention itself
34
+ appears in it as `<@U…>`, the app's own
35
+ id), `ts` (the mention's own timestamp), `thread_ts` (the thread's root, only when the mention was
36
+ posted inside a thread), and `bot_id` and `bot_name` when a bot posted it.
30
37
 
31
38
  ## Answer in the thread
32
39
 
@@ -254,16 +254,20 @@ reviewing until you finish.
254
254
  - **Nothing waits in the session for a deploy.** A `pending` ends the look: the daemon suspends
255
255
  your claim with its session and the issue's volume kept, the pod goes, and the daemon wakes the
256
256
  check (the pod comes back and your session resumes with a new check task) on Dispatch's
257
- `issue.deployed` event for the pull request, or, for `sjawhar/legion`, on a Legion daemon's boot
258
- on a build containing the merge, on Dispatch's `server.started` naming such a build, or on the
259
- build the daemon read from Dispatch at its boot. The woken look's task names the event
260
- (`Dispatch's issue.deployed: run <id> completed <t>`, `the Legion daemon's boot on build <sha>`,
261
- `Dispatch's start on build <sha>`, `Dispatch's build <sha>`), and its `Production:` line names
262
- it too, so the pull request carries the cause. Never poll for the deploy, never wait on a reply,
263
- and never ask anyone for approval to deploy. Open a `dispatch ask` to a person only when the
264
- instructions name nobody for that deploy; it starts with the production gap and why it matters,
265
- then names the required install or restart step, its risk, and outcome-named options, and the
266
- look still ends `pending`.
257
+ `issue.deployed` event for the pull request; when the project's `legion.yaml` names the
258
+ repository's deploy workflow (`projects.<KEY>.deploys`), at each completed run of that workflow
259
+ on the default branch whose named step (such as `Apply`) succeeded after the merge, once per such
260
+ job; or, for `sjawhar/legion`, on a Legion daemon's boot on a build containing the merge, on
261
+ Dispatch's `server.started` naming such a build, or on the build the daemon read from Dispatch at
262
+ its boot. A wake while a look is open is kept, and a `pending` then starts the next look at once.
263
+ The woken look's task names the event (`Dispatch's issue.deployed: run <id> completed <t>`,
264
+ `the deploy workflow's run: <repo> <workflow> run <id> attempt <n>, job <job name>: <step> succeeded at <t>`,
265
+ `the Legion daemon's boot on build <sha>`, `Dispatch's start on build <sha>`,
266
+ `Dispatch's build <sha>`), and its `Production:` line names it too, so the pull request carries
267
+ the cause. Never poll for the deploy, never wait on a reply, and never ask anyone for approval
268
+ to deploy. Open a `dispatch ask` to a person only when the instructions name nobody for that
269
+ deploy; it starts with the production gap and why it matters, then names the required install or
270
+ restart step, its risk, and outcome-named options, and the look still ends `pending`.
267
271
  - **The bounds.** A look is bounded at 2 hours: one that reports nothing ends as a `pending`
268
272
  would. The whole check is bounded by the project's `production_check_hours` (336, 14 days, by
269
273
  default) from the close, and the wake at that bound is the last look, told so in its task
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@sjawhar/pi-legion",
3
- "version": "10.26.0",
3
+ "version": "10.27.1",
4
4
  "license": "Apache-2.0",
5
5
  "type": "module",
6
6
  "omp": {