ilml-plugin-linkedin 1.20.2 → 1.20.4
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/CHANGELOG.md +10 -0
- package/dist/connect.mjs +51 -51
- package/dist/enrichProfile.mjs +10 -10
- package/dist/enrichProfiles.mjs +34 -35
- package/dist/funnelRunner.mjs +38 -38
- package/dist/messageBot.mjs +74 -75
- package/dist/readPost.mjs +10 -10
- package/dist/run.mjs +58 -59
- package/dist/scout.mjs +38 -39
- package/dist/src/saveLoginState.mjs +4 -4
- package/dist/syncAll.mjs +79 -80
- package/dist/syncConnections.mjs +30 -31
- package/dist/verifyAbout.mjs +18 -19
- package/dist/version.mjs +1 -1
- package/dist/viewers.mjs +20 -21
- package/dist/visit.mjs +36 -37
- package/ilml-plugin.json +1 -1
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,15 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
+
## 1.20.4
|
|
4
|
+
- **Fixed: commands no longer fail while LinkedIn is still showing its "setting things up" loading screen.** LinkedIn now often shows a roughly 30-second initialization screen and only then redirects to the page you asked for. The plugin used to give up before that finished — which showed up as false "not logged in / redirected to login" errors on sync, a message box that "never appeared" when sending, and hit-or-miss inbox searches. Every command now waits through that loading screen (up to 45 seconds) before deciding anything, so a slow LinkedIn load no longer looks like a failure. No schema change.
|
|
5
|
+
|
|
6
|
+
## 1.20.3
|
|
7
|
+
- **Fixed: an interrupted or crashed run no longer leaves a hidden browser open that blocks your next command.** Before, if a command was interrupted (Ctrl-C), hit an error, or had its output piped to something that closed early, its automation browser could stay running in the background — and the next `connect` / `sync` / `apply` would then fail with a "the profile is already open" error. Two layers now prevent this:
|
|
8
|
+
- Every command reliably closes its own browser when it exits — on normal finish, Ctrl-C, termination, an uncaught error, or a closed output pipe.
|
|
9
|
+
- As a backstop for a hard kill or power-loss (where nothing gets a chance to run), the next command automatically detects the leftover browser and closes it gracefully before starting — but only when the run that owned it is truly gone, so a second command you start on purpose while another is running is never disturbed.
|
|
10
|
+
|
|
11
|
+
Net effect: a run can no longer strand a browser that jams your next command, and you never need to hunt down and close a stray window by hand. No schema change.
|
|
12
|
+
|
|
3
13
|
## 1.20.2
|
|
4
14
|
- **Discovery is now a normal setting you can turn on — no environment variable needed.** The experimental "build a pool of people to reach next" feature (from 1.20.0) is controlled by a config toggle: `ilml plugin config linkedin set DISCOVERY_ENABLED true`. It's still **off by default**. When on, `connect run` quietly notes LinkedIn's "People also viewed / More profiles for you" suggestions as it works (no extra page loads), and `ilml linkedin connect fill --from-discovered` tops up your connection queue from them, ranked by relevance to you. See `ilml linkedin help connect` for the full command reference and `ilml plugin config linkedin` for the setting. No schema change.
|
|
5
15
|
|