ilml-plugin-linkedin 1.20.15 → 1.20.16

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 CHANGED
@@ -1,5 +1,10 @@
1
1
  # Changelog
2
2
 
3
+ ## 1.20.16
4
+ - **Fixed: `apply` stopped submitting applications — it skipped almost every job as "already applied".** LinkedIn's new job UI dropped an internal attribute the plugin used to tell the left-hand results list apart from the open job, so `apply` mistook the "Applied" badges on OTHER jobs in that list for the open job and skipped brand-new jobs as if you'd already applied — a whole run could end up applying to nothing. It now identifies the results list by its current markers, so the "already applied" check reads only the job you're actually on.
5
+ - **Added a safety net so this can't silently waste a run again.** If the "already applied" check ever misfires like this in future (flagging job after job while submitting nothing), `apply` now notices, distrusts that check for the rest of the run, and falls back to its own record of what it has applied to — so the next time LinkedIn changes its markup, a run keeps applying instead of quietly doing nothing.
6
+ - No schema change.
7
+
3
8
  ## 1.20.15
4
9
  - **Fixed: `apply` could abort an entire run after recovering from a dropped browser connection.** On some heavy job pages the browser connection would time out mid-application; `apply` recovered the page but then hit an internal error while re-reading the job list and stopped the whole run — so one problem job left the rest of your daily quota unused (0 applications sent). It now handles that case cleanly and moves on to the next job, so a single bad page no longer wastes the run.
5
10
  - **`connect fill --from-discovered` now tops up your queue with fresh, still-live profiles instead of stale ones.** The discovery pool builds up over months, and without a personal ranking configured the top-up could surface long-dead entries — profiles that have since been deleted or renamed and now fail to load (a "404"), which then fail to connect. It now prefers the most recently discovered people and skips entries older than about six weeks, so the queue fills with people you can actually reach.