@plune-ai/cli 0.9.0 → 0.9.1

package/README.md

@@ -103,7 +103,7 @@ provider API key is read from the environment based on `provider.type`:
 | `plune login` | Save a [Plune platform](https://plune.ai/platform) API token so `sync` and `ingest` can reach it. The token is **checked against the API before it is saved**, so a wrong one fails here rather than two commands later. Get one at `https://beta.plune.ai` → Settings → API tokens. Flags: `--token <token>` (omit to paste it or pipe it via stdin), `--skip-verify` (save without checking, for offline setup). |
 | `plune logout` | Remove the saved token. |
 | `plune sync` | Upload the latest local run to the platform. Flags: `--file <path>` to send a specific run JSON. |
-| `plune run import <file>` | Turn a **JUnit XML** or **Playwright JSON** report into a run in Plune. Needs no provider key — nothing is generated. Flags: `--format junit\|playwright-json` (detected from the file when omitted), `--key <externalKey>` to land several reports in one run, `--create` to offer unmatched tests to the review queue. |
+| `plune run import <file>` | Turn a **JUnit XML** or **Playwright JSON** report into a run in Plune. Needs no provider key — nothing is generated. Flags: `--format junit\|playwright-json` (detected from the file when omitted), `--key <externalKey>` to land several reports in one run (with `PLUNE_SHARED_RUN=1` — see below), `--create` to offer unmatched tests to the review queue. |
 | `plune ingest [dir]` | Record a [Cairn](https://github.com/plune-ai/cairn) run in Plune. Omit `[dir]` for the newest run under `./runs`, or name the directory holding `report.json`. Generated cases arrive as **review proposals** — nothing is created until a person approves it. |
 
 Global flags: `-c, --config <path>` · `-v, --verbose` · `--no-color`.
@@ -135,14 +135,30 @@ which reports as the run happens and needs no second step:
 reporter: [['list'], ['@plune-ai/playwright']],
 ```
 
-Sharded CI? Give every job the same `--key` (or `PLUNE_RUN`), and close the run with
-`plune run finish <id>` once they are all done.
+### Several jobs, one run
+
+Two suites, or a sharded matrix, report into a single run when every job shares a key **and** knows
+it is not the last one:
+
+```bash
+PLUNE_SHARED_RUN=1 plune run import ./junit.xml     --key "$GITHUB_RUN_ID"
+PLUNE_SHARED_RUN=1 plune run import ./app/junit.xml --key "$GITHUB_RUN_ID"
+plune run finish "$RUN_ID"                          # once, when they are all done
+```
+
+`PLUNE_SHARED_RUN=1` is what stops a job closing a run its siblings are still reporting into; the
+key alone says where the results go, not who ends the run. Leave it unset in the last job and that
+job closes the run instead of the explicit `finish`.
+
+`plune run exec` sets it for you — anything reporting inside it, this command included, joins
+without closing.
 
 ## Optional: keep a history
 
-Everything above works with no account, no network, and no token — that does not change. If you
-also want run history, trends, and a shared dashboard, the three cloud commands push your local
-runs to the [Plune platform](https://plune.ai/platform):
+Everything above works with no account, no network, and no token — that does not change. `run`,
+`report`, `diff` and `init` never open a socket, and that is the half this promise is about. If you
+also want run history, trends, and a shared dashboard, the commands below push your local runs to
+the [Plune platform](https://plune.ai/platform):
 
 ```bash
 plune login          # paste the API token from your platform settings page

package/dist/cli.cjs

@@ -2373,6 +2373,7 @@ async function startRun(passed, expected) {
     rejected: 0,
     unresolved: 0,
     offered: 0,
+    unoffered: 0,
     deferred: 0
   };
   const resolved = /* @__PURE__ */ new Map();
@@ -2465,13 +2466,17 @@ async function startRun(passed, expected) {
   }
   async function offer() {
     if (discoveries.size === 0 || offline || runId === null) return;
-    for (const batch of chunk([...discoveries.values()], DISCOVER_MAX)) {
+    const batches = chunk([...discoveries.values()], DISCOVER_MAX);
+    for (const [index, batch] of batches.entries()) {
       const out = await client.post("/v1/review-items/discovered", {
         discovered: batch,
         runId
       });
       if (!out.ok) {
-        log(`plune: could not offer ${batch.length} unknown tests for review (${out.detail || out.kind}).`);
+        stats.unoffered = batches.slice(index).reduce((n, rest) => n + rest.length, 0);
+        log(
+          `plune: could not offer ${stats.unoffered} unknown test(s) for review (${out.detail || out.kind}).`
+        );
         return;
       }
       stats.offered += out.body.results.filter((r) => r.outcome === "queued").length;
@@ -2501,7 +2506,10 @@ async function startRun(passed, expected) {
       }
     }
     if (unanswered.length > 0) defer(unanswered);
-    if (submissions.length === 0) return;
+    if (submissions.length === 0) {
+      progress();
+      return;
+    }
     const out = await client.post(
       `/v1/runs/${runId}/results`,
       { results: submissions }
@@ -2511,6 +2519,7 @@ async function startRun(passed, expected) {
       stats.duplicate += out.body.counts.duplicate ?? 0;
       stats.conflict += out.body.counts.conflict ?? 0;
       stats.rejected += out.body.counts.rejected ?? 0;
+      progress();
       return;
     }
     if (out.kind === "auth") {
@@ -2523,6 +2532,12 @@ async function startRun(passed, expected) {
     }
     defer(submissions);
   }
+  function progress() {
+    const total = expected?.length ?? 0;
+    if (total <= batchSize) return;
+    const done2 = stats.accepted + stats.duplicate + stats.conflict + stats.rejected + stats.unresolved;
+    log(`plune: ${Math.min(done2, total)} of ${total} results sent`);
+  }
   function summarise() {
     const parts = [`${stats.accepted} accepted`];
     if (stats.duplicate > 0) parts.push(`${stats.duplicate} already reported`);
@@ -3180,7 +3195,8 @@ async function handleRunImport(options) {
     results.map((r) => r.keys)
   );
   for (const result2 of results) await session.add(result2);
-  await session.finish();
+  if (env.keepOpen) await session.leaveOpen();
+  else await session.finish();
   const stats = session.stats;
   const result = {
     runId: session.runId,
@@ -3192,6 +3208,7 @@ async function handleRunImport(options) {
     rejected: stats.rejected,
     unresolved: stats.unresolved,
     offered: stats.offered,
+    unoffered: stats.unoffered,
     deferred: stats.deferred,
     unlocatable: results.filter((r) => r.specRef === void 0).length,
     offering: options.create === true
@@ -3205,7 +3222,11 @@ function runUrl2(apiUrl, id) {
 }
 function describe(result, apiUrl) {
   const lines = [
-    `Imported ${result.parsed} result(s) from a ${result.format} report: ${result.accepted} accepted, ${result.duplicate} already there, ${result.unresolved} unmatched.`
+    // `Read`, not `Imported`: the verb describes the FILE, which was read whatever happened next.
+    // «Imported 555 result(s): 0 accepted, 0 already there, 0 unmatched» was a sentence whose two
+    // halves disagreed, and readers stop at the verb — a run that reached nobody read as one that
+    // worked (#622).
+    `Read ${result.parsed} result(s) from a ${result.format} report: ${result.accepted} accepted, ${result.duplicate} already there, ${result.unresolved} unmatched.`
   ];
   if (result.conflict > 0 || result.rejected > 0) {
     lines.push(`${result.conflict} conflicted and ${result.rejected} were refused.`);
@@ -3219,9 +3240,14 @@ function describe(result, apiUrl) {
     lines.push(`${result.offered} unknown test(s) offered to the review queue.`);
   } else if (result.unresolved > 0 && !result.offering) {
     lines.push("Nothing was offered to the review queue \u2014 pass --create to propose the unmatched tests.");
-  } else if (result.unresolved > 0) {
+  } else if (result.unresolved > 0 && result.unoffered === 0) {
     lines.push("Nothing new to offer \u2014 the unmatched tests are already in the review queue.");
   }
+  if (result.unoffered > 0) {
+    lines.push(
+      `${result.unoffered} more could not be offered \u2014 the review queue is full. Approve or reject what is waiting, then import this report again to offer the rest.`
+    );
+  }
   if (result.unlocatable > 0) {
     lines.push(
       `${result.unlocatable} test(s) carry no file or class in the report, so they cannot be offered for review \u2014 a reviewer would have nowhere to look.`

package/dist/reporter-core.cjs

@@ -239,6 +239,7 @@ async function startRun(passed, expected) {
     rejected: 0,
     unresolved: 0,
     offered: 0,
+    unoffered: 0,
     deferred: 0
   };
   const resolved = /* @__PURE__ */ new Map();
@@ -331,13 +332,17 @@ async function startRun(passed, expected) {
   }
   async function offer() {
     if (discoveries.size === 0 || offline || runId === null) return;
-    for (const batch of chunk([...discoveries.values()], DISCOVER_MAX)) {
+    const batches = chunk([...discoveries.values()], DISCOVER_MAX);
+    for (const [index, batch] of batches.entries()) {
       const out = await client.post("/v1/review-items/discovered", {
         discovered: batch,
         runId
       });
       if (!out.ok) {
-        log(`plune: could not offer ${batch.length} unknown tests for review (${out.detail || out.kind}).`);
+        stats.unoffered = batches.slice(index).reduce((n, rest) => n + rest.length, 0);
+        log(
+          `plune: could not offer ${stats.unoffered} unknown test(s) for review (${out.detail || out.kind}).`
+        );
         return;
       }
       stats.offered += out.body.results.filter((r) => r.outcome === "queued").length;
@@ -367,7 +372,10 @@ async function startRun(passed, expected) {
       }
     }
     if (unanswered.length > 0) defer(unanswered);
-    if (submissions.length === 0) return;
+    if (submissions.length === 0) {
+      progress();
+      return;
+    }
     const out = await client.post(
       `/v1/runs/${runId}/results`,
       { results: submissions }
@@ -377,6 +385,7 @@ async function startRun(passed, expected) {
       stats.duplicate += out.body.counts.duplicate ?? 0;
       stats.conflict += out.body.counts.conflict ?? 0;
       stats.rejected += out.body.counts.rejected ?? 0;
+      progress();
       return;
     }
     if (out.kind === "auth") {
@@ -389,6 +398,12 @@ async function startRun(passed, expected) {
     }
     defer(submissions);
   }
+  function progress() {
+    const total = expected?.length ?? 0;
+    if (total <= batchSize) return;
+    const done2 = stats.accepted + stats.duplicate + stats.conflict + stats.rejected + stats.unresolved;
+    log(`plune: ${Math.min(done2, total)} of ${total} results sent`);
+  }
   function summarise() {
     const parts = [`${stats.accepted} accepted`];
     if (stats.duplicate > 0) parts.push(`${stats.duplicate} already reported`);

package/dist/reporter-core.d.cts

@@ -139,6 +139,14 @@ interface RunStats extends SubmitCounts {
     /** Unresolved tests offered to the review queue under `PLUNE_CREATE=1` (D14). Counted separately
      * from `unresolved` because they are the same tests seen twice, not a second population. */
     offered: number;
+    /**
+     * Unmatched tests that were NOT offered, because the queue would not take them (#627).
+     *
+     * Separate from `offered` rather than a shortfall computed against `unresolved`: the reason a
+     * test went unoffered is the caller's next action, and a subtraction cannot carry a reason. A
+     * queue at capacity means "empty it and import again"; nothing else here means that.
+     */
+    unoffered: number;
     /** Results written to the fallback file instead of the platform. */
     deferred: number;
 }

package/dist/reporter-core.d.ts

@@ -139,6 +139,14 @@ interface RunStats extends SubmitCounts {
     /** Unresolved tests offered to the review queue under `PLUNE_CREATE=1` (D14). Counted separately
      * from `unresolved` because they are the same tests seen twice, not a second population. */
     offered: number;
+    /**
+     * Unmatched tests that were NOT offered, because the queue would not take them (#627).
+     *
+     * Separate from `offered` rather than a shortfall computed against `unresolved`: the reason a
+     * test went unoffered is the caller's next action, and a subtraction cannot carry a reason. A
+     * queue at capacity means "empty it and import again"; nothing else here means that.
+     */
+    unoffered: number;
     /** Results written to the fallback file instead of the platform. */
     deferred: number;
 }

package/dist/reporter-core.js

@@ -196,6 +196,7 @@ async function startRun(passed, expected) {
     rejected: 0,
     unresolved: 0,
     offered: 0,
+    unoffered: 0,
     deferred: 0
   };
   const resolved = /* @__PURE__ */ new Map();
@@ -288,13 +289,17 @@ async function startRun(passed, expected) {
   }
   async function offer() {
     if (discoveries.size === 0 || offline || runId === null) return;
-    for (const batch of chunk([...discoveries.values()], DISCOVER_MAX)) {
+    const batches = chunk([...discoveries.values()], DISCOVER_MAX);
+    for (const [index, batch] of batches.entries()) {
       const out = await client.post("/v1/review-items/discovered", {
         discovered: batch,
         runId
       });
       if (!out.ok) {
-        log(`plune: could not offer ${batch.length} unknown tests for review (${out.detail || out.kind}).`);
+        stats.unoffered = batches.slice(index).reduce((n, rest) => n + rest.length, 0);
+        log(
+          `plune: could not offer ${stats.unoffered} unknown test(s) for review (${out.detail || out.kind}).`
+        );
         return;
       }
       stats.offered += out.body.results.filter((r) => r.outcome === "queued").length;
@@ -324,7 +329,10 @@ async function startRun(passed, expected) {
       }
     }
     if (unanswered.length > 0) defer(unanswered);
-    if (submissions.length === 0) return;
+    if (submissions.length === 0) {
+      progress();
+      return;
+    }
     const out = await client.post(
       `/v1/runs/${runId}/results`,
       { results: submissions }
@@ -334,6 +342,7 @@ async function startRun(passed, expected) {
       stats.duplicate += out.body.counts.duplicate ?? 0;
       stats.conflict += out.body.counts.conflict ?? 0;
       stats.rejected += out.body.counts.rejected ?? 0;
+      progress();
       return;
     }
     if (out.kind === "auth") {
@@ -346,6 +355,12 @@ async function startRun(passed, expected) {
     }
     defer(submissions);
   }
+  function progress() {
+    const total = expected?.length ?? 0;
+    if (total <= batchSize) return;
+    const done2 = stats.accepted + stats.duplicate + stats.conflict + stats.rejected + stats.unresolved;
+    log(`plune: ${Math.min(done2, total)} of ${total} results sent`);
+  }
   function summarise() {
     const parts = [`${stats.accepted} accepted`];
     if (stats.duplicate > 0) parts.push(`${stats.duplicate} already reported`);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@plune-ai/cli",
-  "version": "0.9.0",
+  "version": "0.9.1",
   "description": "AI-powered assertion testing for LLM apps — write assertions, run them against any provider, get a pass/fail report (local, CI, or diff).",
   "keywords": [
     "ai",