@torrent-tv/proxy 2.87.21 → 2.88.1

package/test/encode-exit.test.js

@@ -1,111 +1,111 @@
-/**
- * @file The four things an ffmpeg exit can mean, by their field cases.
- *
- * Each test names the release the classification cost. None of them needs a
- * process: what went wrong in every case was the reading of the exit, not the
- * handling of it.
- */
-
-import assert from "node:assert/strict";
-import test from "node:test";
-
-import { ENCODE_EXIT, classifyEncodeExit } from "../services/encode/encode-exit.js";
-
-test("the predecessor a seek kills says nothing about the session", () => {
-  // The handler decides "is this exit mine" by comparing against the session's
-  // current process — and during a restart that field still names the process
-  // being killed, because the replacement is spawned afterwards. So a SIGKILLed
-  // predecessor passed the check and was read as the session's own run dying:
-  // a spurious "failed" that a segment request landing in that window is
-  // answered 500 for, and on a hardware host a permanent downgrade to libx264.
-  assert.equal(
-    classifyEncodeExit({ superseded: true, code: null, inputUnavailable: false }),
-    ENCODE_EXIT.IGNORED
-  );
-  // Including when it looks like a clean finish.
-  assert.equal(classifyEncodeExit({ superseded: true, code: 0 }), ENCODE_EXIT.IGNORED);
-});
-
-test("exit 0 short of the last segment is a failure, not a finished file", () => {
-  // 2.9.104. A run that had made 188 segments of 624 reported itself complete,
-  // the player consumed what was on disk and froze on the first segment nobody
-  // was making.
-  assert.equal(
-    classifyEncodeExit({ code: 0, producedThrough: 188, lastSegmentIndex: 623 }),
-    ENCODE_EXIT.SHORT
-  );
-  assert.equal(
-    classifyEncodeExit({ code: 0, producedThrough: 623, lastSegmentIndex: 623 }),
-    ENCODE_EXIT.COMPLETE
-  );
-});
-
-test("exit 0 is complete when there is nothing to check it against", () => {
-  // No published last segment, or nothing readable on disk: the claim cannot be
-  // contradicted, so it stands. Guessing "short" here would fail every session
-  // whose directory could not be read.
-  assert.equal(classifyEncodeExit({ code: 0, producedThrough: null, lastSegmentIndex: 623 }), ENCODE_EXIT.COMPLETE);
-  assert.equal(classifyEncodeExit({ code: 0, producedThrough: 12, lastSegmentIndex: null }), ENCODE_EXIT.COMPLETE);
-});
-
-test("data that went away is recoverable, and is not an encoder fault", () => {
-  // Field 2026-08-06: a torrent evicted mid-seek took the film with it, the run
-  // died on `File 0 not found`, and the session answered 500 from then on while
-  // the swarm was there and the data would have come back in seconds.
-  assert.equal(
-    classifyEncodeExit({ code: 1, inputUnavailable: true }),
-    ENCODE_EXIT.INPUT_LOST
-  );
-  // And the distinction is what keeps a working hardware encoder: the caller
-  // downgrades to software on FAILED, so an input that vanished must not
-  // arrive there.
-  assert.notEqual(classifyEncodeExit({ code: 1, inputUnavailable: true }), ENCODE_EXIT.FAILED);
-});
-
-test("anything else is a failure of this target", () => {
-  assert.equal(classifyEncodeExit({ code: 1, inputUnavailable: false }), ENCODE_EXIT.FAILED);
-  assert.equal(classifyEncodeExit({ code: null, inputUnavailable: false }), ENCODE_EXIT.FAILED);
-});
-
-test("every combination answers, and no input produces undefined", () => {
-  for (const superseded of [true, false]) {
-    for (const code of [0, 1, null]) {
-      for (const producedThrough of [null, 0, 10]) {
-        for (const lastSegmentIndex of [null, 0, 10]) {
-          for (const inputUnavailable of [true, false]) {
-            const answer = classifyEncodeExit({
-              superseded,
-              code,
-              producedThrough,
-              lastSegmentIndex,
-              inputUnavailable
-            });
-            assert.ok(
-              Object.values(ENCODE_EXIT).includes(answer),
-              `answered ${String(answer)} for ${JSON.stringify({ superseded, code, producedThrough, lastSegmentIndex, inputUnavailable })}`
-            );
-          }
-        }
-      }
-    }
-  }
-  assert.ok(Object.values(ENCODE_EXIT).includes(classifyEncodeExit()));
-});
-
-test("a run that made nothing has not finished, whatever its exit code", () => {
-  // Field 2026-09-11: a run given #541..#541 was handed a start later than its
-  // own end, wrote 190 bytes that are not a fragment, and exited zero. Making
-  // no segment and being unable to read the directory were the same `null`, and
-  // the second reading — "cannot be contradicted, so it stands" — was applied
-  // to the first.
-  assert.equal(
-    classifyEncodeExit({ code: 0, producedThrough: null, producedCount: 0, lastSegmentIndex: 541 }),
-    ENCODE_EXIT.SHORT
-  );
-  // And the case that reading was written for still stands: nothing readable on
-  // disk is not a claim that nothing was made.
-  assert.equal(
-    classifyEncodeExit({ code: 0, producedThrough: null, producedCount: null, lastSegmentIndex: 623 }),
-    ENCODE_EXIT.COMPLETE
-  );
-});
+/**
+ * @file The four things an ffmpeg exit can mean, by their field cases.
+ *
+ * Each test names the release the classification cost. None of them needs a
+ * process: what went wrong in every case was the reading of the exit, not the
+ * handling of it.
+ */
+
+import assert from "node:assert/strict";
+import test from "node:test";
+
+import { ENCODE_EXIT, classifyEncodeExit } from "../services/encode/encode-exit.js";
+
+test("the predecessor a seek kills says nothing about the session", () => {
+  // The handler decides "is this exit mine" by comparing against the session's
+  // current process — and during a restart that field still names the process
+  // being killed, because the replacement is spawned afterwards. So a SIGKILLed
+  // predecessor passed the check and was read as the session's own run dying:
+  // a spurious "failed" that a segment request landing in that window is
+  // answered 500 for, and on a hardware host a permanent downgrade to libx264.
+  assert.equal(
+    classifyEncodeExit({ superseded: true, code: null, inputUnavailable: false }),
+    ENCODE_EXIT.IGNORED
+  );
+  // Including when it looks like a clean finish.
+  assert.equal(classifyEncodeExit({ superseded: true, code: 0 }), ENCODE_EXIT.IGNORED);
+});
+
+test("exit 0 short of the last segment is a failure, not a finished file", () => {
+  // 2.9.104. A run that had made 188 segments of 624 reported itself complete,
+  // the player consumed what was on disk and froze on the first segment nobody
+  // was making.
+  assert.equal(
+    classifyEncodeExit({ code: 0, producedThrough: 188, lastSegmentIndex: 623 }),
+    ENCODE_EXIT.SHORT
+  );
+  assert.equal(
+    classifyEncodeExit({ code: 0, producedThrough: 623, lastSegmentIndex: 623 }),
+    ENCODE_EXIT.COMPLETE
+  );
+});
+
+test("exit 0 is complete when there is nothing to check it against", () => {
+  // No published last segment, or nothing readable on disk: the claim cannot be
+  // contradicted, so it stands. Guessing "short" here would fail every session
+  // whose directory could not be read.
+  assert.equal(classifyEncodeExit({ code: 0, producedThrough: null, lastSegmentIndex: 623 }), ENCODE_EXIT.COMPLETE);
+  assert.equal(classifyEncodeExit({ code: 0, producedThrough: 12, lastSegmentIndex: null }), ENCODE_EXIT.COMPLETE);
+});
+
+test("data that went away is recoverable, and is not an encoder fault", () => {
+  // Field 2026-08-06: a torrent evicted mid-seek took the film with it, the run
+  // died on `File 0 not found`, and the session answered 500 from then on while
+  // the swarm was there and the data would have come back in seconds.
+  assert.equal(
+    classifyEncodeExit({ code: 1, inputUnavailable: true }),
+    ENCODE_EXIT.INPUT_LOST
+  );
+  // And the distinction is what keeps a working hardware encoder: the caller
+  // downgrades to software on FAILED, so an input that vanished must not
+  // arrive there.
+  assert.notEqual(classifyEncodeExit({ code: 1, inputUnavailable: true }), ENCODE_EXIT.FAILED);
+});
+
+test("anything else is a failure of this target", () => {
+  assert.equal(classifyEncodeExit({ code: 1, inputUnavailable: false }), ENCODE_EXIT.FAILED);
+  assert.equal(classifyEncodeExit({ code: null, inputUnavailable: false }), ENCODE_EXIT.FAILED);
+});
+
+test("every combination answers, and no input produces undefined", () => {
+  for (const superseded of [true, false]) {
+    for (const code of [0, 1, null]) {
+      for (const producedThrough of [null, 0, 10]) {
+        for (const lastSegmentIndex of [null, 0, 10]) {
+          for (const inputUnavailable of [true, false]) {
+            const answer = classifyEncodeExit({
+              superseded,
+              code,
+              producedThrough,
+              lastSegmentIndex,
+              inputUnavailable
+            });
+            assert.ok(
+              Object.values(ENCODE_EXIT).includes(answer),
+              `answered ${String(answer)} for ${JSON.stringify({ superseded, code, producedThrough, lastSegmentIndex, inputUnavailable })}`
+            );
+          }
+        }
+      }
+    }
+  }
+  assert.ok(Object.values(ENCODE_EXIT).includes(classifyEncodeExit()));
+});
+
+test("a run that made nothing has not finished, whatever its exit code", () => {
+  // Field 2026-09-11: a run given #541..#541 was handed a start later than its
+  // own end, wrote 190 bytes that are not a fragment, and exited zero. Making
+  // no segment and being unable to read the directory were the same `null`, and
+  // the second reading — "cannot be contradicted, so it stands" — was applied
+  // to the first.
+  assert.equal(
+    classifyEncodeExit({ code: 0, producedThrough: null, producedCount: 0, lastSegmentIndex: 541 }),
+    ENCODE_EXIT.SHORT
+  );
+  // And the case that reading was written for still stands: nothing readable on
+  // disk is not a claim that nothing was made.
+  assert.equal(
+    classifyEncodeExit({ code: 0, producedThrough: null, producedCount: null, lastSegmentIndex: 623 }),
+    ENCODE_EXIT.COMPLETE
+  );
+});