Keep DeepSearch crawling when the Playwright driver cannot start

The degradation guard in crawl() wrapped only chromium.launch(), while
the driver start sat outside it in the async context manager. A failure
to start the driver therefore propagated out of crawl() and failed the
whole DeepSearch job, contradicting the warning it logs on that path
("pages will use httpx only").

Start the driver inside the guarded block and stop it in the finally, so
an unavailable driver degrades to httpx-only fetching as intended. The
crawl loop body is unchanged apart from indentation.

devplacepy/services/jobs/deepsearch/crawl.py

@@ -302,9 +302,10 @@ async def crawl(
     total = min(len(level_candidates), max_pages)
     cancelled = False
 
+    pw = None
     browser = None
-    async with async_playwright() as pw:
     try:
+        pw = await async_playwright().start()
         browser = await pw.chromium.launch(
             headless=True, args=["--no-sandbox", "--disable-dev-shm-usage"]
         )
@@ -397,4 +398,6 @@ async def crawl(
     finally:
         if browser is not None:
             await browser.close()
+        if pw is not None:
+            await pw.stop()
     return outcome

tests/unit/services/jobs/deepsearch/crawl.py

@@ -110,3 +110,28 @@ def test_crawl_falls_back_to_snippet_when_fetch_empty(monkeypatch):
     )
     assert len(outcome.pages) == 1
     assert outcome.pages[0].source == "search"
+
+
+def test_crawl_degrades_to_httpx_when_the_playwright_driver_cannot_start(monkeypatch):
+    import playwright.async_api as playwright_api
+
+    class FailingDriver:
+        async def start(self):
+            raise RuntimeError("driver missing")
+
+    monkeypatch.setattr(playwright_api, "async_playwright", lambda: FailingDriver())
+
+    async def fake_fetch(url, depth, browser=None):
+        return CrawledPage(
+            url=url, title="T", text="body " * 100, source="httpx", status=200, depth=depth
+        )
+
+    monkeypatch.setattr(crawl_module, "fetch_page", fake_fetch)
+    candidates = [
+        {"url": "https://example.com/a", "title": "A", "description": "", "content": ""}
+    ]
+    outcome = run_async(
+        crawl(candidates, 6, lambda f: None, lambda u: False, _no_stop, query="q", depth=1)
+    )
+    assert len(outcome.pages) == 1
+    assert outcome.pages[0].source == "httpx"