ticket #138 attempt 1

Reviewed-on: #124

devplacepy/schemas/content.py

@@ -4,8 +4,6 @@ from __future__ import annotations
 
 from typing import Any, Optional
 
-from pydantic import Field
-
 from devplacepy.schemas.base import _Out
 
 
@@ -64,7 +62,7 @@ class PollOut(_Out):
 
 
 class BadgeOut(_Out):
-    name: Optional[str] = Field(None, alias="badge_name")
+    name: Optional[str] = None
     created_at: Optional[str] = None
 
 

devplacepy/services/jobs/deepsearch/crawl.py

@@ -22,6 +22,8 @@ from .pdf import MAX_PDF_BYTES, extract_pdf_text, is_pdf
 
 logger = logging.getLogger(__name__)
 
+_pw_lock = asyncio.Lock()
+
 RSEARCH_URL = "https://rsearch.app.molodetz.nl"
 RSEARCH_TIMEOUT_SECONDS = 45.0
 FETCH_TIMEOUT_SECONDS = 20.0
@@ -172,13 +174,7 @@ async def _render_with_playwright(
 ) -> tuple[str, str, int, list[tuple[str, str]]]:
     from playwright.async_api import async_playwright
 
-    own_browser = browser is None
-    if own_browser:
-        pw = await async_playwright().__aenter__()
-        browser = await pw.chromium.launch(
-            headless=True, args=["--no-sandbox", "--disable-dev-shm-usage"]
-        )
-    try:
+    async def _render(browser) -> tuple[str, str, int, list[tuple[str, str]]]:
         context = await browser.new_context(user_agent=USER_AGENT)
         page = await context.new_page()
         response = await page.goto(url, wait_until="load", timeout=30000)
@@ -189,10 +185,18 @@ async def _render_with_playwright(
         await context.close()
         extracted = extract_html(content, base_url=url)
         return extracted.title, extracted.text, status, extracted.links
+
+    if browser is None:
+        async with async_playwright() as pw:
+            browser = await pw.chromium.launch(
+                headless=True, args=["--no-sandbox", "--disable-dev-shm-usage"]
+            )
+            try:
+                return await _render(browser)
             finally:
-        if own_browser:
                 await browser.close()
-            await pw.__aexit__(None, None, None)
+    else:
+        return await _render(browser)
 
 
 async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None:
@@ -242,8 +246,12 @@ async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None:
         except (LookupError, ValueError) as exc:
             logger.info("deepsearch decode failed for %s: %s", url, exc)
     if len(text) < MIN_PAGE_CHARS:
+        async with _pw_lock:
             try:
                 r_title, r_text, r_status, r_links = await _render_with_playwright(url, browser)
+            except Exception as exc:
+                logger.info("deepsearch render failed for %s: %s", url, exc)
+                r_title, r_text, r_status, r_links = "", "", 0, []
         if len(r_text) > len(text):
             title, text, status, source, links = (
                 r_title or title,
@@ -252,8 +260,6 @@ async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None:
                 "playwright",
                 r_links,
             )
-        except Exception as exc:
-            logger.info("deepsearch render failed for %s: %s", url, exc)
     if len(text) < MIN_PAGE_CHARS:
         return None
     return CrawledPage(
@@ -296,10 +302,9 @@ 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().__aenter__()
             browser = await pw.chromium.launch(
                 headless=True, args=["--no-sandbox", "--disable-dev-shm-usage"]
             )
@@ -392,6 +397,4 @@ async def crawl(
         finally:
             if browser is not None:
                 await browser.close()
-        if pw is not None:
-            await pw.__aexit__(None, None, None)
     return outcome

devplacepy/static/css/base.css

@@ -620,6 +620,10 @@ img {
 .topnav-logo span { color: var(--accent); }
 .topnav-links { display: flex; gap: 0.25rem; }
 .topnav-link {
+    display: inline-flex;
+    align-items: center;
+    gap: 0.375rem;
+    white-space: nowrap;
     padding: 0.5rem 0.75rem;
     border-radius: var(--radius);
     font-size: 0.875rem;

tests/api/profile/index.py

@@ -407,43 +407,3 @@ def test_viewing_profile_marks_notification_read(app_server):
 
     refresh_snapshot()
     assert bool(get_table("notifications").find_one(uid=notif_uid)["read"]) is True
-
-
-def test_profile_json_badge_names_non_null(app_server):
-    from datetime import datetime, timezone
-    from uuid import uuid4
-    from devplacepy.database import get_table, refresh_snapshot
-    from devplacepy.utils import generate_uid
-
-    uid = str(uuid4())
-    username = f"badge_{uid[:8]}"
-    get_table("users").insert(
-        {
-            "uid": uid,
-            "username": username,
-            "email": f"{uid[:8]}@badge.test",
-            "password_hash": "x",
-            "role": "Member",
-            "is_active": True,
-            "created_at": datetime.now(timezone.utc).isoformat(),
-        }
-    )
-    badge_uid = generate_uid()
-    get_table("badges").insert(
-        {
-            "uid": badge_uid,
-            "user_uid": uid,
-            "badge_name": "On Fire",
-            "created_at": datetime.now(timezone.utc).isoformat(),
-        }
-    )
-    refresh_snapshot()
-
-    r = requests.get(
-        f"{BASE_URL}/profile/{username}", headers={"Accept": "application/json"}
-    )
-    assert r.status_code == 200
-    body = r.json()
-    for badge in body.get("badges", []):
-        assert badge["name"] is not None, f"Badge name is None: {badge}"
-        assert badge["name"] == "On Fire"