Compare commits

..
Author SHA1 Message Date
Typosaurus 08b791d67a ticket #113 attempt 1 2026-07-23 03:18:03 +00:00
4 changed files with 155 additions and 116 deletions
File diff suppressed because one or more lines are too long
+3 -1
View File
@@ -4,6 +4,8 @@ from __future__ import annotations
from typing import Any, Optional from typing import Any, Optional
from pydantic import Field
from devplacepy.schemas.base import _Out from devplacepy.schemas.base import _Out
@@ -62,7 +64,7 @@ class PollOut(_Out):
class BadgeOut(_Out): class BadgeOut(_Out):
name: Optional[str] = None name: Optional[str] = Field(None, alias="badge_name")
created_at: Optional[str] = None created_at: Optional[str] = None
+15 -18
View File
@@ -22,8 +22,6 @@ from .pdf import MAX_PDF_BYTES, extract_pdf_text, is_pdf
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
_pw_lock = asyncio.Lock()
RSEARCH_URL = "https://rsearch.app.molodetz.nl" RSEARCH_URL = "https://rsearch.app.molodetz.nl"
RSEARCH_TIMEOUT_SECONDS = 45.0 RSEARCH_TIMEOUT_SECONDS = 45.0
FETCH_TIMEOUT_SECONDS = 20.0 FETCH_TIMEOUT_SECONDS = 20.0
@@ -174,7 +172,13 @@ async def _render_with_playwright(
) -> tuple[str, str, int, list[tuple[str, str]]]: ) -> tuple[str, str, int, list[tuple[str, str]]]:
from playwright.async_api import async_playwright from playwright.async_api import async_playwright
async def _render(browser) -> tuple[str, str, int, list[tuple[str, str]]]: 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:
context = await browser.new_context(user_agent=USER_AGENT) context = await browser.new_context(user_agent=USER_AGENT)
page = await context.new_page() page = await context.new_page()
response = await page.goto(url, wait_until="load", timeout=30000) response = await page.goto(url, wait_until="load", timeout=30000)
@@ -185,18 +189,10 @@ async def _render_with_playwright(
await context.close() await context.close()
extracted = extract_html(content, base_url=url) extracted = extract_html(content, base_url=url)
return extracted.title, extracted.text, status, extracted.links 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: finally:
if own_browser:
await browser.close() await browser.close()
else: await pw.__aexit__(None, None, None)
return await _render(browser)
async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None: async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None:
@@ -246,12 +242,8 @@ async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None:
except (LookupError, ValueError) as exc: except (LookupError, ValueError) as exc:
logger.info("deepsearch decode failed for %s: %s", url, exc) logger.info("deepsearch decode failed for %s: %s", url, exc)
if len(text) < MIN_PAGE_CHARS: if len(text) < MIN_PAGE_CHARS:
async with _pw_lock:
try: try:
r_title, r_text, r_status, r_links = await _render_with_playwright(url, browser) 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): if len(r_text) > len(text):
title, text, status, source, links = ( title, text, status, source, links = (
r_title or title, r_title or title,
@@ -260,6 +252,8 @@ async def fetch_page(url: str, depth: int, browser=None) -> CrawledPage | None:
"playwright", "playwright",
r_links, r_links,
) )
except Exception as exc:
logger.info("deepsearch render failed for %s: %s", url, exc)
if len(text) < MIN_PAGE_CHARS: if len(text) < MIN_PAGE_CHARS:
return None return None
return CrawledPage( return CrawledPage(
@@ -302,9 +296,10 @@ async def crawl(
total = min(len(level_candidates), max_pages) total = min(len(level_candidates), max_pages)
cancelled = False cancelled = False
pw = None
browser = None browser = None
async with async_playwright() as pw:
try: try:
pw = await async_playwright().__aenter__()
browser = await pw.chromium.launch( browser = await pw.chromium.launch(
headless=True, args=["--no-sandbox", "--disable-dev-shm-usage"] headless=True, args=["--no-sandbox", "--disable-dev-shm-usage"]
) )
@@ -397,4 +392,6 @@ async def crawl(
finally: finally:
if browser is not None: if browser is not None:
await browser.close() await browser.close()
if pw is not None:
await pw.__aexit__(None, None, None)
return outcome return outcome
+40
View File
@@ -407,3 +407,43 @@ def test_viewing_profile_marks_notification_read(app_server):
refresh_snapshot() refresh_snapshot()
assert bool(get_table("notifications").find_one(uid=notif_uid)["read"]) is True 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"