[release] Bump to version 58.3

- Remove curl-cffi as it doesn't work on FreeBSD

PROJECT.org

@@ -89,6 +89,7 @@ fetching and simple saving.
 **** Scraper
 
 * Backlog [0/22] :vrobbler:project:personal:
+** TODO [#C] After transition to linux add curl_cffi as webpage scrapper again :webpages:metadata:
 ** TODO [#C] Create small utility to clean up tracks scrobbled with wonky playback times :bug:music:scrobbles:
 :PROPERTIES:
 :ID:       702462cf-d54b-48c6-8a7c-78b8de751deb
@@ -604,6 +605,29 @@ independent of the email flow it was originally creatdd for
 
 ** TODO [#B] Is there way to create unique slugs for media instances :media_types:
 
+* Version 58.3 [1/1]
+** DONE [#A] Remove curl-cffi as it doesn't work on FreeBSD :webpages:deps:
+:PROPERTIES:
+:ID:       6bc1b0dd-e449-3d32-a176-46451e793e5d
+:END:
+
+* Version 58.2 [2/2]
+** DONE [#B] Add more robust webpage scraping :webpages:metadata:
+:PROPERTIES:
+:ID:       84d9bfa5-75c0-0718-764e-379f7456602a
+:END:
+** DONE [#B] Time of Day Categories trend :trends:
+:PROPERTIES:
+:ID:       6598074f-2290-46db-967b-29f45d30be29
+:END:
+
+*** Description
+
+Added a "Time of Day Categories" trend that groups scrobbles for Books, Trails,
+Birding Locations, and Board Games into Early Bird (5-10:59am), Day Jay (11am-6:59pm),
+and Night Owl (7pm-4:59am) buckets. Shows both overall and per-media-type breakdowns.
+
+
 * Version 58.1 [1/1]
 ** DONE [#B] Add auto genre tagging for papers :books:papers:metadata:
 :PROPERTIES:

poetry.lock

@@ -967,6 +967,23 @@ prompt-toolkit = ">=3.0.36"
 [package.extras]
 testing = ["pytest (>=7.2.1)", "pytest-cov (>=4.0.0)", "tox (>=4.4.3)"]
 
+[[package]]
+name = "cloudscraper"
+version = "1.2.71"
+description = "A Python module to bypass Cloudflare's anti-bot page."
+optional = false
+python-versions = "*"
+groups = ["main"]
+files = [
+    {file = "cloudscraper-1.2.71-py2.py3-none-any.whl", hash = "sha256:76f50ca529ed2279e220837befdec892626f9511708e200d48d5bb76ded679b0"},
+    {file = "cloudscraper-1.2.71.tar.gz", hash = "sha256:429c6e8aa6916d5bad5c8a5eac50f3ea53c9ac22616f6cb21b18dcc71517d0d3"},
+]
+
+[package.dependencies]
+pyparsing = ">=2.4.7"
+requests = ">=2.9.2"
+requests-toolbelt = ">=0.9.1"
+
 [[package]]
 name = "colorama"
 version = "0.4.6"
@@ -6812,4 +6829,4 @@ cffi = ["cffi (>=1.17,<2.0) ; platform_python_implementation != \"PyPy\" and pyt
 [metadata]
 lock-version = "2.1"
 python-versions = ">=3.11,<3.15"
-content-hash = "3081efc33689e0cb1e2cc12709f2fbfd84a084ee4a083eebba80db2eb80fb0a6"
+content-hash = "beac677c269bb8618ca802e5f92f7558391d8d26f1b2150f3c8a6d3417848cb1"

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "vrobbler"
-version = "58.1"
+version = "58.3"
 description = ""
 authors = ["Colin Powell <colin@unbl.ink>"]
 
@@ -11,6 +11,7 @@ django-extensions = "^3.1.5"
 python-dateutil = "^2.8.2"
 python-dotenv = ">=0.20.0,<2"
 python-json-logger = "^2.0.2"
+cloudscraper = "^1.2.71"
 colorlog = "^6.6.0"
 httpx = "<=0.27.2"
 djangorestframework = "^3.13.1"

vrobbler/apps/trends/templates/trends/_time_of_day_categories.html

@@ -0,0 +1,65 @@
+<div class="row">
+  <div class="col-12">
+    {% if data.total and data.total > 0 %}
+    <h5>Overall</h5>
+    <div class="table-responsive mb-4">
+      <table class="table table-striped table-sm">
+        <thead>
+          <tr>
+            <th>Category</th>
+            <th class="text-end">Scrobbles</th>
+            <th class="text-end">%</th>
+          </tr>
+        </thead>
+        <tbody>
+          {% for slug, info in data.categories.items %}
+          <tr>
+            <td>{{ info.label }}</td>
+            <td class="text-end">{{ info.count }}</td>
+            <td class="text-end">{{ info.pct }}%</td>
+          </tr>
+          {% endfor %}
+          <tr class="table-secondary">
+            <td><strong>Total</strong></td>
+            <td class="text-end"><strong>{{ data.total }}</strong></td>
+            <td class="text-end"></td>
+          </tr>
+        </tbody>
+      </table>
+    </div>
+
+    <h5>By Media Type</h5>
+    {% for mt, mt_data in data.by_media_type.items %}
+    <h6 class="mt-3">{{ mt }}</h6>
+    <div class="table-responsive">
+      <table class="table table-striped table-sm">
+        <thead>
+          <tr>
+            <th>Category</th>
+            <th class="text-end">Scrobbles</th>
+            <th class="text-end">%</th>
+          </tr>
+        </thead>
+        <tbody>
+          {% for slug, info in mt_data.categories.items %}
+          <tr>
+            <td>{{ info.label }}</td>
+            <td class="text-end">{{ info.count }}</td>
+            <td class="text-end">{{ info.pct }}%</td>
+          </tr>
+          {% endfor %}
+          <tr class="table-secondary">
+            <td><strong>Total</strong></td>
+            <td class="text-end"><strong>{{ mt_data.total }}</strong></td>
+            <td class="text-end"></td>
+          </tr>
+        </tbody>
+      </table>
+    </div>
+    {% endfor %}
+
+    {% else %}
+    <p class="text-muted">No data found for Books, Trails, Birding Locations, or Board Games in this period.</p>
+    {% endif %}
+  </div>
+</div>

vrobbler/apps/trends/templates/trends/trend_detail.html

@@ -55,6 +55,9 @@
 {% elif trend.slug == "reading-pace-vs-activity" %}
   {% include "trends/_reading_pace.html" %}
 
+{% elif trend.slug == "time-of-day-categories" %}
+  {% include "trends/_time_of_day_categories.html" %}
+
 {% elif trend.slug == "trending-up" %}
   {% include "trends/_trending_up.html" %}
 

vrobbler/apps/trends/trends/__init__.py

@@ -15,6 +15,7 @@ from trends.trends.mood import (
     compute_mood_weather,
 )
 from trends.trends.reading import compute_reading_pace_vs_activity
+from trends.trends.time_of_day import compute_time_of_day_categories
 from trends.trends.trending import compute_trending_up
 
 TREND_REGISTRY = {}
@@ -44,5 +45,8 @@ compute_peak_hours = register("peak-hours")(compute_peak_hours)
 compute_reading_pace_vs_activity = register("reading-pace-vs-activity")(
     compute_reading_pace_vs_activity
 )
+compute_time_of_day_categories = register("time-of-day-categories")(
+    compute_time_of_day_categories
+)
 compute_trending_up = register("trending-up")(compute_trending_up)
 compute_weekly_rhythm = register("weekly-rhythm")(compute_weekly_rhythm)

vrobbler/apps/trends/trends/time_of_day.py

@@ -0,0 +1,89 @@
+from collections import OrderedDict
+
+from django.db.models import Count, Q
+from django.db.models.functions import Extract
+from scrobbles.models import Scrobble
+
+
+TARGET_MEDIA_TYPES = ["Book", "Trail", "BirdingLocation", "BoardGame"]
+
+CATEGORIES = OrderedDict(
+    [
+        ("early_bird", {"label": "Early Bird", "hours": {5, 6, 7, 8, 9, 10}}),
+        ("day_jay", {"label": "Day Jay", "hours": {11, 12, 13, 14, 15, 16, 17, 18}}),
+        ("night_owl", {"label": "Night Owl", "hours": {19, 20, 21, 22, 23, 0, 1, 2, 3, 4}}),
+    ]
+)
+
+
+def _categorize_hour(hour):
+    for slug, cat in CATEGORIES.items():
+        if hour in cat["hours"]:
+            return slug
+    return None
+
+
+def compute_time_of_day_categories(user, period="last_30"):
+    from trends.utils import get_date_range
+
+    start, end = get_date_range(period)
+    filters = Q(user=user, media_type__in=TARGET_MEDIA_TYPES, timestamp__isnull=False)
+    if start:
+        filters &= Q(timestamp__gte=start)
+    if end:
+        filters &= Q(timestamp__lte=end)
+
+    qs = (
+        Scrobble.objects.filter(filters)
+        .annotate(hour=Extract("timestamp", "hour"))
+        .values("media_type", "hour")
+        .annotate(count=Count("id"))
+        .order_by("media_type", "hour")
+    )
+
+    raw = {}
+    for row in qs:
+        mt = row["media_type"]
+        raw.setdefault(mt, {})[row["hour"]] = row["count"]
+
+    by_media_type = {}
+    grand_totals = {"early_bird": 0, "day_jay": 0, "night_owl": 0}
+    grand_total = 0
+
+    for mt in TARGET_MEDIA_TYPES:
+        mt_data = raw.get(mt, {})
+        cat_counts = {"early_bird": 0, "day_jay": 0, "night_owl": 0}
+        mt_total = 0
+        for hour, count in mt_data.items():
+            slug = _categorize_hour(hour)
+            if slug:
+                cat_counts[slug] += count
+                mt_total += count
+        by_media_type[mt] = {
+            "total": mt_total,
+            "categories": {},
+        }
+        for slug in CATEGORIES:
+            c = cat_counts[slug]
+            by_media_type[mt]["categories"][slug] = {
+                "count": c,
+                "pct": round((c / mt_total * 100), 1) if mt_total else 0,
+                "label": CATEGORIES[slug]["label"],
+            }
+            grand_totals[slug] += c
+        grand_total += mt_total
+
+    categories = {}
+    for slug in CATEGORIES:
+        c = grand_totals[slug]
+        categories[slug] = {
+            "count": c,
+            "pct": round((c / grand_total * 100), 1) if grand_total else 0,
+            "label": CATEGORIES[slug]["label"],
+        }
+
+    return {
+        "categories": categories,
+        "total": grand_total,
+        "by_media_type": by_media_type,
+    }

vrobbler/apps/trends/utils.py

@@ -25,6 +25,7 @@ TIME_BOUND_TRENDS = {
     "mood-weather",
     "peak-hours",
     "reading-pace-vs-activity",
+    "time-of-day-categories",
     "trending-up",
     "weekly-rhythm",
 }

vrobbler/apps/trends/views.py

@@ -55,6 +55,11 @@ TREND_METADATA = {
         "description": "Compare how long you read per session with and without concurrent music.",
         "icon": "📊",
     },
+    "time-of-day-categories": {
+        "title": "Time of Day Categories",
+        "description": "Are you an early bird, day jay, or night owl? Categorized by Books, Trails, Birding Locations, and Board Games.",
+        "icon": "🦉",
+    },
     "trending-up": {
         "title": "Trending Media Types",
         "description": "Which media types have you been consuming more or less of recently?",

vrobbler/apps/webpages/models.py

@@ -45,6 +45,37 @@ class Domain(TimeStampedModel):
         )
 
 
+def _fetch_url_raw(url: str) -> Optional[str]:
+    """Fetch raw HTML for a URL.
+
+    Tries two strategies in order:
+      1. trafilatura (standard, works for most sites)
+      2. cloudscraper (handles Cloudflare JS challenges)
+
+    cloudscraper is lazy-imported so deployments that cannot compile
+    its dependencies (e.g. FreeBSD) still work.
+    """
+    raw = trafilatura.fetch_url(url)
+    if raw:
+        return raw
+
+    logger.debug("trafilatura returned nothing for %s, trying cloudscraper", url)
+    try:
+        import cloudscraper
+    except ImportError:
+        logger.debug("cloudscraper not available")
+    else:
+        try:
+            scraper = cloudscraper.create_scraper()
+            resp = scraper.get(url, timeout=30)
+            resp.raise_for_status()
+            return resp.text
+        except Exception as exc:
+            logger.debug("cloudscraper failed for %s: %s", url, exc)
+
+    return None
+
+
 class WebPage(ScrobblableMixin):
     COMPLETION_PERCENT = getattr(settings, "WEBSITE_COMPLETION_PERCENT", 100)
 
@@ -80,7 +111,7 @@ class WebPage(ScrobblableMixin):
 
     def _update_extract_from_web(self, raw_text: str = "", force=True):
         if not raw_text:
-            raw_text = requests.get(self.url, headers=headers).text
+            raw_text = _fetch_url_raw(self.url)
         if not self.extract or force:
             self.extract = trafilatura.extract(raw_text)
             self.save(update_fields=["extract"])
@@ -259,7 +290,7 @@ class WebPage(ScrobblableMixin):
             return False
 
     def fetch_data_from_web(self, save=True, force=True):
-        raw_text = trafilatura.fetch_url(self.url)
+        raw_text = _fetch_url_raw(self.url)
         if not self.extract or force:
             self.extract = trafilatura.extract(
                 raw_text,