feat(edu): parse event times in /diary and drop weekend days

- fetch each event's time via Playwright (click event-link, read span.data, close fancybox) and render as 'title (HH:MM)'
- '08:00' placeholder renders as localized 'unknown' (time_unknown key in ru/uk/en)
- diary week view shows Mon-Fri only (title ends at Friday)
- diary month view skips Sat/Sun by weekday_idx with name-based fallback
- schedule keyboard drops Sat/Sun day buttons

edu_master/webinar-checker/checker.py

@@ -115,6 +115,7 @@ TRANSLATIONS = {
         'month': '📅 Весь месяц',
         'no_events': 'Нет событий',
         'no_lessons': 'Нет уроков',
+        'time_unknown': 'неизвестно',
         'free_period': 'Свободно',
         'unsubscribed': '🔕 Вы отписались от уведомлений.',
         'diary_title': '📅 <b>Дневник</b> — выберите период:',
@@ -186,6 +187,7 @@ TRANSLATIONS = {
         'month': '📅 Весь місяць',
         'no_events': 'Немає подій',
         'no_lessons': 'Немає уроків',
+        'time_unknown': 'невідомо',
         'free_period': 'Вільно',
         'unsubscribed': '🔕 Ви відписалися від сповіщень.',
         'diary_title': '📅 <b>Щоденник</b> — виберіть період:',
@@ -257,6 +259,7 @@ TRANSLATIONS = {
         'month': '📅 Whole month',
         'no_events': 'No events',
         'no_lessons': 'No lessons',
+        'time_unknown': 'unknown',
         'free_period': 'Free',
         'unsubscribed': '🔕 You have unsubscribed from notifications.',
         'diary_title': '📅 <b>Diary</b> — choose a period:',
@@ -490,7 +493,10 @@ def get_diary_keyboard(user_id: int | None = None):
 
 
 def _parse_calendar_html(table_html: str) -> tuple:
-    """Parse calendar HTML table into (month_text, {day_num: {weekday, events}})."""
+    """Parse calendar HTML table into (month_text, {day_num: {weekday, events}}).
+
+    Each event is a dict: {'title': str, 'id': site event id | None, 'time': 'HH:MM' | None}.
+    """
     days = {}
     weekdays = []
     rows = re.findall(r'<tr[^>]*>(.*?)</tr>', table_html, re.DOTALL)
@@ -522,22 +528,101 @@ def _parse_calendar_html(table_html: str) -> tuple:
                     continue
                 day_num = dm.group(1)
 
-                # Extract events: title attribute (full name) of ALL <a> tags inside the cell
+                # Extract events: title attribute (full name) of ALL <a> tags inside the cell.
+                # Each event keeps its site data-event-id and a 'time' slot (filled later
+                # from the AJAX popup by fetch_diary_data). 'time' is 'HH:MM' or None.
                 events = []
                 for a_match in re.finditer(r'<a[^>]*>(.*?)</a>', cell, re.DOTALL):
                     a_tag = a_match.group(0)
                     # Prefer the title attribute (contains full name, not truncated)
                     title_m = re.search(r'title\s*=\s*"([^"]*)"', a_tag)
                     et = title_m.group(1).strip() if title_m else re.sub(r'<[^>]+>', '', a_match.group(1)).strip()
-                    if et:
+                    if not et:
-                        events.append(et)
+                        continue
+                    id_m = re.search(r'data-event-id\s*=\s*"?(\d+)"?', a_tag)
+                    events.append(
+                        {
+                            'title': et,
+                            'id': id_m.group(1) if id_m else None,
+                            'time': None,
+                        }
+                    )
 
                 weekday = weekdays[col_idx] if col_idx < len(weekdays) else ''
-                days[day_num] = {'weekday': weekday, 'events': events}
+                days[day_num] = {'weekday': weekday, 'weekday_idx': col_idx, 'events': events}
 
     return month_text, days
 
 
+async def _collect_event_times(page) -> dict:
+    """Click every calendar `a.event-link` and read the time from the AJAX popup.
+
+    The site shows event details only after a click fills the fancybox container
+    (`div.event-full-info span.data`). Returns {event_id: 'HH:MM'} for events
+    that have a date; events without one are simply skipped. A failed click on a
+    single event never breaks the whole diary parse.
+    """
+    times_by_id: dict[str, str] = {}
+    try:
+        links = await page.query_selector_all('a.event-link')
+    except Exception as e:
+        logger.warning(f'Failed to list diary event links: {e}')
+        return times_by_id
+
+    for link in links:
+        event_id = None
+        event_deadline = time.monotonic() + 10
+        try:
+            event_id = await link.get_attribute('data-event-id')
+            if not event_id or event_id in times_by_id:
+                continue
+            await link.evaluate('(el) => el.click()')
+            await page.wait_for_selector(
+                f'div.event-full-info[data-event-full-info-id="{event_id}"] span.data',
+                timeout=3000,
+                state='visible',
+            )
+            if time.monotonic() > event_deadline:
+                break
+            time_text = await page.evaluate(
+                """(id) => {
+                    const nodes = document.querySelectorAll(
+                        'div.event-full-info[data-event-full-info-id="' + id + '"] span.data'
+                    );
+                    for (const el of nodes) {
+                        if (el.getClientRects().length > 0 || el.offsetParent !== null) {
+                            return el.textContent.trim();
+                        }
+                    }
+                    return null;
+                }""",
+                event_id,
+            )
+            if time_text:
+                m = re.search(r'(\d{1,2}:\d{2})', time_text)
+                if m:
+                    times_by_id[event_id] = m.group(1)
+        except Exception as e:
+            logger.debug(f'Failed to read time for diary event {event_id}: {e}')
+            if time.monotonic() > event_deadline:
+                break
+        finally:
+            # Close the fancybox popup (click the close button, else Escape).
+            closed = False
+            with contextlib.suppress(Exception):
+                close_btn = await page.query_selector('#fancybox-close')
+                if close_btn is not None:
+                    await close_btn.click(timeout=2000)
+                    closed = True
+            if not closed:
+                with contextlib.suppress(Exception):
+                    await page.keyboard.press('Escape')
+            await page.wait_for_timeout(150)
+
+    logger.info(f'Diary event times collected for {len(times_by_id)} events')
+    return times_by_id
+
+
 async def fetch_diary_data(phpsessid: str) -> dict | None:
     logger.info('Fetching diary data via Playwright...')
     try:
@@ -570,6 +655,16 @@ async def fetch_diary_data(phpsessid: str) -> dict | None:
                         f.write(table_html)
 
                     month_text, days = _parse_calendar_html(table_html)
+
+                    # Read event times by opening each event's AJAX popup.
+                    times_by_id = await _collect_event_times(page)
+                    if times_by_id:
+                        for day_data in days.values():
+                            for ev in day_data.get('events', []):
+                                eid = ev.get('id')
+                                if eid and eid in times_by_id:
+                                    ev['time'] = times_by_id[eid]
+
                     logger.info(
                         f'Diary parsed: month={month_text!r}, days_with_events={sum(1 for d in days.values() if d["events"])}/{len(days)}'
                     )
@@ -596,6 +691,25 @@ def _parse_diary_month(text: str) -> str:
     return text.strip()
 
 
+def _format_event_line(event, lang: str) -> str:
+    """Render one diary event line: 📌 title, with time in parens when known.
+
+    '08:00' is the site's placeholder for "no time specified" — show a localized
+    'time_unknown' mark instead. Unknown/absent time renders as before, no parens.
+    """
+    if isinstance(event, dict):
+        title = escape(str(event.get('title') or ''))
+        tme = event.get('time')
+    else:
+        title = escape(str(event))
+        tme = None
+    if tme == '08:00':
+        return f'📌 {title} ({_tr(lang, "time_unknown")})'
+    if tme:
+        return f'📌 {title} ({escape(str(tme))})'
+    return f'📌 {title}'
+
+
 def format_diary_day(data: dict, day_num: int, lang: str = 'en') -> str:
     days = data.get('days', {})
     month_str = _parse_diary_month(data.get('monthFullText', ''))
@@ -605,7 +719,7 @@ def format_diary_day(data: dict, day_num: int, lang: str = 'en') -> str:
         lines.append(_tr(lang, 'no_events'))
     else:
         for e in day_data['events']:
-            lines.append(f'📌 {escape(e)}')
+            lines.append(_format_event_line(e, lang))
     lines.append(f'\n🔗 {DIARY_URL}')
     return '\n'.join(lines)
 
@@ -614,12 +728,12 @@ def format_diary_week(data: dict, today: datetime, lang: str = 'en') -> str:
     days = data.get('days', {})
     _parse_diary_month(data.get('monthFullText', ''))
     monday = today - timedelta(days=today.weekday())
-    sunday = monday + timedelta(days=6)
+    friday = monday + timedelta(days=4)
     start = f'{monday.day}.{monday.month}'
-    end = f'{sunday.day}.{sunday.month}'
+    end = f'{friday.day}.{friday.month}'
     short = DIARY_WEEKDAYS_SHORT.get(lang, DIARY_WEEKDAYS_SHORT['en'])
     lines = [_tr(lang, 'diary_week_title', start=start, end=end) + '\n']
-    for i in range(7):
+    for i in range(5):
         d = monday + timedelta(days=i)
         day_data = days.get(str(d.day))
         lines.append(f'─ <b>{short[i]} {d.day}.{d.month}</b> ─')
@@ -627,7 +741,7 @@ def format_diary_week(data: dict, today: datetime, lang: str = 'en') -> str:
             lines.append(_tr(lang, 'no_events') + '\n')
         else:
             for e in day_data['events']:
-                lines.append(f'📌 {escape(e)}')
+                lines.append(_format_event_line(e, lang))
             lines.append('')
     lines.append(f'🔗 {DIARY_URL}')
     return '\n'.join(lines)
@@ -639,6 +753,17 @@ def format_diary_month(data: dict, lang: str = 'en') -> str:
     lines = [f'📅 <b>{month_str}</b>\n']
     for day_num in sorted(days.keys(), key=int):
         day_data = days[day_num]
+        if day_data.get('weekday_idx', 0) >= 5:
+            continue
+        if day_data.get('weekday', '').strip().lower() in (
+            'субота',
+            'суббота',
+            'saturday',
+            'неділя',
+            'воскресенье',
+            'sunday',
+        ):
+            continue
         events = day_data.get('events', [])
         weekday = day_data.get('weekday', '')
         lines.append(f'─ <b>{weekday} {day_num}</b> ─')
@@ -646,7 +771,7 @@ def format_diary_month(data: dict, lang: str = 'en') -> str:
             lines.append(_tr(lang, 'no_events') + '\n')
         else:
             for e in events:
-                lines.append(f'📌 {escape(e)}')
+                lines.append(_format_event_line(e, lang))
             lines.append('')
     lines.append(f'🔗 {DIARY_URL}')
     return '\n'.join(lines)
@@ -676,7 +801,7 @@ def get_schedule_day_keyboard(user_id: int | None = None):
     tomorrow = today + timedelta(days=1)
     today_label = _tr(lang, 'today')
     tomorrow_label = _tr(lang, 'tomorrow')
-    short_keys = ['day_mon', 'day_tue', 'day_wed', 'day_thu', 'day_fri', 'day_sat', 'day_sun']
+    short_keys = ['day_mon', 'day_tue', 'day_wed', 'day_thu', 'day_fri']
     keyboard = [
         [
             InlineKeyboardButton(
@@ -706,14 +831,6 @@ def get_schedule_day_keyboard(user_id: int | None = None):
             InlineKeyboardButton(
                 _tr(lang, short_keys[4]), callback_data=f'schedule_day_{SCHEDULE_WEEKDAYS_CANONICAL[4]}'
             ),
-            InlineKeyboardButton(
-                _tr(lang, short_keys[5]), callback_data=f'schedule_day_{SCHEDULE_WEEKDAYS_CANONICAL[5]}'
-            ),
-        ],
-        [
-            InlineKeyboardButton(
-                _tr(lang, short_keys[6]), callback_data=f'schedule_day_{SCHEDULE_WEEKDAYS_CANONICAL[6]}'
-            ),
         ],
     ]
     return InlineKeyboardMarkup(keyboard)