update

.dockerignore

@@ -0,0 +1,7 @@
+# flyctl launch added from .gitignore
+**/.env
+**/.vscode
+**/.telethon
+**/__pycache__
+**/venv
+fly.toml

.github/workflows/fly-deploy.yml

@@ -0,0 +1,18 @@
+# See https://fly.io/docs/app-guides/continuous-deployment-with-github-actions/
+
+name: Fly Deploy
+on:
+  push:
+    branches:
+      - main
+jobs:
+  deploy:
+    name: Deploy app
+    runs-on: ubuntu-latest
+    concurrency: deploy-group    # optional: ensure only one action runs at a time
+    steps:
+      - uses: actions/checkout@v4
+      - uses: superfly/flyctl-actions/setup-flyctl@master
+      - run: flyctl deploy --remote-only
+        env:
+          FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }}

README.md

@@ -1,12 +1,17 @@
 # LightsOff Scheduler
 Imports https://poweron.loe.lviv.ua/ into Google Calendar USING AI(omg-omg🔥)
 
-### RUN
+### Run
 
 ```
 cp .env.template .env # update .env with values
 docker-compose --build run lightsoff-scheduler
 ```
+OR
+```
+python -m pip install -r requirements.txt
+export $(cat .env | xargs) && python telethon_app.py
+```
 
 ### Execution sample
 ![image](https://github.com/user-attachments/assets/526658df-353b-4495-b702-16a446e56521)

fly.toml

@@ -0,0 +1,16 @@
+# fly.toml app configuration file generated for lightsoff-scheduler on 2024-09-06T09:35:39+03:00
+#
+# See https://fly.io/docs/reference/configuration/ for information about how to use this file.
+#
+
+app = 'lightsoff-scheduler'
+primary_region = 'waw'
+
+[build]
+
+[[vm]]
+  size = 'shared-cpu-1x'
+
+[[env]]
+  TELEGRAM_CHANNEL = "@lvivoblenergo"
+  LOG_LEVEL = "INFO"

gcal.py

@@ -1,5 +1,5 @@
 
-from datetime import datetime, time
+from datetime import datetime, time, timedelta
 import os
 import logging
 from googleapiclient.discovery import build
@@ -12,17 +12,19 @@
 credentials = service_account.Credentials.from_service_account_file(GOOGLE_API_KEY_NAME, scopes=SCOPES)
 service = build('calendar', 'v3', credentials=credentials)
 
+def to_str(_datetime):
+    return _datetime.strftime("%Y-%m-%dT%H:%M:%S")
 
 def insert_outage(calendar_id, begin, end, summary="Power Outage", description="Scheduled power outage"):
     event = {
         "summary": summary,
         "description": description,
         "start": {
-            "dateTime": begin,
+            "dateTime": to_str(begin),
             "timeZone": "Europe/Kiev"
         },
         "end": {
-            "dateTime": end,
+            "dateTime": to_str(end),
             "timeZone": "Europe/Kiev"
         },
         "reminders": {
@@ -41,14 +43,15 @@ def insert_outage(calendar_id, begin, end, summary="Power Outage", description="
 def clear_events_for_day(calendar_id, date):
     """Clear all events for a specific day."""
     logging.info(f'Clearing day {date} for calendar {calendar_id}')
-    start_time = datetime.combine(date, time.min).isoformat() + 'Z'
+    start_time = (datetime.combine(date, time.min) - timedelta(seconds=1)).isoformat() + 'Z'
     end_time = datetime.combine(date, time.max).isoformat() + 'Z'
 
     events_result = service.events().list(
         calendarId=calendar_id, timeMin=start_time, timeMax=end_time, singleEvents=True, orderBy='startTime').execute()
 
     events = events_result.get('items', [])
 
+    print(f"Found {len(events)} events for this day.")
     if not events:
         logging.info('No events found for this day.')
     else:

ocr.py

@@ -1,27 +1,42 @@
 import anthropic
 import logging
 import json
-from datetime import datetime
+from datetime import datetime, timedelta
+import csv
+import io
 
 
-# TODO maybe will move this to Google Calendar API instead of emailing  .ics file
-def ocr_image_to_calendar_api(image_b64, group="3.2"):
+def ocr_image_to_calendar_api_multi(image_b64):
     text = _ocr_image(
-        image_b64, f"I've uploaded an image of a power outage schedule table. Please analyze it and create Google Calendar API request to update calendar for the power outages for group {group}. The image shows hours of the day across the top and group numbers in the first column. Orange cells indicate 'no energy' periods. The date is highlighted at the top. Combine consecutive outage periods into single events. Add a 10-minute notification before each event. Only respond with Google API json requests as an array add a separator '->' and date in format '%Y-%m-%d'.")
-    parts = text.split("->")
-    return parts[0], parts[1]
-
-
-def ocr_image_to_calendar_api_multi(image_b64, group="3.2"):
-    text = _ocr_image(
-        image_b64, f"Analyze image of a power outage schedule table which shows hours of the day across the top and group numbers in 1st column. Orange cells indicate 'no energy' periods. The date is highlighted at the top. Combine consecutive outage periods into single events. Only respond with a minified json object key - group number, value - array of pairs [start datetime, end], then add a separator '->' and date in format '%Y-%m-%d'.")
+        image_b64, 'I uploaded an image of a power outage schedule table which shows hours of the day across the top and group numbers in 1st column. The image may be split into 2 12-hour sub-tables. Orange cells indicate outage periods. The date is highlighted at the top. Combine consecutive outage periods into single events. Only respond with a valid minified json of format {"group": [[full datetime start, full datetime end],[full datetime start, full datetime end]]}, then add a separator "->" and date in format "%Y-%m-%d".')
+    logging.debug("Received text from OCR: %s", text)
     parts = text.split("->")
     return json.loads(parts[0].strip()), datetime.strptime(parts[1].strip(), "%Y-%m-%d")
 
 
+def ocr_image_to_timeframes(image_b64):
+    text = _ocr_image(
+        image_b64, 'Don\'t interpret the image text. Turn the large green-orange table into CSV, use 0 for orange, 1 for green, add headers, no prompt, add separator "FOR_DATE:" and date %Y-%m-%d from the top of the image.')
+    logging.info("Received text from OCR: %s", text)
+    parts = text.split("FOR_DATE:")
+    day = datetime.strptime(parts[1].strip(), "%Y-%m-%d")
+    csv_string = parts[0].strip()
+    csv_reader = csv.reader(io.StringIO(csv_string))
 
-def ocr_image_to_ics(image_b64, group="3.2"):
-    return _ocr_image(image_b64, f"I've uploaded an image of a power outage schedule table. Please analyze it and create an .ics file for the power outages for group {group}. The image shows hours of the day across the top and group numbers in the first column. Orange cells indicate 'no energy' periods. Combine consecutive outage periods into single events. Add a 10-minute notification before each event. Provide the .ics file content in a format I can easily copy and save. The date for these events is shown at the top of the image - please use that date when creating the events. Only send the .ics file content, no additional text.")
+    result = {}
+    next(csv_reader, None)  # skip the headers
+    for row in csv_reader:
+        group = row[0]
+        result[group] = []
+        for i, cell in enumerate(row[1:]):
+            if cell == "0":
+                start = day + timedelta(hours=i)
+                end = day + timedelta(hours=i + 1)
+                if i != 0 and len(result[group]) > 0 and result[group][-1] is not None and result[group][-1][1] == start:
+                    result[group][-1][1] = end
+                else:
+                    result[group].append([start, end])
+    return result, day
 
 
 def mock_ocr_to_calendar_api_multi():
@@ -30,92 +45,6 @@ def mock_ocr_to_calendar_api_multi():
     return json.loads(parts[0].strip()), datetime.strptime(parts[1].strip(), "%Y-%m-%d")
 
 
-def mock_ocr_to_calendar_api():
-    text = """
-[
-  {
-    "summary": "Power Outage",
-    "description": "Scheduled power outage for group 3.2",
-    "start": {
-      "dateTime": "2024-09-05T03:00:00+03:00",
-      "timeZone": "Europe/Kiev"
-    },
-    "end": {
-      "dateTime": "2024-09-05T05:00:00+03:00",
-      "timeZone": "Europe/Kiev"
-    },
-    "reminders": {
-      "useDefault": false,
-      "overrides": [
-        {
-          "method": "popup",
-          "minutes": 10
-        }
-      ]
-    }
-  },
-  {
-    "summary": "Power Outage",
-    "description": "Scheduled power outage for group 3.2",
-    "start": {
-      "dateTime": "2024-09-05T15:00:00+03:00",
-      "timeZone": "Europe/Kiev"
-    },
-    "end": {
-      "dateTime": "2024-09-05T17:00:00+03:00",
-      "timeZone": "Europe/Kiev"
-    },
-    "reminders": {
-      "useDefault": false,
-      "overrides": [
-        {
-          "method": "popup",
-          "minutes": 10
-        }
-      ]
-    }
-  }
-]
-->2024-09-05
-"""
-    parts = text.split("->")
-    return parts[0].strip(), parts[1].strip()
-
-
-def mock_ocr_to_ics():
-    return """BEGIN:VCALENDAR
-VERSION:2.0
-PRODID:-//Your Organization//EN
-BEGIN:VEVENT
-SUMMARY:Power Outage (Group 3.2)
-DTSTART:20240905T030000
-DTEND:20240905T050000
-DTSTAMP:20240905T000000Z
-UID:[email protected]
-DESCRIPTION:Scheduled power outage for Group 3.2
-BEGIN:VALARM
-ACTION:DISPLAY
-DESCRIPTION:Power outage starting in 10 minutes
-TRIGGER:-PT10M
-END:VALARM
-END:VEVENT
-BEGIN:VEVENT
-SUMMARY:Power Outage (Group 3.2)
-DTSTART:20240905T170000
-DTEND:20240905T190000
-DTSTAMP:20240905T000000Z
-UID:[email protected]
-DESCRIPTION:Scheduled power outage for Group 3.2
-BEGIN:VALARM
-ACTION:DISPLAY
-DESCRIPTION:Power outage starting in 10 minutes
-TRIGGER:-PT10M
-END:VALARM
-END:VEVENT
-END:VCALENDAR
-"""
-
-
 def _ocr_image(image_b64, prompt):
     client = anthropic.Anthropic()  # defaults to os.environ.get("ANTHROPIC_API_KEY)
     logging.info("Sending image to Anthropics API for OCR...")

telethon_app.py

@@ -7,17 +7,36 @@
 import json
 
 from telethon import TelegramClient, events
+from telethon.sessions import StringSession
 
 api_id = os.environ.get("TELEGRAM_API_ID")
 api_hash = os.environ.get("TELEGRAM_API_HASH")
 phone_number = os.environ.get("TELEGRAM_PHONE_NUMBER")
 channel_username = os.environ.get("TELEGRAM_CHANNEL", '@lvivoblenergo')
+session_string = os.environ.get("TELEGRAM_SESSION_STRING")
 
-subscribers = os.environ.get("EMAIL_SUBSCRIBERS").split(',')
+calendars = json.loads(os.environ.get("CALENDARS_MAPPING_JSON"))
 
-client = TelegramClient('.telethon/session_name', api_id, api_hash)
+session = StringSession(session_string) if session_string else StringSession()
+client = TelegramClient(session, api_id, api_hash)
 
-calendars = json.loads(os.environ.get("CALENDARS_MAPPING_JSON"))
+
+def clean_downloads():
+    import os
+    import shutil
+    folder = '.telethon/downloads'
+    for filename in os.listdir(folder):
+        file_path = os.path.join(folder, filename)
+        try:
+            if os.path.isfile(file_path) or os.path.islink(file_path):
+                os.unlink(file_path)
+            elif os.path.isdir(file_path):
+                shutil.rmtree(file_path)
+        except Exception as e:
+            logging.error('Failed to delete %s. Reason: %s' % (file_path, e))
+
+def to_str(_datetime):
+    return _datetime.strftime("%Y-%m-%dT%H:%M:%S")
 
 @client.on(events.NewMessage(chats=channel_username))
 async def handler(event):
@@ -29,15 +48,11 @@ async def handler(event):
         with open(photo_path, "rb") as file:
             file_data = file.read()
             image_base64 = base64.b64encode(file_data).decode("ascii")
-            # ics = ocr.ocr_image_to_ics(image_base64)
-            # ics = ocr.mock_ocr_to_ics()
-            # mail.send_email_with_invite(ics, subscribers)
-
-            # events, day = ocr.mock_ocr_to_calendar_api()
-            # events, day = ocr.ocr_image_to_calendar_api(image_base64)
             # events, day = ocr.mock_ocr_to_calendar_api_multi()
 
-            events, day = ocr.ocr_image_to_calendar_api_multi(image_base64)
+            #events, day = ocr.ocr_image_to_calendar_api_multi(image_base64)
+            events, day = ocr.ocr_image_to_timeframes(image_base64)
+            logging.debug(f"OCR results: {events}, {day}")
 
             for group, timeframes in events.items():
                 calendar_id = calendars[group]
@@ -50,12 +65,14 @@ async def handler(event):
                     gcal.insert_outage(calendar_id=calendar_id, begin=begin, end=end,
                                        summary=f"Power Outage ({group})", description=f"Scheduled power outage for {group}")
             logging.info("Updated schedules in Google Calendar")
-            # todo delete file
-
+            logging.debug("Cleaning downloads folder...")
+            clean_downloads()
+            logging.debug("Done cleaning downloads folder...")
 
 def main():
     logging.info("Starting client...")
     client.start(phone_number)
+    logging.debug("Session string: %s", client.session.save())
     client.run_until_disconnected()