Upgrade to PTB V21.10

.github/workflows/build_image.yml

@@ -0,0 +1,41 @@
+name: DockerBuildAndPush
+
+on:
+  push:
+    branches:
+      - master
+      - developement
+      - ptb-async
+
+env:
+  IMAGE_NAME: transcriberbot
+
+jobs:
+  push:
+    runs-on: ubuntu-latest
+    if: github.event_name == 'push'
+
+    steps:
+      - uses: actions/checkout@v2
+
+      - name: Login to ghcr registry
+        run: echo "${{ secrets.GITHUB_TOKEN }}" | docker login ghcr.io -u $ --password-stdin
+
+      - name: Build image
+        run: docker build . --file Dockerfile --tag $IMAGE_NAME
+
+      - name: Push image
+        run: |
+          IMAGE_ID=ghcr.io/${{ github.repository_owner }}/$IMAGE_NAME
+          # Change all uppercase to lowercase
+          IMAGE_ID=$(echo $IMAGE_ID | tr '[A-Z]' '[a-z]')
+          # Strip git ref prefix from version
+          VERSION=$(echo "${{ github.ref }}" | sed -e 's,.*/\(.*\),\1,')
+          # Strip "v" prefix from tag name
+          [[ "${{ github.ref }}" == "refs/tags/"* ]] && VERSION=$(echo $VERSION | sed -e 's/^v//')
+          # Use Docker `latest` tag convention
+          [ "$VERSION" == "master" ] && VERSION=latest
+          echo IMAGE_ID=$IMAGE_ID
+          echo VERSION=$VERSION
+          docker tag $IMAGE_NAME $IMAGE_ID:$VERSION
+          docker push $IMAGE_ID:$VERSION

.gitignore

@@ -1,5 +1,4 @@
 # TranscriberBot-specific ignores
-config/
 media/
 
 # Generic data-related ignores

.python-version

@@ -0,0 +1 @@
+transcriber-bot-wonda

Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.9-slim
+FROM python:3.12.0-slim
 
 # Set global configs
 WORKDIR /

config/app.json

@@ -49,12 +49,24 @@
     "webm"
   ],
 
+  "ocr": {
+    "tesseract_path": "/usr/share/tesseract-ocr/5/tessdata/"
+  },
+
   "antiflood": {
     "age_threshold": 10,
     "flood_ratio": 2,
     "max_flood_ratio": 6,
     "time_threshold_warning": 4,
     "time_threshold_flood": 5,
     "timeout": 10
+  },
+
+  "whisper": {
+    "api_endpoint": "http://127.0.0.1:8000"
+  },
+
+  "logging": {
+    "level": "APP"
   }
 }

config/sentry.json

@@ -0,0 +1,3 @@
+{
+  "dsn": "xxx"
+}

requirements.txt

@@ -1,8 +1,9 @@
-python-telegram-bot==12.3.0
+python-telegram-bot
 coloredlogs
 pillow
 watchdog
 tesserocr
 pydub
 zbarlight
 requests
+sentry-sdk

run.sh

@@ -0,0 +1,15 @@
+#!/bin/sh
+
+docker pull ghcr.io/charslab/transcriberbot:ptb-async
+docker run \
+   -e LC_ALL=C \
+   -d --restart unless-stopped \
+   --name "transcriberbot-async" \
+   -v "$(pwd)"/data:/data \
+   -v "$(pwd)"/config:/config \
+   -v "$(pwd)"/values:/values \
+   -v "$(pwd)"/media:/media \
+   --cpus=4.0 \
+   --memory=3000m \
+   -u "$(id -u):1337" \
+   ghcr.io/charslab/transcriberbot:ptb-async

src/antiflood/__init__.py

@@ -2,4 +2,4 @@
 from antiflood.antiflood import register_flood_warning_callback
 from antiflood.antiflood import register_flood_started_callback
 from antiflood.antiflood import register_flood_ended_callback
-from antiflood.antiflood import init
+from antiflood.antiflood import init

src/antiflood/antiflood.py

@@ -4,11 +4,11 @@
 
 logger = logging.getLogger(__name__)
 
-flood_ratio = 2             # messages/seconds
+flood_ratio = 2  # messages/seconds
 max_flood_ratio = 10
 time_threshold_warning = 5  # ratio > flood_ratio for {time_threshold_warning} seconds
-time_threshold_flood = 10   # ratio > flood_ratio for {time_threshold_flood} seconds
-timeout = 4                 # flood ends after ratio < flood_ratio for {timeout} seconds
+time_threshold_flood = 10  # ratio > flood_ratio for {time_threshold_flood} seconds
+timeout = 4  # flood ends after ratio < flood_ratio for {timeout} seconds
 
 callback_flood_warning = None
 callback_flood_started = None
@@ -20,74 +20,77 @@
 # chat_id -> (level, ratio, msg_num, duration, last_update)
 stats = {}
 
-def register_flood_warning_callback(callback):
-  global callback_flood_warning
-  callback_flood_warning = callback
-
-def register_flood_started_callback(callback):
-  global callback_flood_started
-  callback_flood_started = callback
-
-def register_flood_ended_callback(callback):
-  global callback_flood_ended
-  callback_flood_ended = callback
-
-def init():
-  global flood_ratio, max_flood_ratio, time_threshold_warning, time_threshold_flood, timeout
-  flood_ratio = config.get_config_prop("app")["antiflood"]["flood_ratio"]
-  max_flood_ratio = config.get_config_prop("app")["antiflood"]["max_flood_ratio"]
-  time_threshold_warning = config.get_config_prop("app")["antiflood"]["time_threshold_warning"]
-  time_threshold_flood = config.get_config_prop("app")["antiflood"]["time_threshold_flood"]
-  timeout = config.get_config_prop("app")["antiflood"]["timeout"]
-
-  logger.info("Ratio: %d", flood_ratio)
-  logger.info("Max flood ratio: %d", max_flood_ratio)
-  logger.info("Thr warning: %d", time_threshold_warning)
-  logger.info("Thr flood: %d", time_threshold_flood)
-  logger.info("Timeout: %d", timeout)
 
-def on_chat_msg_received(chat_id):
-  global flood_ratio, time_threshold_warning, time_threshold_flood, timeout
-  global callback_flood_warning, callback_flood_started, callback_flood_ended
-
-  curr_time = time.time()
-
-  if chat_id not in stats:
-    stats[chat_id] = [LEVEL_NORMAL, 1.0, 1, 0.0, curr_time]
+def register_flood_warning_callback(callback):
+    global callback_flood_warning
+    callback_flood_warning = callback
 
-  else:
-    level, ratio, msg_num, duration, last_update = stats[chat_id]
-    updated_duration = duration + curr_time - last_update
-    msg_num += 1
-    curr_ratio = msg_num / updated_duration
 
-    if curr_ratio < flood_ratio and updated_duration > timeout:
-      curr_ratio, updated_duration, msg_num = 0, 0, 0
-      level = LEVEL_NORMAL
-      if callback_flood_ended:
-        callback_flood_ended(chat_id)
+def register_flood_started_callback(callback):
+    global callback_flood_started
+    callback_flood_started = callback
 
-    elif updated_duration > 1 and curr_ratio > max_flood_ratio and level < LEVEL_FLOOD:
-      level = LEVEL_FLOOD
-      logger.warning("Flood ratio for chat %d is over the top", chat_id)
-      if callback_flood_started:
-        callback_flood_started(chat_id)
 
-    elif curr_ratio > flood_ratio:
-      if updated_duration >= time_threshold_flood and level < LEVEL_FLOOD:
-        logger.warning("Flood detected for chat %d", chat_id)
-        level = LEVEL_FLOOD
-        if callback_flood_started:
-          callback_flood_started(chat_id)
+def register_flood_ended_callback(callback):
+    global callback_flood_ended
+    callback_flood_ended = callback
 
-      elif updated_duration >= time_threshold_warning and level < LEVEL_WARNING:
-        logger.info("Potential flood for chat %d", chat_id)
-        level = LEVEL_WARNING
-        if callback_flood_warning is not None:
-          callback_flood_warning(chat_id)
 
-    stats[chat_id] = (level, curr_ratio, msg_num, updated_duration, curr_time)
+def init():
+    global flood_ratio, max_flood_ratio, time_threshold_warning, time_threshold_flood, timeout
+    flood_ratio = config.get_config_prop("app")["antiflood"]["flood_ratio"]
+    max_flood_ratio = config.get_config_prop("app")["antiflood"]["max_flood_ratio"]
+    time_threshold_warning = config.get_config_prop("app")["antiflood"]["time_threshold_warning"]
+    time_threshold_flood = config.get_config_prop("app")["antiflood"]["time_threshold_flood"]
+    timeout = config.get_config_prop("app")["antiflood"]["timeout"]
 
-    logger.info("stats[{}]: {}".format(chat_id, stats[chat_id]))
+    logger.info("Ratio: %d", flood_ratio)
+    logger.info("Max flood ratio: %d", max_flood_ratio)
+    logger.info("Thr warning: %d", time_threshold_warning)
+    logger.info("Thr flood: %d", time_threshold_flood)
+    logger.info("Timeout: %d", timeout)
 
 
+def on_chat_msg_received(chat_id):
+    global flood_ratio, time_threshold_warning, time_threshold_flood, timeout
+    global callback_flood_warning, callback_flood_started, callback_flood_ended
+
+    curr_time = time.time()
+
+    if chat_id not in stats:
+        stats[chat_id] = [LEVEL_NORMAL, 1.0, 1, 0.0, curr_time]
+
+    else:
+        level, ratio, msg_num, duration, last_update = stats[chat_id]
+        updated_duration = duration + curr_time - last_update
+        msg_num += 1
+        curr_ratio = msg_num / updated_duration
+
+        if curr_ratio < flood_ratio and updated_duration > timeout:
+            curr_ratio, updated_duration, msg_num = 0, 0, 0
+            level = LEVEL_NORMAL
+            if callback_flood_ended:
+                callback_flood_ended(chat_id)
+
+        elif updated_duration > 1 and curr_ratio > max_flood_ratio and level < LEVEL_FLOOD:
+            level = LEVEL_FLOOD
+            logger.warning("Flood ratio for chat %d is over the top", chat_id)
+            if callback_flood_started:
+                callback_flood_started(chat_id)
+
+        elif curr_ratio > flood_ratio:
+            if updated_duration >= time_threshold_flood and level < LEVEL_FLOOD:
+                logger.warning("Flood detected for chat %d", chat_id)
+                level = LEVEL_FLOOD
+                if callback_flood_started:
+                    callback_flood_started(chat_id)
+
+            elif updated_duration >= time_threshold_warning and level < LEVEL_WARNING:
+                logger.info("Potential flood for chat %d", chat_id)
+                level = LEVEL_WARNING
+                if callback_flood_warning is not None:
+                    callback_flood_warning(chat_id)
+
+        stats[chat_id] = (level, curr_ratio, msg_num, updated_duration, curr_time)
+
+        logger.info("stats[{}]: {}".format(chat_id, stats[chat_id]))

src/audiotools/__init__.py

@@ -1 +1 @@
-from audiotools.speech import transcribe
+from audiotools.speech import transcribe