Merge pull request #3 from talkdai/feature/phone-number-independency

Fixes independency for phone numbers

.dockerignore

@@ -0,0 +1,162 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+data/db/
+data/

.gitignore

@@ -174,4 +174,5 @@ poetry.toml
 pyrightconfig.json
 
 # End of https://www.toptal.com/developers/gitignore/api/python
-n
+n
+project_config.toml

Dockerfile

@@ -0,0 +1,11 @@
+FROM ghcr.io/talkdai/dialog:latest
+
+# Copy the plugin code
+WORKDIR /plugin
+COPY . .
+
+RUN pip install -e .
+
+WORKDIR /app/src/
+
+RUN mkdir /app/static

main.py → dialog_whatsapp/__init__.py

@@ -3,19 +3,22 @@
 import hashlib
 
 from typing import Any
-from dialog.llm import get_llm_class
 from fastapi import APIRouter, Body, HTTPException, Query, Depends, Request
 from openai import OpenAI
 
 from .settings import (
     WHATSAPP_API_TOKEN,
     WHATSAPP_ACCOUNT_NUMBER,
-    API_HOST
+    WHATSAPP_VERIFY_TOKEN,
+    API_HOST,
+    PROJECT_CONFIG,
+    DATASET,
+    OPENAI_API_KEY,
+    ROUTE_SUFFIX
 )
-from dialog.settings import PROJECT_CONFIG
+from dialog.llm.default import DialogLLM
 
 from dialog.models.helpers import create_session
-from plugins.whats_audio.responses import whatsapp_get_response
 from uuid import uuid4
 
 client = OpenAI()
@@ -84,22 +87,33 @@ def transcribe_audio(audio_id):
     logger.info(f"Got transcript back from OpenAI - {transcript}")
     return transcript.text
 
-@router.get("/whats-audio")
+@router.get(f"/tenant-{ROUTE_SUFFIX}")
 async def whats_audio_get(request: Request):
-    content = await whatsapp_get_response(request)
-    return content
+    logging.info(request.query_params.get("hub.verify_token"))
+    logging.info(WHATSAPP_VERIFY_TOKEN)
+    if request.query_params.get("hub.verify_token") == WHATSAPP_VERIFY_TOKEN:
+        return int(request.query_params.get("hub.challenge"))
 
-@router.post("/whats-audio")
+    raise HTTPException(status_code=404)
+
+@router.post(f"/tenant-{ROUTE_SUFFIX}")
 async def whats_audio_post(body: Any = Body(None)):
     logger.info("Started")
     value = body["entry"][0]["changes"][0]["value"]
+    logging.info("Body: ", body)
 
     try:
         from_number = value["messages"][0]["from"]
         message_type = value["messages"][0].get("type")
     except KeyError:
         return {}
 
+    phone_number_id = value["metadata"]["phone_number_id"]
+    logging.info(f"Got message for number {phone_number_id}")
+    if phone_number_id != WHATSAPP_ACCOUNT_NUMBER:
+        logger.info("Phone number not supported")
+        return {"status": "success"}
+
     if message_type not in ["audio", "text"]:
         logger.info("Message type not supported")
         send_text_to_whatsapp("Ainda não consigo entender este tipo de mensagem", from_number)
@@ -114,10 +128,19 @@ async def whats_audio_post(body: Any = Body(None)):
 
     logger.info(f"Got message from {from_number} - message: {message}")
 
-    create_session(identifier=from_number)
+    session_id = from_number
+    if DATASET:
+        session_id = f"{DATASET}_{from_number}"
+
+    create_session(identifier=session_id)
 
-    LLM = get_llm_class()
-    llm = LLM(config=PROJECT_CONFIG, session_id=from_number)
+    llm = DialogLLM(
+        config=PROJECT_CONFIG,
+        session_id=session_id,
+        dataset=DATASET,
+        llm_key=OPENAI_API_KEY
+    )
+    logger.info(f"Project config: {PROJECT_CONFIG}")
     processed_message = llm.process(message)
     processed_message = processed_message["text"]
     logger.info("Processed message: %s", processed_message)
@@ -141,4 +164,7 @@ async def whats_audio_post(body: Any = Body(None)):
     else:
         send_text_to_whatsapp(processed_message, from_number)
 
-    return {}
+    return {}
+
+def register_plugin(app):
+    app.include_router(router, prefix="/whatsapp")

responses.py → dialog_whatsapp/responses.py

@@ -8,8 +8,8 @@
     WHATSAPP_VERIFY_TOKEN,
     WHATSAPP_API_TOKEN,
     WHATSAPP_ACCOUNT_NUMBER,
+    PROJECT_CONFIG
 )
-from dialog.settings import PROJECT_CONFIG
 
 from dialog.models.helpers import create_session
 
@@ -26,7 +26,6 @@ async def whatsapp_get_response(request):
 
     raise HTTPException(status_code=404)
 
-
 async def whatsapp_post_response(request, body):
     value = body["entry"][0]["changes"][0]["value"]
     try:
@@ -41,7 +40,7 @@ async def whatsapp_post_response(request, body):
         "Authorization": f"Bearer {WHATSAPP_API_TOKEN}",
         "Content-Type": "application/json",
     }
-    url = f"https://graph.facebook.com/v17.0/{WHATSAPP_ACCOUNT_NUMBER}/messages"
+    url = f"https://graph.facebook.com/v17.0/{phone_number_id}/messages"
 
     create_session(identifier=from_number)
 

dialog_whatsapp/settings.py

@@ -0,0 +1,36 @@
+import os
+import json
+import tomllib
+import logging
+
+from pathlib import Path
+from decouple import config
+from dialog.settings import PROJECT_CONFIG as DIALOG_PROJECT_CONFIG
+
+file_path = os.path.abspath(__file__)
+dir_path = os.path.dirname(file_path)
+
+if "config.json" in os.listdir(dir_path):
+    logging.info("Loading config file")
+    config_json = json.load(open(dir_path + "/config.json"))
+    WHATSAPP_VERIFY_TOKEN = config_json.get("WHATSAPP_VERIFY_TOKEN", "1234567890")
+    WHATSAPP_API_TOKEN = config_json.get("WHATSAPP_API_TOKEN", "1234567890")
+    WHATSAPP_ACCOUNT_NUMBER = config_json.get("WHATSAPP_ACCOUNT_NUMBER", "")
+    API_HOST = config_json.get("API_HOST")
+    PROJECT_CONFIG = tomllib.loads(Path(dir_path + "/" + config_json.get("PROJECT_CONFIG")).read_text())
+    DATASET = config_json.get("DATASET")
+    OPENAI_API_KEY = config_json.get("OPENAI_API_KEY")
+    ROUTE_SUFFIX = config_json.get("ROUTE_SUFFIX")
+else:
+    WHATSAPP_VERIFY_TOKEN = config("WHATSAPP_VERIFY_TOKEN", "1234567890")
+    WHATSAPP_API_TOKEN = config("WHATSAPP_API_TOKEN", "1234567890")
+    WHATSAPP_ACCOUNT_NUMBER = config("WHATSAPP_ACCOUNT_NUMBER", "")
+    API_HOST = config("API_HOST", "http://localhost:8000")
+    PROJECT_CONFIG = config(
+        "PROJECT_CONFIG",
+        cast=lambda filename: tomllib.loads(Path(filename).read_text()),
+        default={}
+    )
+    DATASET = config("DATASET", None)
+    OPENAI_API_KEY = config("OPENAI_API_KEY", None)
+    ROUTE_SUFFIX = 1

docker-compose.yml

@@ -0,0 +1,33 @@
+version: '3.3'
+services:
+  db:
+    image: ankane/pgvector:latest
+    restart: always
+    ports:
+      - '5432:5432'
+    environment:
+      POSTGRES_USER: talkdai
+      POSTGRES_PASSWORD: talkdai
+      POSTGRES_DB: talkdai
+    volumes:
+       - ./data/db:/var/lib/postgresql/data
+       - ./etc/db-ext-vector.sql:/docker-entrypoint-initdb.d/db-ext-vector.sql
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready", "-d", "talkdai", "-U", "talkdai"]
+      interval: 30s
+      timeout: 60s
+      retries: 5
+      start_period: 80s
+  dialog:
+    build: .
+    stdin_open: true
+    tty: true
+    ports:
+      - '8000:8000'
+    depends_on:
+      db:
+        condition: service_healthy
+    env_file:
+      - .env
+    volumes:
+      - ./data:/data/

project_config.sample.toml

@@ -0,0 +1,13 @@
+[model]
+model = "gpt-3.5-turbo-1106"
+temperature = 0.1
+
+[prompt]
+header = """Você está impersonando o Vitor, um operador de atendimento chamado Vitor."""
+question_signalizer = "Responda a pergunta:"
+
+fallback = """
+Você está impersonando o Vitor, um operador de atendimento chamado Vitor.
+Seja breve em suas respostas limitando-se ao máximo a 2 frases. Nunca fale que
+você é um modelo (AI), assistente virtual ou qualquer coisa do genero.
+"""