Merge pull request #53 from justinmcp/feat/docker-env

Changes to environment loading.

Dockerfile

@@ -23,5 +23,7 @@ RUN pip install --target="/install" /src
 FROM python:${PY_VERSION}-slim
 ARG PY_VERSION
 COPY --from=builder /install /usr/local/lib/python${PY_VERSION}/site-packages
-COPY gpt3discord.py /bin/gpt3discord
-CMD ["python3", "/bin/gpt3discord"]
+RUN mkdir -p /opt/gpt3discord/etc
+COPY gpt3discord.py /opt/gpt3discord/bin/
+COPY image_optimizer_pretext.txt conversation_starter_pretext.txt conversation_starter_pretext_minimal.txt /opt/gpt3discord/share/
+CMD ["python3", "/opt/gpt3discord/bin/gpt3discord.py"]

README.md

@@ -302,12 +302,15 @@ To build:
 - Clone repository and build *(hopefully eventually we'll add CI to automatically build + push to docker hub)*
   - `docker build -t gpt3discord .`
   - *From repository root or supply path to repository*
-- Make a env file to bind mount to /bin/.env
+- Make a .env file to bind mount to `/opt/gpt3discord/etc/environment`
 - Optional: Make a data directory + bind mount it
-  - Add `DATA_DIR=/data` to env file
+  - Add `DATA_DIR=/data` to env file -> `usage.txt` is made here
+  - Add `SHARE_DIR=/data/share` to env file -> this is where `conversation starters, optimizer pretext and the 'openers' folder` is loaded from
+  - If `SHARE_DIR` is not included it'll load from the files added during the docker image build
 - Run via docker:
-  - `docker run [-d] --name gpt3discord -v env_file:/bin/.env [-v /containers/gpt3discord:/data] gpt3discord`
-  - You can also mount a second volume and set `DATA_DIR` in the env file to keep persistent data
+  - `docker run [-d] --name gpt3discord -v env_file:/opt/gpt3discord/etc/environment [-v /containers/gpt3discord:/data] [-v /containers/gpt3discord/share:/data/share] gpt3discord`
+  - You can also mount extra volumes and set `DATA_DIR` and `SHARE_DIR` in the env file to keep persistent data
+  - `env_file` should be replaced with where you have your .env file stored on your machine
 
 This can also be run via screen/tmux or detached like a daemon.
 

cogs/gpt_3_commands_and_converser.py

@@ -71,7 +71,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
         self.pinecone_service = pinecone_service
 
         try:
-            conversation_file_path = data_path / "conversation_starter_pretext.txt"
+            conversation_file_path = EnvService.find_shared_file("conversation_starter_pretext.txt")
             # Attempt to read a conversation starter text string from the file.
             with conversation_file_path.open("r") as f:
                 self.CONVERSATION_STARTER_TEXT = f.read()
@@ -80,9 +80,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
                 )
             assert self.CONVERSATION_STARTER_TEXT is not None
 
-            conversation_file_path_minimal = (
-                data_path / "conversation_starter_pretext_minimal.txt"
-            )
+            conversation_file_path_minimal = EnvService.find_shared_file("conversation_starter_pretext_minimal.txt")
             with conversation_file_path_minimal.open("r") as f:
                 self.CONVERSATION_STARTER_TEXT_MINIMAL = f.read()
                 print(
@@ -1136,7 +1134,7 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
             ):  # only load in files if it's included in the command, if not pass on as normal
                 if opener_file.endswith(".txt"):
                     # Load the file and read it into opener
-                    opener_file = f"openers{separator}{opener_file}"
+                    opener_file = EnvService.find_shared_file(f"openers{separator}{opener_file}")
                     opener_file = await self.load_file(opener_file, ctx)
                     if (
                         not opener
@@ -1204,7 +1202,8 @@ class GPT3ComCon(discord.Cog, name="GPT3ComCon"):
                 thread_message,
             )
             self.awaiting_responses.remove(user_id_normalized)
-            self.awaiting_thread_responses.remove(thread.id)
+            if thread.id in self.awaiting_thread_responses:
+                self.awaiting_thread_responses.remove(thread.id)
 
         self.conversation_thread_owners[user_id_normalized] = thread.id
 

cogs/image_prompt_optimizer.py

@@ -34,9 +34,7 @@ class ImgPromptOptimizer(discord.Cog, name="ImgPromptOptimizer"):
         self.deletion_queue = deletion_queue
 
         try:
-            image_pretext_path = (
-                self.converser_cog.data_path / "image_optimizer_pretext.txt"
-            )
+            image_pretext_path = EnvService.find_shared_file("image_optimizer_pretext.txt")
             # Try to read the image optimizer pretext from
             # the file system
             with image_pretext_path.open("r") as file:

gpt3discord.py

@@ -5,7 +5,6 @@ from pathlib import Path
 
 import discord
 import pinecone
-from dotenv import load_dotenv
 from pycord.multicog import apply_multicog
 import os
 
@@ -16,9 +15,6 @@ if sys.platform == "win32":
 else:
     separator = "/"
 
-print("The environment file is located at " + os.getcwd() + separator + ".env")
-load_dotenv(dotenv_path=os.getcwd() + separator + ".env")
-
 from cogs.draw_image_generation import DrawDallEService
 from cogs.gpt_3_commands_and_converser import GPT3ComCon
 from cogs.image_prompt_optimizer import ImgPromptOptimizer
@@ -26,6 +22,7 @@ from models.deletion_service_model import Deletion
 from models.message_model import Message
 from models.openai_model import Model
 from models.usage_service_model import UsageService
+from models.env_service_model import EnvService
 
 __version__ = "4.0.1"
 
@@ -86,12 +83,12 @@ async def on_application_command_error(
 
 
 async def main():
-    data_path = Path(os.environ.get("DATA_DIR", os.getcwd()))
+    data_path = EnvService.environment_path_with_fallback("DATA_DIR")
     debug_guild = int(os.getenv("DEBUG_GUILD"))
     debug_channel = int(os.getenv("DEBUG_CHANNEL"))
 
     if not data_path.exists():
-        raise OSError(f"{data_path} does not exist ... create it?")
+        raise OSError(f"Data path: {data_path} does not exist ... create it?")
 
     # Load the main GPT3 Bot service
     bot.add_cog(

models/autocomplete_model.py

@@ -5,6 +5,7 @@ import re
 import discord
 from models.usage_service_model import UsageService
 from models.openai_model import Model
+from models.env_service_model import EnvService
 
 usage_service = UsageService(Path(os.environ.get("DATA_DIR", os.getcwd())))
 model = Model(usage_service)
@@ -51,7 +52,7 @@ class File_autocompleter:
         try:
             return [
                 file
-                for file in os.listdir("openers")
+                for file in os.listdir(EnvService.find_shared_file("openers"))
                 if file.startswith(ctx.value.lower())
             ][
                 :25

models/env_service_model.py

@@ -1,7 +1,24 @@
+import os
+import sys
+from pathlib import Path
 from dotenv import load_dotenv
 
-load_dotenv()
-import os
+
+def app_root_path():
+    app_path = Path(sys.argv[0]).resolve()
+    try:
+        if app_path.parent.name == "bin": # Installed in unixy hierachy
+            return app_path.parents[1]
+    except IndexError:
+        pass
+    return app_path.parent
+
+# None will let direnv do its' thing
+env_paths = [Path(".env"), app_root_path() / "etc/environment", None]
+
+for env_path in env_paths:
+    print("Loading environment from " + str(env_path))
+    load_dotenv(dotenv_path=env_path)
 
 
 class EnvService:
@@ -9,6 +26,33 @@ class EnvService:
     def __init__(self):
         self.env = {}
 
+    @staticmethod
+    def environment_path_with_fallback(env_name, relative_fallback = None):
+        dir = os.getenv(env_name)
+        if dir != None:
+            return Path(dir).resolve()
+
+        if relative_fallback:
+            app_relative = (app_root_path() / relative_fallback).resolve()
+            if app_relative.exists():
+                return app_relative
+
+        return Path.cwd()
+
+    @staticmethod
+    def find_shared_file(file_name):
+        share_file_paths = []
+        share_dir = os.getenv("SHARE_DIR") 
+        if share_dir != None:
+            share_file_paths.append(share_dir) 
+        share_file_paths.extend([app_root_path() / "share" / file_name, app_root_path() / file_name, Path(file_name)])
+        
+        for share_file_path in share_file_paths:
+            if share_file_path.exists():
+                return share_file_path.resolve()
+
+        raise ValueError(f"Unable to find shared data file {file_name}")
+
     @staticmethod
     def get_allowed_guilds():
         # ALLOWED_GUILDS is a comma separated list of guild ids