Merge pull request #154 from griptape-ai:dev

Dev

README.md

@@ -100,9 +100,14 @@ You can previous and download more examples [here](examples/README.md).
 18. Persistent Display Text: https://youtu.be/9229bN0EKlc?si=Or2eu3Nuh7lxgfEU
 19. Convert an Agent to a Tool.. and give it to another Agent: https://youtu.be/CcRot5tVAU8?si=lA0v5kDH51nYWwgG
 20. Text-To-Speech Nodes: https://youtu.be/PP1uPkRmvoo?si=QSaWNCRsRaIERrZ4
+21. Update to Managing Your API Keys in ComfyUI Griptape Nodes: https://www.youtube.com/watch?v=v80A7rtIjko
 
 ## Recent Changelog
 
+### Nov 4, 2024
+* Fixed bug where OPENAI_API_KEY was still required, and was causing some install issues.
+* Added video to README with how to manage api keys.
+
 ### Nov 1, 2024
 * Major reworking of how API keys are set. Now you can use the ComfyUI Settings window and add your API keys there. This should simplify things quite a bit as you no longer need to create a `.env` file in your ComfyUI folder.
 

js/versions.js

@@ -1,11 +1,18 @@
 export const versions = {
-  "version": "0.34.0e",
-  "releaseDate": "2024-10-31",
+  "version": "0.34.0f",
+  "releaseDate": "2024-11-4",
   "name": "ComfyUI-Griptape",
   "description": "Griptape integration for ComfyUI",
   "author": "Jason Schleifer",
   "repository": "https://github.com/griptape-ai/ComfyUI-Griptape",
   "changelog": [
+    {
+      "version": "0.34.0f",
+      "date": "2024-11-4",
+      "changes": [
+        "Fixed install issue",
+      ]
+    }, 
     {
       "version": "0.34.0e",
       "date": "2024-11-1",

nodes/config/gtUIStructureConfig.py

@@ -17,28 +17,7 @@
 from ...py.griptape_settings import GriptapeSettings
 from ..gtUIBase import gtUIBase
 
-# Check for the default OPENAI_API_KEY being set
 default_env = "OPENAI_API_KEY"
-settings = GriptapeSettings()
-has_openai_key = settings.get_settings_key_or_use_env(default_env) is not None
-if not has_openai_key:
-    default_chat_prompt_driver = DummyPromptDriver()
-    default_image_generation_driver = DummyImageGenerationDriver()
-    default_embedding_driver = DummyEmbeddingDriver()
-    default_text_to_speech_driver = DummyTextToSpeechDriver()
-    default_audio_transcription_driver = DummyAudioTranscriptionDriver()
-else:
-    default_chat_prompt_driver = OpenAiChatPromptDriver(model="gpt-4o")
-    default_image_generation_driver = OpenAiImageGenerationDriver(model="dall-e-3")
-    default_chat_prompt_driver = OpenAiChatPromptDriver(model="gpt-4o")
-    default_image_generation_driver = OpenAiImageGenerationDriver(model="dall-e-3")
-    default_embedding_driver = OpenAiEmbeddingDriver()
-    default_text_to_speech_driver = OpenAiTextToSpeechDriver(
-        model="tts-1", voice="alloy"
-    )
-    default_audio_transcription_driver = OpenAiAudioTranscriptionDriver(
-        model="whisper-1"
-    )
 
 
 class gtUIStructureConfig(gtUIBase):
@@ -47,6 +26,32 @@ class gtUIStructureConfig(gtUIBase):
     """
 
     def __init__(self):
+        # Check for the default OPENAI_API_KEY being set
+        settings = GriptapeSettings()
+        has_openai_key = settings.get_settings_key_or_use_env(default_env) is not None
+        if not has_openai_key:
+            self.default_chat_prompt_driver = DummyPromptDriver()
+            self.default_image_generation_driver = DummyImageGenerationDriver()
+            self.default_embedding_driver = DummyEmbeddingDriver()
+            self.default_text_to_speech_driver = DummyTextToSpeechDriver()
+            self.default_audio_transcription_driver = DummyAudioTranscriptionDriver()
+        else:
+            self.default_chat_prompt_driver = OpenAiChatPromptDriver(model="gpt-4o")
+            self.default_image_generation_driver = OpenAiImageGenerationDriver(
+                model="dall-e-3"
+            )
+            self.default_chat_prompt_driver = OpenAiChatPromptDriver(model="gpt-4o")
+            self.default_image_generation_driver = OpenAiImageGenerationDriver(
+                model="dall-e-3"
+            )
+            self.default_embedding_driver = OpenAiEmbeddingDriver()
+            self.default_text_to_speech_driver = OpenAiTextToSpeechDriver(
+                model="tts-1", voice="alloy"
+            )
+            self.default_audio_transcription_driver = OpenAiAudioTranscriptionDriver(
+                model="whisper-1"
+            )
+
         pass
 
     @classmethod
@@ -75,17 +80,17 @@ def INPUT_TYPES(s):
     def create(self, **kwargs):
         self.run_envs(kwargs)
 
-        prompt_driver = kwargs.get("prompt_driver", default_chat_prompt_driver)
+        prompt_driver = kwargs.get("prompt_driver", self.default_chat_prompt_driver)
         image_generation_driver = kwargs.get(
-            "image_generation_driver", default_image_generation_driver
+            "image_generation_driver", self.default_image_generation_driver
         )
-        embedding_driver = kwargs.get("embedding_driver", default_embedding_driver)
+        embedding_driver = kwargs.get("embedding_driver", self.default_embedding_driver)
         vector_store_driver = kwargs.get("vector_store_driver", None)
         text_to_speech_driver = kwargs.get(
-            "text_to_speech_driver", default_text_to_speech_driver
+            "text_to_speech_driver", self.default_text_to_speech_driver
         )
         audio_transcription_driver = kwargs.get(
-            "audio_transcription_driver", default_audio_transcription_driver
+            "audio_transcription_driver", self.default_audio_transcription_driver
         )
 
         drivers = {}

nodes/tasks/gtUIAudioTranscriptionTask.py

@@ -19,9 +19,6 @@
 
 default_prompt = "{{ input_string }}"
 
-settings = GriptapeSettings()
-OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
-
 
 class gtUIAudioTranscriptionTask(gtUIBaseAudioTask):
     DESCRIPTION = "Transcribe an audio file."
@@ -66,8 +63,11 @@ def run(self, **kwargs):
             print(f"Error loading audio file: {e}")
         if audio_artifact:
             if not driver and not agent:
+                settings = GriptapeSettings()
+                OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
                 audio_transcription_driver = OpenAiAudioTranscriptionDriver(
-                    model="whisper-1"
+                    model="whisper-1",
+                    api_key=OPENAI_API_KEY,
                 )
             elif not driver and agent:
                 audio_transcription_driver = (

nodes/tasks/gtUIParallelImageQueryTask.py

@@ -9,17 +9,16 @@
     PromptTask,
 )
 
-from ...py.griptape_settings import GriptapeSettings
 from ..agent.gtComfyAgent import gtComfyAgent as Agent
 from ..utilities import (
     convert_tensor_to_base_64,
 )
 from .gtUIBaseImageTask import gtUIBaseImageTask
 
 default_prompt = "{{ input_string }}"
-settings = GriptapeSettings()
+# settings = GriptapeSettings()
 
-OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
+# OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
 
 
 def do_start_task(task: CodeExecutionTask) -> BaseArtifact:

nodes/tasks/gtUIPromptImageGenerationTask.py

@@ -21,9 +21,6 @@
 from .gtUIBaseTask import gtUIBaseTask
 
 default_prompt = "{{ input_string }}"
-settings = GriptapeSettings()
-
-OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
 
 
 class gtUIPromptImageGenerationTask(gtUIBaseTask):
@@ -63,6 +60,8 @@ def run(self, **kwargs):
             if isinstance(
                 agent.drivers_config.image_generation_driver, DummyImageGenerationDriver
             ):
+                settings = GriptapeSettings()
+                OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
                 # create a default driver
                 driver = OpenAiImageGenerationDriver(
                     model="dall-e-3",

nodes/tasks/gtUIPromptImageVariationTask.py

@@ -22,8 +22,6 @@
 from .gtUIBaseImageTask import gtUIBaseImageTask
 
 default_prompt = "{{ input_string }}"
-settings = GriptapeSettings()
-OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
 
 
 class gtUIPromptImageVariationTask(gtUIBaseImageTask):
@@ -58,6 +56,9 @@ def run(self, **kwargs):
 
         prompt_text = self.get_prompt_text(STRING, input_string)
         if not driver:
+            settings = GriptapeSettings()
+            OPENAI_API_KEY = settings.get_settings_key_or_use_env("OPENAI_API_KEY")
+
             driver = OpenAiImageGenerationDriver(
                 api_key=OPENAI_API_KEY,
                 model="dall-e-2",

nodes/utils/ollama_utils.py

@@ -22,7 +22,6 @@ def run_ollama_command(command):
     # Remove any newlines and extra spaces
     full_command_str = str(" ".join(full_command.split()))
 
-    # ic(full_command_str)
     try:
         result = subprocess.run(full_command_str, capture_output=True, text=True)
         output = result.stdout + result.stderr

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "comfyui-griptape"
-version = "1.0.16"
+version = "1.0.17"
 description = "Griptape LLM(Large Language Model) Nodes for ComfyUI."
 authors = ["Jason Schleifer <[email protected]>"]
 readme = "README.md"
@@ -26,7 +26,7 @@ priority = "explicit"
 [project]
 name = "comfyui-griptape"
 description = "Griptape LLM(Large Language Model) Nodes for ComfyUI."
-version = "1.0.16"
+version = "1.0.17"
 license = {file = "LICENSE"}
 dependencies = ["griptape[all]==0.34.0", "python-dotenv"]