Fixed issue with Gemini API

Author: AJaySi

README.md

@@ -1,11 +1,28 @@
-# AI Content Generation Toolkit - Alwrity
-![](https://github.com/AJaySi/AI-Blog-Writer/blob/main/workspace/keyword_blog.gif)
+# How to Use AI Content Generation Toolkit - Alwrity
+
+1). Visit alwrity.com, Under tools section, you will find some AI content writing tools, which are Free & No-Signup.
+**Note:** Although, this is limited, as is our wallet & Resources.
+
+2). For complete AI content toolkit, alwrity offers a commandline App. Its a BYOK model(Bring Your Own Key).
+**Note:** (🗯️ commandline, byok and shit) ... Now, before you run away 🏃💨   
+
+If you have 💻 Laptop + 🛜 Internet + 20 minutes, you will be generating blogs, articles etc with just few words.
+
+**[Getting Started for Absolute Begginers]**(https://www.alwrity.com/post/getting-started-with-alwrity-ai-writer)
+
+Getting started for pretentious Developers : Continue Reading.....
+
+**If you still Get stuck, Open a issue here & say pretty please** :: https://github.com/AJaySi/AI-Writer/issues
+
 
 ## Introduction
 
 Alwrity automates and enhances the process of blog creation, optimization, and management. 
 Leveraging AI technologies, it assists content creators and digital marketers in generating, formatting, and uploading blog content efficiently. The toolkit integrates advanced AI models for text generation, image creation, and data analysis, streamlining the content creation pipeline.
 
+# AI Content Generation Toolkit - Alwrity
+![](https://github.com/AJaySi/AI-Blog-Writer/blob/main/workspace/keyword_blog.gif)
+
 ---
 
 ## Getting Started 🚀 🤞🤞🤞

alwrity.py

@@ -5,6 +5,7 @@
 
 import typer
 from prompt_toolkit.shortcuts import checkboxlist_dialog, message_dialog, input_dialog
+from prompt_toolkit.shortcuts import radiolist_dialog
 from prompt_toolkit import prompt
 from prompt_toolkit.styles import Style
 from prompt_toolkit.shortcuts import radiolist_dialog
@@ -20,7 +21,8 @@
 
 app = typer.Typer()
 
-from lib.utils.alwrity_utils import blog_from_audio, blog_from_keyword, do_web_research, do_web_research, write_story, essay_writer, blog_tools, competitor_analysis
+from lib.utils.alwrity_utils import blog_from_audio, blog_from_keyword, do_web_research, do_web_research
+from lib.utils.alwrity_utils import write_story, essay_writer, blog_tools, competitor_analysis, image_to_text_writer, image_generator
 
 
 def prompt_for_time_range():
@@ -61,10 +63,11 @@ def start_interactive_mode():
         ("AI Blog Writer", "AI Blog Writer"),
         ("AI Story Writer", "AI Story Writer"),
         ("AI Essay Writer", "AI Essay Writer"),
+        ("AI Image to Text Writer", "AI Image to Text Writer"),
         ("Online Blog Tools/Apps", "Online Blog Tools/Apps"),
         ("Do keyword Research", "Do keyword Research"),
         ("Competitor Analysis", "Competitor Analysis"),
-        ("Create Blog Images(TBD)", "Create Blog Images(TBD)"),
+        ("Create Blog Images", "Create Blog Images"),
         ("AI Social Media(TBD)", "AI Social Media(TBD)"),
         ("Quit", "Quit")
     ]
@@ -76,10 +79,12 @@ def start_interactive_mode():
             write_story()
         elif mode == 'AI Essay Writer':
             essay_writer()
+        elif mode == 'AI Image to Text Writer':
+            image_to_text_writer()
         elif mode == 'Do keyword Research':
             do_web_research()
-        elif mode == 'Create Blog Images(TBD)':
-            faq_generator()
+        elif mode == 'Create Blog Images':
+            image_generator()
         elif mode == 'Competitor Analysis':
             competitor_analysis()
         elif mode == 'Online Blog Tools/Apps':
@@ -105,8 +110,7 @@ def check_search_apis():
     """
 
     # Use rich.print for styling and hyperlinking
-    print("\n\n🙋♂️  🙋♂️   Before doing web research, ensure the following API keys are available:")
-    print("Blogen uses Basic, Semantic, Neural web search using above APIs for contextual blog generation.\n")
+    print("Alwrity uses Basic, Semantic, Neural web search using above APIs for contextual blog generation.\n")
 
     api_keys = {
         "METAPHOR_API_KEY": "Metaphor AI Key (Get it here: [link=https://dashboard.exa.ai/login]Metaphor API[/link])",
@@ -121,7 +125,6 @@ def check_search_apis():
             if os.getenv(key) is None:
                 # Use rich.print for styling and hyperlinking
                 print(f"[bold red]✖ 🚫 {key} is missing:[/bold red] [blue underline]Get {key} API Key[/blue underline]")
-                typer.echo(f"[bold red]✖ 🚫 {key} is missing:[/bold red] [link={key}]Get {key} API Key[/link]")
                 missing_keys.append((key, description))
 
     if missing_keys:
@@ -142,10 +145,12 @@ def get_api_key(api_key: str, api_description: str):
         api_key (str): The name of the API key variable.
         api_description (str): The description of the API key.
     """
-    user_input = typer.prompt(f"\n\n🙆💩💩🙆 - Please enter {api_key} API Key:")
+    print("\n\n")
+    print(f"[bold green] 🙋 Attention Here: 🙋 -- {api_description}")
+    user_input = typer.prompt(f"💩 -**Please Enter(copy/paste) {api_key} API Key** - Here🙋:")
     with open(".env", "a") as env_file:
         env_file.write(f"{api_key}={user_input}\n")
-    print(f"✅ {api_description} API Key added to .env file.")
+        print(f"✅ API Key added to .env file.")
 
 
 def write_blog():
@@ -172,22 +177,20 @@ def check_llm_environs():
     """ Function to check which LLM api is given. """
     # Load .env file
     load_dotenv(Path('.env'))
-    # Check if GPT_PROVIDER is defined in .env file
     gpt_provider = os.getenv("GPT_PROVIDER")
 
     # Disable unsupported GPT providers
     supported_providers = ['google', 'openai', 'mistralai']
     if gpt_provider is None or gpt_provider.lower() not in map(str.lower, supported_providers):
         # Prompt user to select a provider
-        selected_provider = radiolistbox(
-            msg='Select your preferred GPT provider:',
-            title='GPT Provider Selection',
-            choices=["Google", "OpenAI", "MistralAI/WIP", "Ollama (TBD)"],
-            default_choice="Google"
-        )
-        gpt_provider = selected_provider
+        gpt_provider = radiolist_dialog(
+            title="Select your GPT Provider(llm) from 'google', 'openai', 'mistralai'",
+            values=[("google", "Google Gemini Pro"), ("openai", "OpenAI- ChatGPT"), ("mistralai", "MistralAI/WIP")]).run()
         # Update .env file
         os.environ["GPT_PROVIDER"] = gpt_provider
+        with open(".env", "a") as env_file:
+            env_file.write(f"GPT_PROVIDER=gpt_provider\n")
+            print(f"✅ API Key added to .env file.")
 
     if gpt_provider.lower() == "google":
         api_key_var = "GEMINI_API_KEY"
@@ -198,9 +201,9 @@ def check_llm_environs():
     elif gpt_provider.lower() == "mistralai":
         api_key_var = "MISTRAL_API_KEY"
         missing_api_msg = "To get your MistralAI API key, please visit: https://mistralai.com/api"
-    else:
-        print("Unrecognised/Unsupported GPT provider. Check your main_config and environs.")
-        exit(1)
+
+    if api_key_var not in os.environ:
+        get_api_key(api_key_var, missing_api_msg)
 
 
 def check_internet():
@@ -233,13 +236,10 @@ def create_env_file():
 
 
 if __name__ == "__main__":
-    print("Checking Internet, lets get the basics right.")
     check_internet()
-    print("Create .env file, if not Present working directory")
     create_env_file()
-    print("Check Metaphor, Tavily, YOU.com Search API keys.")
+    os.system("clear" if os.name == "posix" else "cls")
     check_search_apis()
-    print("Check LLM details & AI Model to use.")
     check_llm_environs()
     os.environ["SEARCH_SAVE_FILE"] = os.path.join(os.getcwd(), "workspace",
             "web_research_report" + "_" + datetime.now().strftime("%Y-%m-%d_%H-%M-%S"))

lib/gpt_providers/image_to_text_gen/gemini_image_details.py

@@ -132,7 +132,7 @@ def ai_essay_generator(essay_title, selected_essay_type, selected_education_leve
         load_dotenv(Path('../.env'))
         genai.configure(api_key=os.getenv('GEMINI_API_KEY'))
         # Initialize the generative model
-        model = genai.GenerativeModel('gemini-1.0-pro')
+        model = genai.GenerativeModel('gemini-pro')
 
         # Generate prompts
         try:

lib/image_to_text/gpt_vision_image_details.py lib/gpt_providers/image_to_text_gen/openai_vision_image_details.py

@@ -19,7 +19,7 @@
 )
 
 
-#@retry(wait=wait_random_exponential(min=1, max=60), stop=stop_after_attempt(6))
+@retry(wait=wait_random_exponential(min=1, max=60), stop=stop_after_attempt(6))
 def gemini_text_response(prompt, temperature, top_p, n, max_tokens):
     """ Common functiont to get response from gemini pro Text. """
     #FIXME: Include : https://github.com/google-gemini/cookbook/blob/main/quickstarts/rest/System_instructions_REST.ipynb

lib/gpt_providers/text_generation/ai_essay_writer.py

@@ -0,0 +1,41 @@
+from PIL import Image
+import requests
+
+# Ensure you sign up for an account to obtain an API key:
+# https://platform.stability.ai/
+# Your API key can be found here after account creation:
+# https://platform.stability.ai/account/keys
+
+
+def generate_stable_diffusion_image(prompt):
+    """
+    Generate images using Stable Diffusion API based on a given prompt.
+
+    Args:
+        prompt (str): The prompt to generate the image.
+        image_dir (str): The directory where the image will be saved.
+
+    Raises:
+        Warning: If the adult content classifier is triggered.
+        Exception: For any issues during image generation or saving.
+    """
+    api_key = os.getenv('STABILITY_API_KEY')
+
+    response = requests.post(
+        f"https://api.stability.ai/v2beta/stable-image/generate/sd3",
+        headers={
+            "authorization": f"Bearer {api_key}",
+            "accept": "image/*"
+        },
+        files={"none": ''},
+        data={
+            "prompt": prompt,
+            "output_format": "webp",
+        },
+    )
+
+    if response.status_code == 200:
+        with open("./dog-wearing-glasses.jpeg", 'wb') as file:
+            file.write(response.content)
+    else:
+        raise Exception(str(response.json()))

lib/gpt_providers/text_generation/gemini_pro_text.py

@@ -20,11 +20,12 @@
         format="<level>{level}</level>|<green>{file}:{line}:{function}</green>| {message}"
     )
 
-from .gpt_providers.openai_gpt_provider import generate_dalle2_images, generate_dalle3_images, openai_chatgpt
-from .stabl_diff_img2html import generate_stable_diffusion_image
+#from .gen_dali2_images
+from .gen_dali3_images import generate_dalle3_images
+from .gen_stabl_diff_img import generate_stable_diffusion_image
 
 
-def generate_image(user_prompt, image_dir, image_engine="dalle3"):
+def generate_image(user_prompt, image_engine="dalle3"):
     """
     The generation API endpoint creates an image based on a text prompt.
 
@@ -40,18 +41,14 @@ def generate_image(user_prompt, image_dir, image_engine="dalle3"):
     Must be one of "url" or "b64_json". Defaults to "url".
     --> user (str): A unique identifier representing your end-user, which will help OpenAI to monitor and detect abuse.
     """
-    logger.info(f"Generated blog images will be stored at: {image_dir=}")
-
     img_prompt = generate_img_prompt(user_prompt) 
     # call the OpenAI API to generate image from prompt.
-    logger.info(f"Calling openai.image.generate with prompt: {img_prompt}")
+    logger.info(f"Calling image.generate with prompt: {img_prompt}")
 
-    if 'dalle2' in image_engine:
-        image_stored_at = generate_dalle2_images(img_prompt, image_dir)
-    elif 'dalle3' in image_engine:
-        image_stored_at = generate_dalle3_images(img_prompt, image_dir)
-    elif 'stable_diffusion' in image_engine:
-        image_stored_at = generate_stable_diffusion_image(img_prompt, image_dir)
+    if 'Dalle3' in image_engine:
+        image_stored_at = generate_dalle3_images(img_prompt)
+    elif 'Stable Diffusion' in image_engine:
+        image_stored_at = generate_stable_diffusion_image(img_prompt)
 
     return image_stored_at
 
@@ -72,5 +69,5 @@ def generate_img_prompt(user_prompt):
             Advice for creating prompt for image from the given text(no more than 150 words).
             Reply with only one answer and no descrition. Generate image prompt for the below text.
             Text: {user_prompt}"""
-    response = openai_chatgpt(prompt)
+    response = (prompt)
     return response

lib/gpt_providers/image_generation/gen_dali2_images.py lib/gpt_providers/text_to_image_generation/gen_dali2_images.py

@@ -5,6 +5,8 @@
 
 from prompt_toolkit.shortcuts import checkboxlist_dialog, message_dialog, input_dialog
 from prompt_toolkit import prompt
+from prompt_toolkit.completion import WordCompleter
+from prompt_toolkit.validation import Validator, ValidationError
 from prompt_toolkit.shortcuts import radiolist_dialog
 
 from lib.ai_web_researcher.gpt_online_researcher import gpt_web_researcher
@@ -13,6 +15,7 @@
 from lib.ai_writers.speech_to_blog.main_audio_to_blog import generate_audio_blog
 from lib.gpt_providers.text_generation.ai_story_writer import ai_story_generator
 from lib.gpt_providers.text_generation.ai_essay_writer import ai_essay_generator
+from lib.gpt_providers.text_to_image_generation.generate_image_from_prompt import generate_image
 
 
 def blog_from_audio():
@@ -258,6 +261,95 @@ def blog_tools():
             ).run()
 
 
+def image_generator():
+    """ Generate image from given text """
+    print("Enter your long string below---")
+    img_prompt = prompt("Enter text to create image from:: ")
+
+    img_models = WordCompleter(['Stability-Stable-Diffusion', 'Dalle2', 'Dalle3'], ignore_case=True)
+    print("Choose between:: Stable-Diffusion, Dalle2, Dalle3")
+    img_model = prompt('Choose the image model to use for generation: ', completer=img_models, validator=ModelTypeValidator())
+
+    print(f"{img_prompt}----{img_model}")
+    try:
+        generate_image(img_prompt, img_model)
+    except Exception as err:
+        print(f"Failed to generate image: {err}")
+
+
+
+class ModelTypeValidator(Validator):
+    def validate(self, document):
+        if document.text.lower() not in ['stability-stable-diffusion', 'dalle2', 'dalle3']:
+            raise ValidationError(message='Please choose a valid Text to image model.')
+
+
+def image_to_text_writer():
+    """ IMage to Text Content Generation"""
+    os.system("clear" if os.name == "posix" else "cls")
+    text = "_______________________________________________________________________\n"
+    text += "\n⚠️    Alert!   💥❓💥\n"
+    text += "Provide Inputs Below to Continue..\n"
+    text += "_______________________________________________________________________\n\n"
+    print(text)
+
+    print("Make sure the file path is correct and the file is one of the following image types: PNG, JPEG, WEBP, HEIC, HEIF.\n")
+    
+    file_location = prompt('⚠️  Enter the image file location: ', validator=FileTypeValidator())
+    if file_location:
+        writing_completer = WordCompleter(['Blog', 'Food Recipe', 'Alt Text', 'Marketing Copy'], ignore_case=True)
+        print("Choose between 'Blog', 'Food Recipe', 'Alt Text', 'Marketing Copy'")
+        writing_type = prompt('Select the type of writing: ', completer=writing_completer, validator=WritingTypeValidator())
+    
+    prompt_gemini = None
+    if writing_type.lower() == 'blog':
+        prompt_gemini = "Given an image of a product and its target audience, write an engaging marketing description",
+    elif writing_type.lower() == 'food recipe':
+        prompt_gemini = """I have the ingredients above. Not sure what to cook for lunch.
+        Show me a list of foods with the recipes.
+        Accurately identify the baked good in the image and provide an appropriate and recipe consistent with your analysis.
+        Write a short, engaging blog post based on this picture.
+        It should include a description of the meal in the photo and talk about my journey meal prepping.
+        """
+    elif writing_type.lower() == 'alt text':
+        prompt_gemini = """Given an image from my blog, generate 3 different ALT texts.
+        The image alt text should be of maximum 2 lines. It should be descriptive and SEO optimised."""
+    elif writing_type.lower() == 'marketing copy':
+        prompt_gemini = "Given an image of a product and its target audience, write an engaging marketing description"
+
+    print("TBD/FIXME: Will be taken up soon..")
+
+
+class WritingTypeValidator(Validator):
+    def validate(self, document):
+        writing_type = document.text.strip().lower()
+        if writing_type not in ['blog', 'food recipe', 'alt text', 'marketing copy']:
+            raise ValidationError(message="Please select a valid writing type: Blog, Food Recipe, Alt Text, or Marketing Copy.")
+
+
+class FileTypeValidator(Validator):
+    def validate(self, document):
+        file_path = document.text.strip()
+        if not os.path.exists(file_path):
+            raise ValidationError(message="File does not exist.")
+        elif not self.is_valid_file_type(file_path):
+            raise ValidationError(message="Unsupported file type or MIME type. Please select an image file.")
+    
+    def is_valid_file_type(self, file_path):
+        # Define supported MIME types for image files
+        supported_types = ['image/png', 'image/jpeg', 'image/webp', 'image/heic', 'image/heif']
+        file_mime_type = self.get_file_mime_type(file_path)
+        return file_mime_type in supported_types
+    
+    def get_file_mime_type(self, file_path):
+        # Placeholder function to get the MIME type of the file
+        # You can use libraries like magic or mimetypes for this purpose
+        # Example:
+        # import magic
+        # mime = magic.Magic(mime=True)
+        # return mime.from_file(file_path)
+        return 'image/png'  # Placeholder value for demonstration
+
 
 def competitor_analysis():
     text = "_______________________________________________________________________\n"