ag2 0.9.1a1__py3-none-any.whl → 0.9.2__py3-none-any.whl

autogen/agentchat/contrib/captainagent/tools/information_retrieval/extract_pdf_image.py

@@ -0,0 +1,53 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+import os
+
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["PyMuPDF"], ["os"])
+def extract_pdf_image(pdf_path: str, output_dir: str, page_number=None):
+    """Extracts images from a PDF file and saves them to the specified output directory.
+
+    Args:
+        pdf_path (str): The path to the PDF file.
+        output_dir (str): The directory to save the extracted images.
+        page_number (int, optional): The page number to extract images from. If not provided, extract images from all pages.
+    """
+    import fitz  # PyMuPDF library
+
+    # Open the PDF file
+    doc = fitz.open(pdf_path)
+
+    # Create the output directory if it doesn't exist
+    os.makedirs(output_dir, exist_ok=True)
+
+    # Extract images from the PDF file
+    images = []
+    if page_number is not None:
+        page = doc[page_number - 1]  # Adjust page number to 0-based index
+        for img in page.get_images():
+            xref = img[0]
+            base_image = doc.extract_image(xref)
+            image_bytes = base_image["image"]
+            images.append(image_bytes)
+    else:
+        for page in doc:
+            for img in page.get_images():
+                xref = img[0]
+                base_image = doc.extract_image(xref)
+                image_bytes = base_image["image"]
+                images.append(image_bytes)
+
+    # Save the extracted images
+    for i, image_bytes in enumerate(images):
+        image_path = os.path.join(output_dir, f"image_{i}.png")
+        with open(image_path, "wb") as f:
+            f.write(image_bytes)
+
+    # Print the total number of images saved
+    print(f"Saved a total of {len(images)} images")
+
+    # Close the PDF file
+    doc.close()

autogen/agentchat/contrib/captainagent/tools/information_retrieval/extract_pdf_text.py

@@ -0,0 +1,38 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["PyMuPDF"])
+def extract_pdf_text(pdf_path, page_number=None):
+    """Extracts text from a specified page or the entire PDF file.
+
+    Args:
+        pdf_path (str): The path to the PDF file.
+        page_number (int, optional): The page number to extract (starting from 0). If not provided,
+            the function will extract text from the entire PDF file.
+
+    Returns:
+        str: The extracted text.
+    """
+    import fitz
+
+    # Open the PDF file
+    doc = fitz.open(pdf_path)
+
+    # Extract text from the entire PDF file or a specific page
+    text = ""
+    if page_number is None:
+        # Extract content from the entire PDF file
+        for page in doc:
+            text += page.get_text()
+    else:
+        # Extract content from a specific page
+        page = doc[page_number]
+        text = page.get_text()
+
+    # Close the PDF file
+    doc.close()
+
+    return text

autogen/agentchat/contrib/captainagent/tools/information_retrieval/get_wikipedia_text.py

@@ -0,0 +1,21 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def get_wikipedia_text(title):
+    """Retrieves the text content of a Wikipedia page. It does not support tables and other complex formatting.
+
+    Args:
+        title (str): The title of the Wikipedia page.
+
+    Returns:
+        str or None: The text content of the Wikipedia page if it exists, None otherwise.
+    """
+    import wikipediaapi
+
+    wiki_wiki = wikipediaapi.Wikipedia("Mozilla/5.0 (merlin@example.com)", "en")
+    page = wiki_wiki.page(title)
+
+    if page.exists():
+        return page.text
+    else:
+        return None

autogen/agentchat/contrib/captainagent/tools/information_retrieval/get_youtube_caption.py

@@ -0,0 +1,34 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+# alternative api: https://rapidapi.com/omarmhaimdat/api/youtube-v2
+
+
+def get_youtube_caption(video_id: str) -> str:
+    """Retrieves the captions for a YouTube video.
+
+    Args:
+        video_id (str): The ID of the YouTube video.
+
+    Returns:
+        str: The captions of the YouTube video in text format.
+
+    Raises:
+        KeyError: If the RAPID_API_KEY environment variable is not set.
+    """
+    import os
+
+    import requests
+
+    rapid_api_key = os.environ["RAPID_API_KEY"]
+    video_url = f"https://www.youtube.com/watch?v={video_id}"
+    url = "https://youtube-transcript3.p.rapidapi.com/api/transcript-with-url"
+
+    querystring = {"url": video_url, "lang": "en", "flat_text": "true"}
+
+    headers = {"X-RapidAPI-Key": rapid_api_key, "X-RapidAPI-Host": "youtube-transcript3.p.rapidapi.com"}
+
+    response = requests.get(url, headers=headers, params=querystring)
+    response = response.json()
+    print(response)
+    return response["transcript"]

autogen/agentchat/contrib/captainagent/tools/information_retrieval/image_qa.py

@@ -0,0 +1,60 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+import os
+
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["transformers", "torch", "pillow"], ["transformers", "torch", "os"])
+def image_qa(image, question, ckpt="Salesforce/blip-vqa-base"):
+    """Perform question answering on an image using a pre-trained VQA model.
+
+    Args:
+        image (Union[str, Image.Image]): The image to perform question answering on. It can be either file path to the image or a PIL Image object.
+        question: The question to ask about the image.
+        ckpt: The checkpoint name to use. Default is "Salesforce/blip-vqa-base".
+
+    Returns:
+        dict: The generated answer text.
+    """
+    import torch
+    from PIL import Image
+    from transformers import BlipForQuestionAnswering, BlipProcessor
+
+    def image_processing(img):
+        if isinstance(img, Image.Image):
+            return img.convert("RGB")
+        elif isinstance(img, str):
+            if os.path.exists(img):
+                return Image.open(img).convert("RGB")
+            else:
+                full_path = img
+                if os.path.exists(full_path):
+                    return Image.open(full_path).convert("RGB")
+                else:
+                    raise FileNotFoundError
+
+    def text_processing(file_path):
+        # Check the file extension
+        if file_path.endswith(".txt"):
+            with open(file_path) as file:
+                content = file.read()
+        else:
+            # if the file is not .txt, then it is a string, directly return the string
+            return file_path
+        return content
+
+    image = image_processing(image)
+    question = text_processing(question)
+
+    processor = BlipProcessor.from_pretrained(ckpt)
+    model = BlipForQuestionAnswering.from_pretrained(ckpt, torch_dtype=torch.float16).to("cuda")
+
+    raw_image = image
+
+    inputs = processor(raw_image, question, return_tensors="pt").to("cuda", torch.float16)
+    out = model.generate(**inputs)
+    result_formatted = processor.decode(out[0], skip_special_tokens=True)
+
+    return result_formatted

autogen/agentchat/contrib/captainagent/tools/information_retrieval/optical_character_recognition.py

@@ -0,0 +1,61 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+import os
+
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["easyocr"], ["os"])
+def optical_character_recognition(image):
+    """Perform optical character recognition (OCR) on the given image.
+
+    Args:
+        image (Union[str, Image.Image]): The image to perform OCR on. It can be either a file path or an Image object.
+
+    Returns:
+        str: The extracted text from the image.
+
+    Raises:
+        FileNotFoundError: If the image file path does not exist.
+    """
+    import io
+
+    import easyocr
+    from PIL import Image
+
+    def image_processing(img):
+        if isinstance(img, Image.Image):
+            return img.convert("RGB")
+        elif isinstance(img, str):
+            if os.path.exists(img):
+                return Image.open(img).convert("RGB")
+            else:
+                full_path = img
+                if os.path.exists(full_path):
+                    return Image.open(full_path).convert("RGB")
+                else:
+                    raise FileNotFoundError
+
+    reader = easyocr.Reader(["en"])  # Load the OCR model into memory
+
+    if isinstance(image, str):
+        # If image is a path, use it directly
+        if not os.path.exists(image):
+            raise FileNotFoundError
+        image_path_or_bytes = image
+    else:
+        # If image is an Image object, convert it to a bytes stream
+        buffer = io.BytesIO()
+        image = image_processing(image)  # Process the image if needed
+        image.save(buffer, format="JPEG")
+        buffer.seek(0)
+        image_path_or_bytes = buffer
+
+    # Read text from the image or image path
+    result = reader.readtext(image_path_or_bytes)
+
+    # Extract only the text from the result
+    result_text = [text for _, text, _ in result]
+
+    return ", ".join(result_text)

autogen/agentchat/contrib/captainagent/tools/information_retrieval/perform_web_search.py

@@ -0,0 +1,47 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def perform_web_search(query, count=10, offset=0):
+    """Perform a web search using Bing API.
+
+    Args:
+        query (str): The search query.
+        count (int, optional): Number of search results to retrieve. Defaults to 10.
+        offset (int, optional): Offset of the first search result. Defaults to 0.
+
+    Returns:
+        The name, URL and snippet of each search result.
+    """
+    import os
+
+    import requests
+
+    # Get the Bing API key from the environment variable
+    bing_api_key = os.getenv("BING_API_KEY")
+
+    # Check if the API key is available
+    if not bing_api_key:
+        raise ValueError("Bing API key not found in environment variable")
+
+    # Set up the API request
+    url = "https://api.bing.microsoft.com/v7.0/search"
+    headers = {
+        "Ocp-Apim-Subscription-Key": bing_api_key,
+    }
+    params = {
+        "q": query,
+        "count": count,  # Number of search results to retrieve
+        "offset": offset,  # Offset of the first search result
+    }
+
+    # Send the API request
+    response = requests.get(url, headers=headers, params=params)
+    response.raise_for_status()
+
+    # Process the search results
+    search_results = response.json()
+    for index, result in enumerate(search_results["webPages"]["value"]):
+        print(f"Search Result {index + 1}:")
+        print(result["name"])
+        print(result["url"])
+        print(result["snippet"])

autogen/agentchat/contrib/captainagent/tools/information_retrieval/scrape_wikipedia_tables.py

@@ -0,0 +1,33 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def scrape_wikipedia_tables(url: str, header_keyword: str):
+    """Scrapes Wikipedia tables based on a given URL and header keyword.
+
+    Args:
+        url: The URL of the Wikipedia page to scrape.
+        header_keyword: The keyword to search for in the headers of the page.
+
+    Returns:
+        list: A list of lists representing the scraped table data. Each inner list represents a row in the table,
+              with each element representing a cell value.
+    """
+    import requests
+    from bs4 import BeautifulSoup
+
+    response = requests.get(url)
+    response.raise_for_status()
+    soup = BeautifulSoup(response.content, "html.parser")
+    headers = soup.find_all(["h1", "h2", "h3", "h4", "h5", "h6"])
+    data = []
+    for header in headers:
+        if header_keyword.lower() in header.text.lower():
+            table = header.find_next_sibling("table", class_="wikitable")
+            if table:
+                rows = table.find_all("tr")
+                for row in rows:
+                    cols = row.find_all(["th", "td"])
+                    cols = [ele.text.strip() for ele in cols]
+                    data.append([ele for ele in cols if ele])
+                break
+    return data

autogen/agentchat/contrib/captainagent/tools/information_retrieval/transcribe_audio_file.py

@@ -0,0 +1,21 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["openai-whisper"])
+def transcribe_audio_file(file_path):
+    """Transcribes the audio file located at the given file path.
+
+    Args:
+        file_path (str): The path to the audio file.
+
+    Returns:
+        str: The transcribed text from the audio file.
+    """
+    import whisper
+
+    model = whisper.load_model("base")
+    result = model.transcribe(file_path)
+    return result["text"]

autogen/agentchat/contrib/captainagent/tools/information_retrieval/youtube_download.py

@@ -0,0 +1,35 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def youtube_download(url: str):
+    """Downloads a YouTube video and returns the download link.
+
+    Args:
+        url: The URL of the YouTube video.
+
+    Returns:
+        str: The download link for the audio.
+    """
+    import os
+
+    import requests
+
+    endpoint = "https://youtube-mp3-downloader2.p.rapidapi.com/ytmp3/ytmp3/"
+
+    querystring = {"url": url}
+
+    headers = {
+        "X-RapidAPI-Key": os.environ.get("RAPIDAPI_KEY"),
+        "X-RapidAPI-Host": "youtube-mp3-downloader2.p.rapidapi.com",
+    }
+
+    response = requests.get(endpoint, headers=headers, params=querystring)
+    response = response.json()
+
+    if "link" in response:
+        return response["link"]
+    else:
+        print("Error: Unable to retrieve download link.")
+        print(response)
+        # or you can return an error message
+        # return "Error: Unable to retrieve download link."

autogen/agentchat/contrib/captainagent/tools/math/calculate_circle_area_from_diameter.py

@@ -0,0 +1,21 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["sympy"])
+def calculate_circle_area_from_diameter(diameter):
+    """Calculate the area of a circle given its diameter.
+
+    Args:
+    diameter (float): The diameter of the circle.
+
+    Returns:
+    float: The area of the circle.
+    """
+    from sympy import pi
+
+    radius = diameter / 2
+    area = pi * radius**2
+    return area

autogen/agentchat/contrib/captainagent/tools/math/calculate_day_of_the_week.py

@@ -0,0 +1,18 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def calculate_day_of_the_week(total_days: int, starting_day: str):
+    """Calculates the day of the week after a given number of days starting from a specified day.
+
+    Args:
+        total_days: The number of days to calculate.
+        starting_day: The starting day of the week, should be one of 'Monday', 'Tuesday', 'Wednesday', etc.
+
+    Returns:
+        str: The day of the week after the specified number of days.
+    """
+    days_of_week = ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday"]
+
+    start_index = days_of_week.index(starting_day)
+    end_index = (start_index + total_days) % 7
+    return days_of_week[end_index]

autogen/agentchat/contrib/captainagent/tools/math/calculate_fraction_sum.py

@@ -0,0 +1,28 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def calculate_fraction_sum(
+    fraction1_numerator: int, fraction1_denominator: int, fraction2_numerator: int, fraction2_denominator: int
+):
+    """Calculates the sum of two fractions and returns the result as a mixed number.
+
+    Args:
+        fraction1_numerator: The numerator of the first fraction.
+        fraction1_denominator: The denominator of the first fraction.
+        fraction2_numerator: The numerator of the second fraction.
+        fraction2_denominator: The denominator of the second fraction.
+
+    Returns:
+        str: The sum of the two fractions as a mixed number in the format 'a b/c'
+    """
+    from fractions import Fraction
+
+    fraction1 = Fraction(fraction1_numerator, fraction1_denominator)
+    fraction2 = Fraction(fraction2_numerator, fraction2_denominator)
+    result = fraction1 + fraction2
+    mixed_number = result.numerator // result.denominator
+    mixed_fraction_numerator = result.numerator % result.denominator
+    if mixed_fraction_numerator > 0:
+        return f"{mixed_number} {Fraction(mixed_fraction_numerator, result.denominator)}"
+    else:
+        return str(mixed_number)

autogen/agentchat/contrib/captainagent/tools/math/calculate_matrix_power.py

@@ -0,0 +1,31 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["sympy"])
+def calculate_matrix_power(matrix, power):
+    """Calculate the power of a given matrix.
+
+    Args:
+        matrix (list): An array of numbers that represents the matrix.
+        power (int): The power to which the matrix is raised.
+
+    Returns:
+        Matrix: The resulting matrix after raising to power.
+
+    Raises:
+        ValueError: If the power is negative and the matrix is not invertible.
+    """
+    from sympy import Matrix, eye
+
+    m = Matrix(matrix)
+    if power == 0:
+        return eye(m.shape[0])
+    elif power < 0:
+        if not m.is_invertible():
+            raise ValueError("Matrix is not invertible.")
+        return m.inverse() ** (-power)
+    elif power > 0:
+        return m**power

autogen/agentchat/contrib/captainagent/tools/math/calculate_reflected_point.py

@@ -0,0 +1,16 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def calculate_reflected_point(point):
+    """Calculates the reflection point of a given point about the line y=x.
+
+    Args:
+        point (dict): A dictionary representing the coordinates of the point.
+            The dictionary should have keys 'x' and 'y' representing the x and y coordinates respectively.
+
+    Returns:
+        dict: A dictionary representing the coordinates of the reflected point. Its keys are 'x' and 'y'.
+    """
+    # Swap x and y for reflection about y=x
+    reflected_point = {"x": point["y"], "y": point["x"]}
+    return reflected_point

autogen/agentchat/contrib/captainagent/tools/math/complex_numbers_product.py

@@ -0,0 +1,25 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["sympy"])
+def complex_numbers_product(complex_numbers):
+    """Calculates the product of a list of complex numbers.
+
+    Args:
+        complex_numbers (list): A list of dictionaries representing complex numbers.
+            Each dictionary should have 'real' and 'imag' keys representing the real
+            and imaginary parts of the complex number.
+
+    Returns:
+        complex: The simplified product of the complex numbers.
+
+    """
+    from sympy import I, simplify
+
+    result = 1
+    for c in complex_numbers:
+        result *= c["real"] + I * c["imag"]
+    return simplify(result)

autogen/agentchat/contrib/captainagent/tools/math/compute_currency_conversion.py

@@ -0,0 +1,23 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+from autogen.coding.func_with_reqs import with_requirements
+
+
+@with_requirements(["sympy"])
+def compute_currency_conversion(amount, exchange_rate):
+    """Compute the currency conversion of the given amount using the provided exchange rate.
+
+    Args:
+    amount (float): The amount to be converted.
+    exchange_rate (float): The exchange rate to use for the conversion, represented as the amount of second currency equivalent to one unit of the first currency.
+
+    Returns:
+    float: The converted amount.
+
+    """
+    from sympy import Rational
+
+    # Calculate the converted amount using the given exchange rate
+    converted_amount = Rational(amount, exchange_rate)
+    return float(converted_amount)

autogen/agentchat/contrib/captainagent/tools/math/count_distinct_permutations.py

@@ -0,0 +1,27 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def count_distinct_permutations(sequence):
+    """Counts the number of distinct permutations of a sequence where items may be indistinguishable.
+
+    Args:
+        sequence (iterable): The sequence for which to count the distinct permutations.
+
+    Returns:
+        int: The number of distinct permutations.
+
+    Example:
+        >>> count_distinct_permutations("aab")
+        3
+        >>> count_distinct_permutations([1, 2, 2])
+        3
+    """
+    from collections import Counter
+    from math import factorial
+
+    counts = Counter(sequence)
+    total_length = sum(counts.values())
+    permutations = factorial(total_length)
+    for count in counts.values():
+        permutations //= factorial(count)
+    return permutations

autogen/agentchat/contrib/captainagent/tools/math/evaluate_expression.py

@@ -0,0 +1,28 @@
+# Copyright (c) 2023 - 2025, AG2ai, Inc., AG2ai open-source projects maintainers and core contributors
+#
+# SPDX-License-Identifier: Apache-2.0
+def evaluate_expression(expression):
+    """Evaluates a mathematical expression with support for floor function notation and power notation.
+
+    Args:
+        expression (str): The mathematical expression to evaluate. It can only contain one symbol 'x'.
+
+    Returns:
+        Union[sympy.Expr, str]: The evaluated result as a sympy expression if successful,
+        otherwise an error message as a string.
+
+    """
+    from sympy import symbols, sympify
+
+    # Replace power with ** for sympy
+    expression = expression.replace("^", "**")
+    # Replace the floor function notation
+    expression = expression.replace("\\lfloor", "floor(").replace("\\rfloor", ")")
+    try:
+        # Create a symbol 'x' for use in case it is in the expression
+        symbols("x")
+        # Evaluate the expression
+        result = sympify(expression)
+        return result
+    except Exception as e:
+        return str(e)