Untitled

#%% import
import openai
from docx import Document
from docx.shared import Pt
from docx.enum.text import WD_PARAGRAPH_ALIGNMENT
from docx.oxml import OxmlElement

# Set your OpenAI GPT-3 API key
api_key = 'YOUR_API_KEY'
openai.api_key = api_key

#%% Function to read content from a DOCX file
def read_docx(file_path):
    doc = Document(file_path)
    return doc

# Function to proofread text using ChatGPT API
def proofread_text(text_to_proofread):
    prompt = f"Automatically proofread the following text keeping in mind grammar, clarity, flow, professional style, but also try to maintain the author's style as much as possible. Track changes throughout the document and generate a .docx file with tracked changes.\n\n{text_to_proofread}"

    # Make a request to the OpenAI API
    response = openai.Completion.create(
        engine="text-davinci-003",
        prompt=prompt,
        temperature=0.7,
        max_tokens=2000,  # Increase max_tokens as needed
        top_p=1.0,
        frequency_penalty=0.0,
        presence_penalty=0.0,
        stop=None,  # Add any specific stop words if needed
    )

    # Extract the proofread suggestions from the API response
    proofread_suggestions = response.choices[0].text.strip()

    return proofread_suggestions

#%% Function to apply tracked changes to a document
def apply_tracked_changes(original_doc, changes):
    for paragraph in original_doc.paragraphs:
        for run in paragraph.runs:
            # Apply changes to text
            if run.text.strip():
                run.clear()
                run.add_text(changes)
                run.font.color.theme_color = None  # Clear theme color to avoid potential formatting issues
                run.font.highlight_color = None

    # Apply changes to formatting
    for paragraph in original_doc.paragraphs:
        if changes:
            # Apply changes to formatting
            run = paragraph.runs[0]
            run.text = changes
            run.font.color.theme_color = None  # Clear theme color to avoid potential formatting issues
            run.font.highlight_color = None

            # Apply tracked changes
            run = paragraph.runs[0]
            if run.text.strip():
                run.clear()
                run.add_text(changes)
                run.font.color.theme_color = None  # Clear theme color to avoid potential formatting issues
                run.font.highlight_color = None
                run.element.getparent().remove(run.element)

    return original_doc

#%% Path to the DOCX file
docx_path = '9693-ao.docx'

# Read content from the DOCX file
original_doc = read_docx(docx_path)

# Split the content into smaller chunks
chunk_size = 1000
chunks = [original_doc.text[i:i + chunk_size] for i in range(0, len(original_doc.text), chunk_size)]

# Proofread each chunk and store the results
proofread_results = []
for chunk in chunks:
    proofread_result = proofread_text(chunk)
    proofread_results.append(proofread_result)

#%% Combine the results
final_proofread_result = "".join(proofread_results)

# Apply tracked changes to the original document
tracked_changes_doc = apply_tracked_changes(original_doc, final_proofread_result)

# Save the document with tracked changes
tracked_changes_doc.save('proofread_output.docx')

print("Proofread document with tracked changes saved to 'proofread_output.docx'.")