app.py

import gradio as gr
from docx import Document
import re

def clean_markdown_heading(text):
    return re.sub(r"^#+\s*", "", text).strip()

def get_color(tag):
    return {
        "docx": "#1f77b4",
        "markdown": "#2ca02c",
        "segment": "#ff7f0e",
        "colon": "#9467bd",
        "stage": "#d62728"
    }.get(tag, "black")

def process_doc(file):
    removed_headings = []
    cleaned_paragraphs = []
    if file is not None:
        doc = Document(file.name)
        for para in doc.paragraphs:
            text = para.text.strip()
            if not text:
                continue
            style_name = para.style.name.lower()
            is_docx_heading = "heading" in style_name
            is_markdown_heading = re.match(r"^#+\s+", text)
            is_segment_heading = re.match(r"(segment\s*\d+)", text.lower())
            is_colon_title = text.endswith(":") and len(text.split()) <= 6
            is_stage_direction = re.match(r"^\*\[.*\]\*$", text)
            if (
                is_docx_heading
                or is_markdown_heading
                or is_segment_heading
                or is_colon_title
                or is_stage_direction
            ):
                if is_markdown_heading:
                    clean_text = clean_markdown_heading(text)
                    removed_headings.append((clean_text, "markdown"))
                elif is_stage_direction:
                    clean_text = re.sub(r"^\*\[|\]\*$", "", text).strip()
                    removed_headings.append((clean_text, "stage"))
                elif is_docx_heading:
                    removed_headings.append((text, "docx"))
                elif is_segment_heading:
                    removed_headings.append((text, "segment"))
                elif is_colon_title:
                    removed_headings.append((text, "colon"))
            else:
                cleaned_paragraphs.append(text)
    cleaned_text = "\n\n".join(cleaned_paragraphs)
    if removed_headings:
        headings_output = "<br><br>".join(
            f"<span style='color:{get_color(tag)}; font-weight:600'>{text}</span>"
            for text, tag in removed_headings
        )
    else:
        headings_output = "<span style='color:gray'> 🔴 No headings found 🔴 </span>"
    return headings_output, cleaned_text

# ✅ Custom CSS targeting only the cleaned output textbox
custom_css = """
#cleaned_output textarea {
    background-color: black !important;
    color: white !important;
}
"""

with gr.Blocks(theme=gr.themes.Soft(), css=custom_css) as app:
    with gr.Row():
        with gr.Column(scale=1): pass
        with gr.Column(scale=2):
            gr.Markdown("## Upload Docx Here to Remove Titles")
        with gr.Column(scale=1): pass

    with gr.Row():
        with gr.Column(scale=1): pass
        with gr.Column(scale=2):
            file_input = gr.File(
                file_types=[".docx"],
                label="📤 Upload DOCX File",
                height=60
            )
            process_btn = gr.Button("🚀 Click to Process", variant="primary")
        with gr.Column(scale=1): pass

    gr.Markdown("---")

    with gr.Row():
        with gr.Column():
            gr.Markdown("### ===========------->  Removed Headings  <-------===========")
            headings_output = gr.Markdown()
        with gr.Column():
            # ✅ elem_id added to target this textbox specifically
            cleaned_output = gr.Textbox(lines=25, elem_id="cleaned_output", buttons=["copy"])

    process_btn.click(
        fn=process_doc,
        inputs=[file_input],
        outputs=[headings_output, cleaned_output]
    )

app.launch()