import streamlit as st from PIL import Image, ImageDraw import img2pdf from io import BytesIO from docx import Document from docx.shared import Inches, Pt from docx.enum.text import WD_PARAGRAPH_ALIGNMENT import nbformat from nbconvert import HTMLExporter from weasyprint import HTML import pandas as pd import markdown from gtts import gTTS import speech_recognition as sr import fitz # PyMuPDF import os st.title("Hari Files Converter") # CSS for styling cards st.markdown(""" """, unsafe_allow_html=True) # Conversion functions def image_to_pdf(uploaded_file): image = Image.open(uploaded_file) pdf_bytes = img2pdf.convert(image.filename, dpi=300) return pdf_bytes def docx_to_html(docx_path): document = Document(docx_path) html_content = "" for para in document.paragraphs: style = para.style font_size = Pt(12).pt if not (style.font and style.font.size) else style.font.size.pt alignment = para.alignment or WD_PARAGRAPH_ALIGNMENT.LEFT align = "left" if alignment == WD_PARAGRAPH_ALIGNMENT.LEFT else \ "center" if alignment == WD_PARAGRAPH_ALIGNMENT.CENTER else \ "right" if alignment == WD_PARAGRAPH_ALIGNMENT.RIGHT else "justify" html_content += f"

{para.text}

" html_content += "" return html_content def docx_to_pdf_func(uploaded_file): temp_docx_path = "temp.docx" with open(temp_docx_path, "wb") as f: f.write(uploaded_file.getbuffer()) html_content = docx_to_html(temp_docx_path) output_pdf = BytesIO() HTML(string=html_content).write_pdf(output_pdf) os.remove(temp_docx_path) return output_pdf.getvalue() def ipynb_to_pdf(uploaded_file): notebook_content = uploaded_file.read().decode('utf-8') notebook = nbformat.reads(notebook_content, as_version=4) html_exporter = HTMLExporter() html_exporter.template_name = 'classic' html_data, _ = html_exporter.from_notebook_node(notebook) output_pdf = BytesIO() HTML(string=html_data).write_pdf(output_pdf) return output_pdf.getvalue() def docx_to_image(uploaded_file): document = Document(uploaded_file) text = "\n".join([para.text for para in document.paragraphs]) image = Image.new('RGB', (1600, 1200), color=(255, 255, 255)) d = ImageDraw.Draw(image) d.text((10, 10), text, fill=(0, 0, 0)) output_image = BytesIO() image.save(output_image, format="PNG", dpi=(300, 300)) return output_image.getvalue() def image_to_docx(uploaded_file): document = Document() image = Image.open(uploaded_file) max_width = Inches(6) width, height = image.size aspect_ratio = height / width document.add_picture(uploaded_file, width=max_width, height=max_width * aspect_ratio) output_docx = BytesIO() document.save(output_docx) return output_docx.getvalue() def pdf_to_image(uploaded_file): pdf_document = fitz.open(stream=uploaded_file.read(), filetype="pdf") image_buffers = [] for page_number in range(len(pdf_document)): page = pdf_document.load_page(page_number) pix = page.get_pixmap() img_buffer = BytesIO() pix.save(img_buffer) image_buffers.append(img_buffer.getvalue()) return image_buffers def text_to_pdf(uploaded_file): text = uploaded_file.read().decode("utf-8") pdf = FPDF() pdf.add_page() pdf.set_font("Arial", size=12) pdf.multi_cell(0, 10, text) output_pdf = BytesIO() pdf.output(output_pdf) return output_pdf.getvalue() def markdown_to_pdf(uploaded_file): markdown_text = uploaded_file.read().decode("utf-8") html_content = markdown.markdown(markdown_text) output_pdf = BytesIO() HTML(string=html_content).write_pdf(output_pdf) return output_pdf.getvalue() def html_to_image(uploaded_file): html_content = uploaded_file.read().decode("utf-8") output_image = BytesIO() HTML(string=html_content).write_png(output_image) return output_image.getvalue() def csv_to_excel(uploaded_file): df = pd.read_csv(uploaded_file) output_excel = BytesIO() df.to_excel(output_excel, index=False, engine='openpyxl') return output_excel.getvalue() def excel_to_csv(uploaded_file): df = pd.read_excel(uploaded_file) output_csv = BytesIO() df.to_csv(output_csv, index=False) return output_csv.getvalue() def audio_to_text(uploaded_file): recognizer = sr.Recognizer() with sr.AudioFile(uploaded_file) as source: audio_data = recognizer.record(source) text = recognizer.recognize_google(audio_data) return text.encode('utf-8') def text_to_speech(uploaded_file): text = uploaded_file.read().decode("utf-8") tts = gTTS(text) output_audio = BytesIO() tts.write_to_fp(output_audio) return output_audio.getvalue() def epub_to_pdf(uploaded_file): from ebooklib import epub from bs4 import BeautifulSoup book = epub.read_epub(uploaded_file) html_content = "" for item in book.get_items_of_type(ebooklib.ITEM_DOCUMENT): soup = BeautifulSoup(item.get_body_content(), 'html.parser') html_content += str(soup) output_pdf = BytesIO() HTML(string=html_content).write_pdf(output_pdf) return output_pdf.getvalue() # Display conversion options as cards conversions = { "Image to PDF": image_to_pdf, "DOCX to PDF": docx_to_pdf_func, "IPYNB to PDF": ipynb_to_pdf, "DOCX to Image": docx_to_image, "Image to DOCX": image_to_docx, "PDF to Image": pdf_to_image, "Text to PDF": text_to_pdf, "Markdown to PDF": markdown_to_pdf, "HTML to Image": html_to_image, "CSV to Excel": csv_to_excel, "Excel to CSV": excel_to_csv, "Audio to Text": audio_to_text, "Text to Speech": text_to_speech, "EPUB to PDF": epub_to_pdf, } # Render each conversion option as a separate card for conversion_name, conversion_func in conversions.items(): with st.container(): st.markdown(f"

", unsafe_allow_html=True) st.markdown(f"

{conversion_name}

", unsafe_allow_html=True) uploaded_file = st.file_uploader(f"Upload file for {conversion_name}", key=conversion_name) if uploaded_file is not None: if st.button(f"Convert {conversion_name}", key=f"button_{conversion_name}"): if conversion_name == "PDF to Image": images = conversion_func(uploaded_file) for i, img in enumerate(images): st.image(img, caption=f"Page {i+1}") st.download_button( label=f"Download Page {i+1}", data=img, file_name=f"output_page_{i+1}.png", mime="image/png" ) else: output_data = conversion_func(uploaded_file) output_mime = "application/pdf" if "PDF" in conversion_name else "image/png" if "Image" in conversion_name else "text/plain" if "Text" in conversion_name or "Markdown" in conversion_name else "application/vnd.openxmlformats-officedocument.wordprocessingml.document" if "DOCX" in conversion_name else "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet" if "Excel" in conversion_name else "audio/mpeg" if "Speech" in conversion_name else "application/pdf" output_ext = "pdf" if "PDF" in conversion_name else "png" if "Image" in conversion_name else "txt" if "Text" in conversion_name or "Markdown" in conversion_name else "docx" if "DOCX" in conversion_name else "xlsx" if "Excel" in conversion_name else "mp3" if "Speech" in conversion_name else "pdf" st.download_button( label="Download Converted File", data=output_data, file_name=f"output.{output_ext}", mime=output_mime ) st.markdown("

", unsafe_allow_html=True)