Spaces:

Projetweb
/

Ai-prject

Runtime error

App Files Files Community

rayhane123 commited on 14 days ago

Commit

6c95da8

verified ·

1 Parent(s): ec71fe6

Upload 10 files

Browse files

Files changed (11) hide show

.gitattributes +2 -0
translation/__pycache__/main.cpython-312.pyc +0 -0
translation/app.py +8 -0
translation/requirements.txt +9 -0
translation/static/__pycache__/main.cpython-312.pyc +0 -0
translation/static/__pycache__/main.cpython-39.pyc +0 -0
translation/static/index.html +128 -0
translation/static/main.py +145 -0
translation/static/uploads/TD1 (1).docx +0 -0
translation/static/uploads/Untitled.pdf +3 -0
translation/static/uploads/chapitre 2 RECONNAISSANCE DE FORMES ET CLASSIFICATION.pdf +3 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+translation/static/uploads/chapitre[[:space:]]2[[:space:]]RECONNAISSANCE[[:space:]]DE[[:space:]]FORMES[[:space:]]ET[[:space:]]CLASSIFICATION.pdf filter=lfs diff=lfs merge=lfs -text
+translation/static/uploads/Untitled.pdf filter=lfs diff=lfs merge=lfs -text

translation/__pycache__/main.cpython-312.pyc ADDED Viewed

Binary file (7.41 kB). View file

translation/app.py ADDED Viewed

	@@ -0,0 +1,8 @@

+import sys
+import os
+# Add the "translation" folder to Python's path
+sys.path.append(os.path.join(os.path.dirname(__file__), "translation"))
+# Import FastAPI app from main.py
+from translation.main import app

translation/requirements.txt ADDED Viewed

	@@ -0,0 +1,9 @@

+fastapi
+uvicorn
+transformers
+torch
+textwrap
+pymupdf
+python-docx
+openpyxl
+python-pptx

translation/static/__pycache__/main.cpython-312.pyc ADDED Viewed

Binary file (7.41 kB). View file

translation/static/__pycache__/main.cpython-39.pyc ADDED Viewed

Binary file (318 Bytes). View file

translation/static/index.html ADDED Viewed

	@@ -0,0 +1,128 @@

+<!DOCTYPE html>
+<html lang="fr">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Document Translator - IA Edition</title>
+    <link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet">
+    <script src="https://code.jquery.com/jquery-3.6.0.min.js"></script>
+    <script src="https://cdn.jsdelivr.net/npm/sweetalert2@11"></script>
+    <script src="https://cdnjs.cloudflare.com/ajax/libs/jspdf/2.5.1/jspdf.umd.min.js"></script>
+    <style>
+        body {
+            background: linear-gradient(135deg, #1e3c72, #2a5298);
+            color: white;
+            text-align: center;
+            padding: 40px;
+        }
+        .container {
+            max-width: 600px;
+            background: rgba(255, 255, 255, 0.1);
+            padding: 20px;
+            border-radius: 15px;
+            backdrop-filter: blur(10px);
+            box-shadow: 0px 0px 15px rgba(255, 255, 255, 0.2);
+        }
+        .result-box {
+            background: white;
+            color: black;
+            padding: 15px;
+            border-radius: 10px;
+            text-align: left;
+            max-height: 300px;
+            overflow-y: auto;
+            margin-top: 15px;
+            font-size: 16px;
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>📝 AI Document Translator</h1>
+        <p class="file-info">Formats supportés : <b>TXT, PDF, DOCX, PPTX, XLSX</b></p>
+        <form id="uploadForm">
+            <input type="file" id="fileInput" name="file" class="form-control mt-3" required>
+            <select id="src_lang" name="src_lang" class="form-control mt-3">
+                <option >Francais</option>
+                <option >Anglais</option>
+                <option >Arabe</option>
+                <option >Espagnol</option>
+            </select>
+            <select id="tgt_lang" name="tgt_lang" class="form-control mt-3">
+                <option >Francais</option>
+                <option >Anglais</option>
+                <option >Arabe</option>
+                <option >Espagnol</option>
+            </select>
+            <button type="submit" class="btn btn-success mt-3">Traduire</button>
+        </form>
+        <div class="menu mt-3">
+            <button type="button" class="btn btn-info" id="openFile">📂 Ouvrir le fichier</button>
+            <button id="copyText" class="btn btn-secondary" disabled>📋 Copier</button>
+            <button id="speakText" class="btn btn-dark" disabled>🔊 Écouter</button>
+            <button id="downloadTxt" class="btn btn-warning" disabled>⬇️ Télécharger en TXT</button>
+            <button id="downloadPdf" class="btn btn-danger" disabled>📄 Télécharger en PDF</button>
+            <button id="summarizeText" class="btn btn-primary" disabled>✂️ Résumer</button>
+        </div>
+        <h2 class="mt-4">Résultat :</h2>
+        <div id="result" class="result-box">Aucun texte traduit pour l’instant...</div>
+    </div>
+    <script>
+        $(document).ready(function() {
+            $('#uploadForm').submit(async function(event) {
+                event.preventDefault();
+                let formData = new FormData(this);
+                Swal.fire({ title: 'Traduction en cours...', didOpen: () => { Swal.showLoading(); }});
+                let response = await fetch('/upload/', { method: 'POST', body: formData });
+                let result = await response.json();
+                Swal.close();
+                if (result.translated_text) {
+                    $('#result').html(`<b>✨ Traduction :</b><br>${result.translated_text}`);
+                    $('#copyText, #speakText, #downloadTxt, #downloadPdf, #summarizeText').prop("disabled", false);
+                } else {
+                    $('#result').html(`❌ Erreur: ${result.detail}`);
+                }
+            });
+            $('#speakText').click(function() {
+                let text = $('#result').text().trim();
+                if (!text) return;
+                let langMap = { "Francais": "fr-FR", "Anglais": "en-US", "Arabe": "ar-SA", "Espagnol": "es-ES" };
+                let lang = $('#tgt_lang').val();
+                let utterance = new SpeechSynthesisUtterance(text);
+                utterance.lang = langMap[lang] || "fr-FR";
+                utterance.rate = 1;
+                utterance.pitch = 1;
+                utterance.volume = 1;
+                speechSynthesis.speak(utterance);
+            });
+            $('#summarizeText').click(async function() {
+                let text = $('#result').text().trim();
+                if (!text) return;
+                Swal.fire({ title: 'Résumé en cours...', didOpen: () => { Swal.showLoading(); }});
+                let response = await fetch('/summarize/', {
+                    method: 'POST',
+                    headers: { 'Content-Type': 'application/json' },
+                    body: JSON.stringify({ text: text })
+                });
+                let result = await response.json();
+                Swal.close();
+                if (result.summary) {
+                    Swal.fire("Résumé", result.summary, "info");
+                } else {
+                    Swal.fire("Erreur", "Impossible de résumer le texte.", "error");
+                }
+            });
+        });
+    </script>
+</body>
+</html>

translation/static/main.py ADDED Viewed

	@@ -0,0 +1,145 @@

+from fastapi import FastAPI, File, UploadFile, HTTPException, Form
+from fastapi.responses import HTMLResponse
+from fastapi.staticfiles import StaticFiles
+from transformers import pipeline
+import textwrap
+import fitz  # PyMuPDF for PDF handling
+from docx import Document
+import openpyxl  # For Excel
+from pptx import Presentation
+from functools import lru_cache
+import os
+# Initialize FastAPI app
+app = FastAPI()
+# Set the correct path for static files
+STATIC_DIR = r"C:\Users\User\doc_translation_service\translation\static"
+# Ensure the static directory exists
+if not os.path.exists(STATIC_DIR):
+    os.makedirs(STATIC_DIR)
+# Mount static files (serves index.html)
+app.mount("/static", StaticFiles(directory=STATIC_DIR), name="static")
+@app.get("/", response_class=HTMLResponse)
+async def read_root():
+    index_path = os.path.join(STATIC_DIR, "index.html")
+    try:
+        with open(index_path, "r", encoding="utf-8") as file:
+            return HTMLResponse(content=file.read())
+    except FileNotFoundError:
+        raise HTTPException(status_code=404, detail="index.html not found in static folder.")
+# Language codes (ISO 639-1)
+LANGUAGE_CODES = {
+    "Anglais": "en",
+    "Francais": "fr",
+    "Arabe": "ar",
+    "Espagnol": "es",
+}
+# Available translation models
+AVAILABLE_MODELS = {
+    "fr-en": "Helsinki-NLP/opus-mt-fr-en",
+    "en-fr": "Helsinki-NLP/opus-mt-en-fr",
+    "ar-en": "Helsinki-NLP/opus-mt-ar-en",
+    "en-ar": "Helsinki-NLP/opus-mt-en-ar",
+    "es-en": "Helsinki-NLP/opus-mt-es-en",
+    "en-es": "Helsinki-NLP/opus-mt-en-es",
+}
+# Cache model loading
+@lru_cache(maxsize=10)
+def load_translator(src_code: str, tgt_code: str):
+    model_key = f"{src_code}-{tgt_code}"
+    if model_key in AVAILABLE_MODELS:
+        return pipeline("translation", model=AVAILABLE_MODELS[model_key])
+    elif src_code != "en" and tgt_code != "en":
+        return (
+            pipeline("translation", model=AVAILABLE_MODELS.get(f"{src_code}-en")),
+            pipeline("translation", model=AVAILABLE_MODELS.get(f"en-{tgt_code}"))
+        )
+    else:
+        raise ValueError(f"No model available for {src_code} -> {tgt_code}")
+# Split text into chunks
+def chunk_text(text, max_length=400):
+    return textwrap.wrap(text, max_length)
+# Extract text based on file type
+def extract_text(file: UploadFile):
+    try:
+        if file.filename.endswith(".txt"):
+            return file.file.read().decode("utf-8")
+        elif file.filename.endswith(".pdf"):
+            doc = fitz.open(stream=file.file.read(), filetype="pdf")
+            return "\n".join([page.get_text() for page in doc])
+        elif file.filename.endswith(".docx"):
+            doc = Document(file.file)
+            return "\n".join([para.text for para in doc.paragraphs])
+        elif file.filename.endswith(".xlsx"):
+            wb = openpyxl.load_workbook(file.file)
+            text = ""
+            for sheet in wb.sheetnames:
+                ws = wb[sheet]
+                for row in ws.iter_rows():
+                    text += "\t".join([str(cell.value or "") for cell in row]) + "\n"
+            return text
+        elif file.filename.endswith(".pptx"):
+            prs = Presentation(file.file)
+            text = ""
+            for slide in prs.slides:
+                for shape in slide.shapes:
+                    if hasattr(shape, "text"):
+                        text += shape.text + "\n"
+            return text
+        else:
+            raise HTTPException(status_code=400, detail="File type not supported.")
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error extracting text: {str(e)}")
+@app.post("/upload/")
+async def upload_file(
+    file: UploadFile = File(...),
+    src_lang: str = Form(...),
+    tgt_lang: str = Form(...)
+):
+    text = extract_text(file)
+    if not text.strip():
+        raise HTTPException(status_code=400, detail="No text extracted from the file.")
+    src_code = LANGUAGE_CODES.get(src_lang)
+    tgt_code = LANGUAGE_CODES.get(tgt_lang)
+    if not src_code or not tgt_code:
+        raise HTTPException(status_code=400, detail=f"Unsupported language: {src_lang} -> {tgt_lang}")
+    try:
+        # Load translation model
+        translator = load_translator(src_code, tgt_code)
+        # If indirect translation via English
+        if isinstance(translator, tuple):
+            translator1, translator2 = translator
+            intermediate_text = "\n".join([translator1(chunk)[0]['translation_text'] for chunk in chunk_text(text)])
+            translated_text = "\n".join([translator2(chunk)[0]['translation_text'] for chunk in chunk_text(intermediate_text)])
+        else:
+            translated_text = "\n".join([translator(chunk)[0]['translation_text'] for chunk in chunk_text(text)])
+        return {"translated_text": translated_text}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Internal error: {str(e)}")

translation/static/uploads/TD1 (1).docx ADDED Viewed

Binary file (53.4 kB). View file

translation/static/uploads/Untitled.pdf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:eb03deee8b41e265cde7eb86f14deae4cc505d9fef18a3cdaccde123f7ca6ae8
+size 349803

translation/static/uploads/chapitre 2 RECONNAISSANCE DE FORMES ET CLASSIFICATION.pdf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:cad0c7e09e566f603390d90611e4fd0c969d6920a064b4cf74fdbcde58fc8c7e
+size 1665388