fix: auto-compress large browser uploads and accept csv.gz

Author: Mihai - Alexandru Chindriș

src/web/app.py

@@ -263,8 +263,20 @@ def upload():
         if not file or file.filename == "":
             return render_template("upload.html", error="No file selected.")
 
+        filename = file.filename.lower()
+        if not (filename.endswith(".csv") or filename.endswith(".csv.gz")):
+            return render_template(
+                "upload.html",
+                error="Unsupported file type. Please upload a .csv or .csv.gz file.",
+            )
+        compression = "gzip" if filename.endswith(".gz") else None
+
         # Stream CSV in chunks to keep memory bounded for large uploads.
-        chunks = pd.read_csv(file.stream, chunksize=UPLOAD_CHUNK_SIZE)
+        chunks = pd.read_csv(
+            file.stream,
+            chunksize=UPLOAD_CHUNK_SIZE,
+            compression=compression,
+        )
         first_chunk = next(chunks, None)
         if first_chunk is None or first_chunk.empty:
             return render_template("upload.html", error="Uploaded CSV is empty.")

src/web/templates/upload.html

@@ -14,8 +14,9 @@ <h1>Batch Prediction</h1>
 <form method="POST" action="/upload" enctype="multipart/form-data">
     <div class="form-group">
         <label for="file">Select CSV file:</label>
-        <input type="file" name="file" id="file" accept=".csv" required>
+        <input type="file" name="file" id="file" accept=".csv,.gz" required>
     </div>
+    <p id="upload-status" aria-live="polite"></p>
     <button type="submit" class="btn btn-primary">Upload & Predict</button>
 </form>
 
@@ -74,4 +75,69 @@ <h2>Predictions ({{ num_predictions }} instances)</h2>
     </div>
 </div>
 {% endif %}
+
+<script>
+(() => {
+    const form = document.querySelector('form[action="/upload"]');
+    const fileInput = document.getElementById('file');
+    const statusEl = document.getElementById('upload-status');
+    const LARGE_FILE_THRESHOLD = 8 * 1024 * 1024;
+
+    if (!form || !fileInput || !statusEl) {
+        return;
+    }
+
+    const setStatus = (text) => {
+        statusEl.textContent = text;
+    };
+
+    form.addEventListener('submit', async (event) => {
+        const file = fileInput.files && fileInput.files[0];
+        if (!file) {
+            return;
+        }
+
+        if (file.size < LARGE_FILE_THRESHOLD || typeof CompressionStream === 'undefined') {
+            return;
+        }
+
+        event.preventDefault();
+        const submitButton = form.querySelector('button[type="submit"]');
+        if (submitButton) {
+            submitButton.disabled = true;
+        }
+
+        try {
+            setStatus('Large file detected. Compressing before upload...');
+            const gzipStream = file.stream().pipeThrough(new CompressionStream('gzip'));
+            const compressedBlob = await new Response(gzipStream).blob();
+            const gzipFile = new File([compressedBlob], file.name + '.gz', {
+                type: 'application/gzip',
+                lastModified: Date.now(),
+            });
+
+            const formData = new FormData();
+            formData.append('file', gzipFile);
+
+            setStatus('Uploading compressed file...');
+            const response = await fetch('/upload', {
+                method: 'POST',
+                body: formData,
+            });
+
+            const html = await response.text();
+            document.open();
+            document.write(html);
+            document.close();
+        } catch (_error) {
+            setStatus('Compression failed. Falling back to regular upload...');
+            form.submit();
+        } finally {
+            if (submitButton) {
+                submitButton.disabled = false;
+            }
+        }
+    });
+})();
+</script>
 {% endblock %}

tests/test_api.py

@@ -3,6 +3,8 @@
 import json
 import re
 from pathlib import Path
+import gzip
+import io
 
 import pytest
 from src.web.app import app
@@ -71,3 +73,23 @@ def test_upload_csv_and_download_results(client):
     assert download_response.status_code == 200
     assert "attachment" in download_response.headers.get("Content-Disposition", "")
     assert b"Prediction" in download_response.data
+
+
+def test_upload_gzipped_csv(client):
+    """Test batch upload accepts .csv.gz files."""
+    csv_path = Path(__file__).resolve().parents[1] / "data" / "test_set.csv"
+
+    with csv_path.open("rb") as f:
+        raw = f.read()
+    compressed = gzip.compress(raw)
+    file_obj = io.BytesIO(compressed)
+
+    response = client.post(
+        "/upload",
+        data={"file": (file_obj, "test_set.csv.gz")},
+        content_type="multipart/form-data",
+    )
+
+    assert response.status_code == 200
+    html = response.data.decode("utf-8")
+    assert "Predictions (" in html