Got chunked uploading kinda working

First pass of FE chunked uploading (theres a 1 byte less per chunk per file, no idea why)
More formatting!

lockbox/common/constants.py

@@ -39,14 +39,14 @@ CONFIG_KEYS = {
         "verbose_name": "Max per chunk size in bytes",
         "native_type": int,
         "sensitive": False,
-        "default": 1024 * 1024 * 1,  # 20 MB
+        "default": 1024 * 1024 * 2,  # 2 MB
     },
     "MAX_FILE_BYTES": {
         "description": "Max total file size in bytes",
         "verbose_name": "Max upload size in bytes",
         "native_type": int,
         "sensitive": False,
-        "default": 1024 * 1024 * 30,  # 200 MB
+        "default": 1024 * 1024 * 30,  # 300 MB
     },
     "ENABLE_BROWSABLE_API": {
         "description": "REST Framework browsable API is enabled (Always enabled if DEBUG is true)",

lockbox/common/utils.py

@@ -33,7 +33,9 @@ def cast_to_native_type(key, value, native_type):
     try:
         return native_type(value)
     except ValueError as e:
-        message = f"Received unexpected value type for configuration key {key}\nValue: {value}\nExpected type : {native_type}"
+        message = f"Received unexpected value type for configuration key {key}\n\
+            Value: {value}\n\
+            Expected type : {native_type}"
         raise ValueError(message) from e
 
 

lockbox/lockbox/settings.py

@@ -75,8 +75,6 @@ TEMPLATES = [
 WSGI_APPLICATION = "lockbox.wsgi.application"
 
 # Password validation
-# https://docs.djangoproject.com/en/4.2/ref/settings/#auth-password-validators
-
 AUTH_PASSWORD_VALIDATORS = [
     {
         "NAME": "django.contrib.auth.password_validation.UserAttributeSimilarityValidator",

lockbox/lockbox/setup.py

@@ -1,11 +1,15 @@
 from pathlib import Path
 
-# TODO: LOG MEEEEE
-# TODO: Figure out file owner in system, permissions, GUID
-# Whats the default path if not provided? // docker volume
-
 
 def validate_paths(media_path):
+    """Validates media path and validates that we can actually write to that location
+
+    Args:
+        media_path (str): Absolute-path-like string where files are stored
+
+    Raises:
+        e: Any exception that might happen (Permission Denied, Path does not exist, etc.)
+    """    
     try:
         Path(media_path).mkdir(exist_ok=True)
     except Exception as e:

lockbox/static/js/chunked_uploader.js

@@ -1,65 +1,95 @@
-const fileInput = document.getElementById('file-upload');
-const upload_ready = false;
+const fileInput = document.getElementById("file-upload");
+const uploadButton = document.getElementById("upload-button");
+const fileSizeReport = document.getElementById("file-size");
+const progressBar = document.getElementById("progressBar");
 
+fileInput.value = '';
 fileInput.addEventListener('change', handleFileChange);
+uploadButton.addEventListener('click', handleFileUpload);
 
 function handleFileChange(event) {
     const file = event.target.files[0];
     const file_size = file.size;
-
-    console.log("Max file bytes is : ", max_file_bytes);
-    console.log("File size is: ", file_size);
+    fileSizeReport.textContent = "File size is: " + file.size;
 
     if (file_size > max_file_bytes){
-        console.log("PLACEHOLDER: Size too big man.");
+        console.log("File size is too large");
+        // Handle this.
         return
     }
-
-    console.log("PLACEHOLDER: Ready!");
-    handleFileUpload(file);
+    console.log("Ready!");
 }
 
-function handleFileUpload(file) {
-    let start = 0;
-    let end = 0;
-    let chunk;
-    let file_id = null;
-
-    while (start < file.size) {
-        chunk = file.slice(start, start + chunk_size);
-        end = chunk.size - start;
-        console.log("LID: ", file_id);
-        file_id = uploadChunk(chunk, start, end, file.size, file_id);
-        start += chunk_size;
+async function handleFileUpload(event) {
+    let headers = new Headers();
+    headers.append("Content-Type", "application/json");
+    const request_args = {
+        method: "POST",
+        headers: headers,
+        body: JSON.stringify(
+            {
+                "expected_size": fileInput.files[0].size
+            }
+        )
+    };
+    const response = await fetch(uploadPath, request_args);
+    if (!response.ok) {
+        throw new Error(`Response status: ${response.status}`);
     }
+    const file = await response.json();
+    await uploadChunks(file);
 }
 
-function uploadChunk(chunk, start, end, total, file_id=null) {
-    const formData = new FormData();
-    const range_header = `bytes ${start}-${end}/${total}`;
-    formData.append('file', chunk);
+function updateProgressBar(remaining, total) {
+    let current_percent = Math.round((total - remaining) / (total / 100));
+    progressBar.textContent = current_percent;
+}
 
+async function uploadChunks(remoteFile){
+    const chunkPath = chunkPathTemplate.replace("@", remoteFile.lid);
+    let file = fileInput.files[0];
 
-    if (file_id) {
-        formData.append("lid", file_id);
-    }
+    let bytes_remaining = remoteFile.expected_size
+    let last_transfer_position = 0;
+    let to_transfer = remoteFile.max_size_chunk_bytes;
+    console.log("Chunk size is: " + remoteFile.max_size_chunk_bytes);
 
-    let request = new Request(".", {
-        method: 'POST',
-        body: formData,
-        headers: {
-            'X-CSRFToken': csrftoken,
-            'Content-range': range_header
+    while (bytes_remaining >= 0) {
+        updateProgressBar(bytes_remaining, remoteFile.expected_size);
+        if (bytes_remaining <= remoteFile.max_size_chunk_bytes) {
+            to_transfer = bytes_remaining;
+            bytes_remaining = 0;
         }
-    })
-    return _uploadChunk(request);
-}   
+        
+        await uploadChunk(file, [last_transfer_position, last_transfer_position += to_transfer], chunkPath);
+        last_transfer_position += 1;
+        bytes_remaining -= to_transfer;
+    }
+    console.log("Done!")
+    progressBar.textContent = 100;
 
-async function _uploadChunk(request) {
-    const _response = await fetch(request)
-    .then(async (response)=>response.json())
-    .then((data) =>{
-        return data.lid;
-    })
-    return _response;
+}
+
+async function uploadChunk(file, byte_range, chunkPath) {
+    console.log(byte_range);
+    let file_bytes_target = file.slice(byte_range[0], byte_range[1]);
+    let body = new FormData();
+
+    body.append("Content", file_bytes_target);
+
+    let headers = new Headers();
+    headers.append("Content-Disposition", 'attachment; filename="DUMMY"');
+    headers.append("Content-Range", "bytes " + byte_range[0] + "-" + byte_range[1])
+
+    const request_args = {
+        method: "PUT",
+        headers: headers,
+        body: body
+    };
+
+    const response = await fetch(chunkPath, request_args);
+
+    if (!response.ok) {
+        throw new Error(`Response status: ${response.status}`);
+    }
 }

lockbox/storage/migrations/0001_initial.py

@@ -1,4 +1,4 @@
-# Generated by Django 4.2.15 on 2024-09-17 06:52
+# Generated by Django 4.2.15 on 2024-09-17 19:51
 
 import common.utils
 from django.conf import settings
@@ -24,13 +24,13 @@ class Migration(migrations.Migration):
                 ('date_created', models.DateTimeField(blank=True, help_text='date at which this object was created', verbose_name='date created')),
                 ('date_updated', models.DateTimeField(blank=True, help_text='date at which this object was last updated', verbose_name='date updated')),
                 ('mime_type', models.CharField(blank=True, help_text='reported mime-type', max_length=128, null=True, verbose_name='mime-type')),
-                ('file', models.FileField(blank=True, help_text='actual file', null=True, upload_to=storage.models.upload_to_fielpath, verbose_name='file')),
+                ('file', models.FileField(blank=True, help_text='actual file', null=True, upload_to=storage.models._upload_to_fielpath, verbose_name='file')),
                 ('status', models.CharField(choices=[('uploading', 'uploading'), ('completed', 'completed'), ('abandoned', 'abandoned'), ('error', 'error')], default='uploading', help_text='upload status for file', max_length=10, verbose_name='status')),
                 ('datetime_completed', models.DateTimeField(blank=True, help_text="datetime at which this file's upload was completed", null=True, verbose_name='completed on')),
                 ('expires', models.BooleanField(default=False, help_text="will be scrubbed on 'date_expires'", verbose_name='expires')),
                 ('delete_on_expiration', models.BooleanField(default=False, help_text='will be deleted if expired and expires is true', verbose_name='delete on expiration')),
                 ('size', models.PositiveBigIntegerField(blank=True, help_text='total size on disk for this file', null=True, verbose_name='size (bytes)')),
-                ('expected_size', models.PositiveBigIntegerField(blank=True, help_text='expected file size', null=True, verbose_name='expected size (bytes)')),
+                ('expected_size', models.PositiveBigIntegerField(help_text='expected file size', verbose_name='expected size (bytes)')),
                 ('max_size_chunk_bytes', models.PositiveBigIntegerField(default=common.utils.get_max_size_chunk_bytes, help_text='max size of each individual chunk for this file', verbose_name='maximum size of chunks (bytes)')),
                 ('last_end_bytes', models.BigIntegerField(blank=True, help_text='last uploaded bytes position', null=True, verbose_name='last end bytes')),
                 ('owner', models.ForeignKey(blank=True, help_text='Who owns this file', null=True, on_delete=django.db.models.deletion.SET_NULL, related_name='files_owned', to=settings.AUTH_USER_MODEL, verbose_name='owner')),

lockbox/storage/models.py

@@ -6,6 +6,7 @@ from common.constants import UPLOAD_ERROR_CODES, UPLOAD_STATUS_TYPES
 from common.models import LockboxBase
 from common.utils import get_config, get_max_size_chunk_bytes
 from django.conf import settings
+from django.core.exceptions import ValidationError
 from django.db import models, transaction
 from django.utils import timezone
 from django.utils.translation import gettext_lazy as _
@@ -18,7 +19,7 @@ class UploadError(Exception):
         super().__init__(*args, **kwargs)
 
 
-def upload_to_fielpath(instance, filename):
+def _upload_to_fielpath(instance, filename):
     return Path(str(instance.lid)).joinpath(f"{filename}{settings.INCOMPLETE_EXT}")
 
 
@@ -36,7 +37,7 @@ class File(LockboxBase):
         blank=True,
         verbose_name=_("file"),
         help_text=_("actual file"),
-        upload_to=upload_to_fielpath,
+        upload_to=_upload_to_fielpath,
     )
 
     UPLOAD_CHOICES = (
@@ -97,8 +98,8 @@ class File(LockboxBase):
     )
 
     expected_size = models.PositiveBigIntegerField(
-        null=True,
-        blank=True,
+        null=False,
+        blank=False,
         verbose_name=_("expected size (bytes)"),
         help_text=_("expected file size"),
     )
@@ -143,20 +144,20 @@ class File(LockboxBase):
     @property
     def md5(self):
         if self.exists:
-            return md5(open(self.file, "rb").read()).hexdigest()
+            self.file.open("rb")
+            md5_hash = md5(self.file.read()).hexdigest()
+            self.file.close()
+            return md5_hash
         return None
 
     @property
     def abandoned(self):
-        return self.date_created <= timezone.now() + timedelta(
-            minutes=get_config("ABANDONED_DELTA_MINUTES")
-        )
+        return self.date_created <= timezone.now() + timedelta(minutes=get_config("ABANDONED_DELTA_MINUTES"))
 
     @classmethod
     def abandoned_condition():
         return models.Q(
-            date_created__lte=timezone.now()
-            + timedelta(minutes=get_config("ABANDONED_DELTA_MINUTES"))
+            date_created__lte=timezone.now() + timedelta(minutes=get_config("ABANDONED_DELTA_MINUTES"))
         )
 
     @property
@@ -174,6 +175,16 @@ class File(LockboxBase):
         return Path(self.file.path).is_file()
 
     def append_chunk(self, chunk_file, chunk_data):
+        """Append chunks to a file
+
+        Args:
+            chunk_file (UploadedFile): Django provided abstraction that contains the actual file in memory
+            chunk_data (Dict): Additional chunk parameters: start_bytes, end_bytes, size
+
+        Raises:
+            UploadError: The file you are trying to append to is not found in FS
+            UploadError: Expected last_end_bytes + 1 but got a different number
+        """
         # Override in case recently abandoned
         # Will persist if it does not error out.
         self.status = UPLOAD_STATUS_TYPES.UPLOADING
@@ -192,24 +203,32 @@ class File(LockboxBase):
         if self.last_end_bytes and self.last_end_bytes + 1 != chunk_data["start_bytes"]:
             # Client screwed up, this is not where we left
             raise UploadError(
-                "Mismatch in expected chunk", code=UPLOAD_ERROR_CODES.CHUNK_MISMATCH
+                "Mismatch in expected chunk",
+                code=UPLOAD_ERROR_CODES.CHUNK_MISMATCH
             )
 
         self.last_end_bytes = chunk_data["end_bytes"]
+        if chunk_data["start_bytes"] == 0:
+            self.file = chunk_file
+            self.save()
+        else:
+            chunk_file.open("rb")
+            self.file.open("ab")
+            self.file.write(chunk_file.read())
+            self.file.close()
+            self.save()
+
         if self.expected_size == self.last_end_bytes:
-            # File is one shot chunk.
-            if chunk_data["start_bytes"] == 0:
-                self.file = chunk_file
             self.save()
             self.finalize()
             return
 
-        # This is an n + 1 chunk.
-        print("Appending bytes yo")
-        chunk_file.seek(0)
-        self.save()
-
     def finalize(self):
+        """Finalizes the file
+        Sets file status to 'completed'
+        Sets datetime_completed to now
+        Renames file from file.extention.incomplete to file.extention
+        """
         self.refresh_from_db()
         self.status = UPLOAD_STATUS_TYPES.COMPLETED
         self.datetime_completed = timezone.now()
@@ -223,31 +242,17 @@ class File(LockboxBase):
     def save(self, *args, **kwargs):
         if not self.max_size_chunk_bytes:
             self.max_size_chunk_bytes = get_max_size_chunk_bytes()
+
+        if self.expected_size > get_config("MAX_FILE_BYTES"):
+            raise ValidationError(f"Expected size: {self.expected_size} > than config MAX_SIZE_BYTES")     
         return super().save(*args, **kwargs)
 
     def delete(self, *args, **kwargs):
-
         with transaction.atomic():
             if self.file:
                 if Path(self.file.path).is_file():
                     self.file.storage.delete(self.file.path)
+                    # Delete containing directory (UUID)
                     self.file.storage.delete(Path(self.file.path).parent)
             result = super().delete(*args, **kwargs)
         return result
-
-
-# class FileShare(LockboxBase):
-#     file = models.ForeignKey(
-#         "storage.File",
-#         null=False,
-#         blank=False,
-#         on_delete=models.CASCADE,
-#         related_name="shares",
-#     )
-
-#     def __str__(self):
-#         return self.file.name
-
-#     class Meta:
-#         verbose_name = _("share")
-#         verbose_name_plural = _("shares")

lockbox/storage/views_api.py

@@ -1,4 +1,8 @@
-from common.constants import CONTENT_RANGE_HEADER, CONTENT_RANGE_HEADER_PATTERN
+from common.constants import (
+    CONTENT_RANGE_HEADER,
+    CONTENT_RANGE_HEADER_PATTERN,
+    UPLOAD_STATUS_TYPES,
+)
 from django.core.exceptions import ValidationError
 from rest_framework import status
 from rest_framework.decorators import action
@@ -15,6 +19,15 @@ class FileModelViewSet(ModelViewSet):
     queryset = File.objects.all()
     serializer_class = FileSerializer
 
+    @action(detail=True, methods=["GET"])
+    def md5(self, request, pk=None):
+        try:
+            file = File.objects.filter(lid=pk).first()
+        except ValidationError:
+            raise UserValidationError(f"UUID {pk} is not a valid UUID")
+
+        return Response({"md5": file.md5}, status=status.HTTP_200_OK)
+
     @action(detail=True, methods=["PUT"])
     def append_chunk(self, request, filename="DUMMY", format=None, pk=None):
         try:
@@ -25,11 +38,22 @@ class FileModelViewSet(ModelViewSet):
         if not file:
             raise NotFound(f"File with ID {pk} was not found")
 
-        chunk_data = self.get_content_range(request)
+        if file.status in [UPLOAD_STATUS_TYPES.COMPLETED, UPLOAD_STATUS_TYPES.ERROR]:
+            raise UserValidationError(
+                f"Cannot append chunk to file, status is: {file.status}"
+            )
+
+        chunk_data = self._get_content_range(request)
         if not chunk_data:
             raise UserValidationError("Missing content-range headers")
 
-        chunk_file = request.FILES["Content"]
+        chunk_file = request.FILES.get("Content", None)
+
+        if not chunk_file:
+            raise UserValidationError(
+                "Could not find 'Content' in request body"
+            )
+
         if chunk_file.size > file.max_size_chunk_bytes:
             raise UserValidationError(
                 f"Chunk size is greater than files max chunk size: {chunk_file.size} > {file.max_size_chunk_bytes}"
@@ -48,7 +72,7 @@ class FileModelViewSet(ModelViewSet):
             return Response({"code": e.code}, status=status.HTTP_400_BAD_REQUEST)
         return Response(status=status.HTTP_201_CREATED)
 
-    def get_content_range(self, request):
+    def _get_content_range(self, request):
         content_range = request.META.get(CONTENT_RANGE_HEADER, None)
         if not content_range:
             return None

lockbox/storage/views_client.py

@@ -1,3 +1,4 @@
+from common.utils import get_config
 from django.shortcuts import render
 from django.views import View
 
@@ -5,4 +6,7 @@ from django.views import View
 # Static view
 class FileUploadView(View):
     def get(self, request):
-        return render(request, "storage/upload.html")
+        context = {
+            "max_file_bytes": get_config("MAX_FILE_BYTES"),
+        }
+        return render(request, "storage/upload.html", context=context)

lockbox/templates/storage/upload.html

@@ -1,40 +1,39 @@
 {% extends "base.html" %}
-
 {% block title %}Upload a file{% endblock %}
-
 {% block postjs  %}
-
 {% load static %}
 {{ block.super }}
 
 <script src="{% static 'js/utils.js' %}"></script>
 
 <script>
-    const chunk_size = {{ max_chunk_bytes }};
     const max_file_bytes = {{ max_file_bytes }};
-    const csrftoken = getCookie('csrftoken');
+    const uploadPath = "{% url 'file-list' %}";
+    const chunkPathTemplate = "{% url 'file-append-chunk' pk='@'%}";
+
+    console.log("upload path is: " + uploadPath);
+    console.log("chunk path is: " + chunkPathTemplate);
+
 </script>
 
 <script src="{% static 'js/chunked_uploader.js' %}"></script>
 
 {% endblock %}
-
-
 {% block content %}
 
 <p> Upload file </p>
-
-<table>
-    {{form}}
-</table>
-
+<p id="max_size">Max size allowed size is: {{max_file_bytes}} bytes</p>
 <p id="file-size"></p>
 
-<form method="post">{% csrf_token %}
-    <input type="file" id="file-upload">
-</form>
+
+<table>
+    <form>
+        <label for="expected_size">Expected Size:</label><br>
+        <input type="file" id="file-upload">
+        <input type="button" id="upload-button" value="Upload">
+    </form>
+</table>
+
+<p id="progressBar"></p>
 
 {% endblock %}
-
-
-