I want to create endpoint that will stream video stream that is stored in azure CloudBlob. Here is snippet of my code:
public async Task<IActionResult> GetVideo(string videoId)
{
var videoStream = await _contentStorage.Get(videoId);
var fileStreamResult = new FileStreamResult(videoStream, mimeType);
fileStreamResult.EnableRangeProcessing = true;
return fileStreamResult;
}
and in ContentStorage
public async Task<StoredContent> Get(string id)
{
var block = _blobContainer.GetBlobClient(id);
var ms = await block.OpenReadAsync();
return ms;
}
I had everything working fine except iPhones and safari, after some debugging it turned out that my endpoint is returning 200 http code, but it should 206 - partial content. So I made some changes into my code, here is some snippet:
public async Task<IActionResult> GetVideo(string videoId)
{
var videoStream = await _contentStorage.Get(videoId);
var ms = new MemoryStream();
await videoStream.CopyToAsync(ms);
var fileStreamResult = new FileStreamResult(ms, mimeType);
fileStreamResult.EnableRangeProcessing = true;
return fileStreamResult;
}
Now when I test it on iphone or by postman response is 206, and it's working fine. But I thing that copping the video stream into new memorystream is a valid approach.
Correct me if I'm wrong but I understand this code as for every partial of the video, I'm downloading whole video from blob storage, cut it and then return just the piece within range.
It's not sure for me how to handle this case, is there any out of the box solution for that, or do I need to read range header from request and use OpenReadAsync with parameters as position and buffer side? Or there is another way?
Solution for me was to update Azure.Storage.Blobs library. I had 12.6.0 and after update to 12.7.0 it started working as expected, since they added:
Added seekability to BaseBlobClient.OpenRead().
Related
I want to download a storage blob from Azure and stream it to a client via an .NET Web-App. The blob was uploaded correctly and is visible in my Azure storage account.
Surprisingly, the following throws an exception within HttpBaseStream:
[...]
var blobClient = _containerClient.GetBlobClient(Path.Combine(fileName));
var stream = await blobClient.OpenReadAsync();
return stream;
-> When i step further and return a File (return File(stream, MediaTypeNames.Application.Octet);), the download works as intended.
I tried to push the stream into an MemoryStream, which also fails with the same exception:
[...]
var blobClient = _containerClient.GetBlobClient(Path.Combine(fileName));
var stream = new MemoryStream();
await blobClient.DownloadToAsync(stream);
return stream
->When i step further, returning the file results in a timeout.
How can i fix that? Why do i get this exception - i followed the official quickstart guide from Microsoft.
the following throws an exception within HttpBaseStream
It looks like the HTTP result type is attempting to set the Content-Length header and is reading Length to do so. That would be the natural thing to do. However, it would also be natural to handle the NotSupportedException and just not set Content-Length at all.
If the NotSupportedException only shows up when running in the debugger, then just ignore it.
If the exception is actually thrown to your code (i.e., causing the request to fail), then you'll need to follow the rest of this answer.
First, create a minimal reproducible example and report a bug to the .NET team.
To work around this issue in the meantime, I recommend writing a stream wrapper that returns an already-determined length, which you can get from the Azure blob attributes. E.g.:
public sealed class KnownLengthStreamWrapper : Stream
{
private readonly Stream _stream;
public KnownLengthStreamWrapper(Stream stream, long length)
{
_stream = stream;
Length = length;
}
public override long Length { get; private set; }
... // override all other Stream members and forward to _stream.
}
That should be sufficient to get your app working.
I tried to push the stream into an MemoryStream
This didn't work because you'd need to "rewind" the MemoryStream at some point, e.g.:
var stream = new MemoryStream();
await blobClient.DownloadToAsync(stream);
stream.Position = 0;
return stream;
Check this sample of all the blob options which i have already posted on git working as expected. Reference
public void DownloadBlob(string path)
{
storageAccount = CloudStorageAccount.Parse(CloudConfigurationManager.GetSetting("StorageConnectionString"));
CloudBlobClient client = storageAccount.CreateCloudBlobClient();
CloudBlobContainer container = client.GetContainerReference("images");
CloudBlockBlob blockBlob = container.GetBlockBlobReference(Path.GetFileName(path));
using (MemoryStream ms = new MemoryStream())
{
blockBlob.DownloadToStream(ms);
HttpContext.Current.Response.ContentType = blockBlob.Properties.ContentType.ToString();
HttpContext.Current.Response.AddHeader("Content-Disposition", "Attachment; filename=" + Path.GetFileName(path).ToString());
HttpContext.Current.Response.AddHeader("Content-Length", blockBlob.Properties.Length.ToString());
HttpContext.Current.Response.BinaryWrite(ms.ToArray());
HttpContext.Current.Response.Flush();
HttpContext.Current.Response.Close();
}
}
I am trying to use StartCopyFromUri or StartCopyFromUriAsync to copy a blob from one storage account to another. Even though status.HasCompleted when I try to get the ETag either through
1. var etag = await _siteStorageClient.GetBlobETag(containerPath, asset.BlobName);
//this is the response from WaitForCompletionAsync
2. var etag = complete.GetRawResponse().Headers.Where(x => x.Name == "ETag").FirstOrDefault().Value;
I've tried both methods and both return an Etag that doesn't match what is shown in the properties of the blob when I log in through Azure Portal. It is almost as if the file wasn't done copying(or race condition) when the Etag check was executed. I couldn't find any usage samples on github for the SDK.
Any ideas what could be going awry?
This a similar question but using the older SDK. How to copy a blob from one container to another container using Azure Blob storage SDK
//Storage class
public async Task<CopyFromUriOperation> CopyFile(string containerName, string blobName, Uri sourceUri)
{
var container = _blobServiceClient.GetBlobContainerClient(containerName);
var blockBlobClient = container.GetBlockBlobClient(blobName);
//Made this the synchronous version try and block
//this is the target client
var status = await blockBlobClient.StartCopyFromUriAsync(sourceUri);
while(!status.HasCompleted)
{
//Per Documentation this calls UpdateStatusAsync() periodically
//until status.HasCompleted is true
await status.WaitForCompletionAsync();
}
return status;
}
//Calling Code
var status = await _siteStorageClient.CopyFile(container,BlobName, sasUri);
var etag = await _siteStorageClient.GetBlobETag(container, BlobName);
I was able to get this working after a few tries and troubleshooting. It would only happen in the Azure environment and not when running the web app locally.
Initially the status.WaitForCompletionAsync() was inside the loop and I started getting a socket error. I believe it was getting called too many times and was causing port exhaustion(just speculation at this point).
But this is what is working now.
public async Task<CopyFromUriOperation> CopyFile(string containerName, string blobName,Uri sourceUri)
{
var container = _blobServiceClient.GetBlobContainerClient(containerName);
var blockBlobClient = container.GetBlockBlobClient(blobName);
var status = await blockBlobClient.StartCopyFromUriAsync(sourceUri);
await status.WaitForCompletionAsync();
while(status.HasCompleted == false)
{
await Task.Delay(100);
}
return status;
}
i was trying to create an EPaper PDF viewer using react js as front-end and back-end asp.net core web api, here my API code that return pdf file.
[HttpGet]
public async Task<IActionResult> Get(int id)
{
var epaper = await _context.EPaperFilePages.FirstOrDefaultAsync(x =>
x.Id == id);
if (epaper == null)
{
return NotFound();
}
var inputPath = $"{_hostingEnvironment.WebRootPath}\\EPaper\\{epaper.PageFileName}";
MemoryStream stream = new MemoryStream();
using (FileStream fileStream = new FileStream(inputPath, FileMode.Open))
{
await fileStream.CopyToAsync(stream);
await fileStream.FlushAsync();
}
stream.Position = 0;
return File(stream, System.Net.Mime.MediaTypeNames.Application.Pdf);
}
when i use this api in browser or react code it was loading very slow minimum it was taking 1 min to load the file.
please suggest is there any way to load pdf file faster using any packages or how should i improve my code that loads the pdf file faster.
You can query like detail = Parking.joins(:cars).select('parkings.*, count(cars.parking_id) as Alloted, no_of_places - count(parking_id) as Remaining, cars.*, cars.id as car_id').group('place_name')
joins will do an inner join and leave off parking records where there are no cars.
Instead, can you do an includes? Or a .left_outer_joins
detail = Parking.left_outer_joins(:cars).
select('parking_id, place_name, no_of_places, count(parking_id) as Alloted, no_of_places - count(parking_id) as Remaining').
group('place_name')
I can't use the ServiceStack Client libraries and I've chosen to use the HttpClient PCL library instead. I can do all my Rest calls (and other json calls) without a problem, but I'm now stucked with uploading files.
A snippet of what I am trying to do:
var message = new HttpRequestMessage(restRequest.Method, restRequest.GetResourceUri(BaseUrl));
var content = new MultipartFormDataContent();
foreach (var file in files)
{
byte[] data;
bool success = CxFileStorage.TryReadBinaryFile(file, out data);
if (success)
{
var byteContent = new ByteArrayContent(data);
byteContent.Headers.ContentDisposition = new ContentDispositionHeaderValue("attachment")
{
FileName = System.IO.Path.GetFileName(file) ,
};
content.Add(byteContent);
}
}
message.Content = content;
Problem is now that I get a null reference exception (status 500) when posting. I doesn't get into the service. I see the call in the filterrequest, but that's it.
So I'm wondering what I do wrong and how I can pinpoint what is going wrong. How can I catch the correct error on the ServiceStack layer?
The web socket is written in javascript by my colleague. I managed to connect. First of all I have to log in on the application using a test account. I have to send the email and password through a json. I have installed the Json.Net packet using NuGet.
Some code that I found on my reaserch is this, but I do not understand how to send my data using that segment.
var buffer = new byte[1024];
var segment = new ArraySegment<byte>(buffer);
webSocket.SendAsync(segment, WebSocketMessageType.Text, true, CancellationToken.None);
Of course, I can use an object
User user=new User();
user.Email="bla#bla.com";
user.Password="pass";
string json = JsonConvert.SerializeObject(user);
But it will not be of any use because the method SendAsync accepts only byte type on segment.
All I want is to send that data, and if log in succeeds, I should receive other data (in Json format) about the user.
As a side note, I am quite new to web sockets, I used http protocols from ASP.NET WEB API 2.
I have no idea about Windows Phone 8, but by the code you pasted it seems similar to the regular .NET ClientWebSocket, so here you have some examples:
public static Task SendString(ClientWebSocket ws, String data, CancellationToken cancellation)
{
var encoded = Encoding.UTF8.GetBytes(data);
var buffer = new ArraySegment<Byte>(encoded, 0, encoded.Length);
return ws.SendAsync(buffer, WebSocketMessageType.Text, true, cancellation);
}
public static async Task<String> ReadString(ClientWebSocket ws)
{
ArraySegment<Byte> buffer = new ArraySegment<byte>(new Byte[8192]);
WebSocketReceiveResult result = null;
using (var ms = new MemoryStream())
{
do
{
result = await ws.ReceiveAsync(buffer, CancellationToken.None);
ms.Write(buffer.Array, buffer.Offset, result.Count);
}
while (!result.EndOfMessage);
ms.Seek(0, SeekOrigin.Begin);
using (var reader = new StreamReader(ms, Encoding.UTF8))
return reader.ReadToEnd();
}
}
If something does not compile or exists in WP8, just find an equivalent.
#vtortola is a working example in case your data comes in multiple segmented messages, but if all data comes in a single message you don't need all those streams to read the message, you just need to do this:
public static async Task<String> ReadString(ClientWebSocket socket)
{
var reciveBuffer = new byte[32000];
var result = await socket.ReceiveAsync(new ArraySegment<byte>(reciveBuffer), CancellationToken.None);
if (result.MessageType == WebSocketMessageType.Close)
{
await socket.CloseAsync(WebSocketCloseStatus.NormalClosure, string.Empty, CancellationToken.None);
}
return Encoding.ASCII.GetString(reciveBuffer , 0, result.Count);
}
If your message is splited in multiple segments or you don't know how your message is comming then you have to do like #vtortola
Also if you want to keep receiving messages you can do a while and call ReadString inside, like this:
while (socket.State == WebSocketState.Open)
{
var msg = ReadString(socket)
//do something with your message...
}