We Keep Coding

When loading data it gives an error " InvalidCastException: Specified cast is not valid "

e.g. I want to select a character and save his number
private Storage storage;
void Awake()
{
storage = new Storage();
}
public void SelectChar1()
{
numberChar = 1;//byte
storage.Save(DataPlayerSave);//save works fine
}
on awakening, the number is loaded
private DataPlayerSave dataPlayer;
private byte numberChar;
private void Awake()
{
dataPlayer = (DataPlayerSave)storage.Load(new DataPlayerSave());
numberChar = dataPlayer.numerChar;
}
I tried to divide the data into several parts and one large file, the result is almost always the same (sometimes everything works)
public class Storage()
{
public object Load(object saveDataByDefault)
{
filePath = Application.persistentDataPath + "/saves/GameSave.save";
if (!File.Exists(filePath))
{
if (saveDataByDefault != null)
{
Save(saveDataByDefault);
return saveDataByDefault;
}
}
var file = File.Open(filePath, FileMode.Open);
var saveData = formatter.Deserialize(file);
file.Close();
return saveData;
}
}
There are also similar classes that load data at the beginning of the scene. If there are 2 or more of them, then it gives an error, if 1, then everything works. I tried to set the sequence using the Coroutine did not help.
When loading data it gives an error "InvalidCastException: Specified cast is not valid."
dataPlayer = (DataPlayerSave)storage.Load(new DataPlayerSave());

if formatter is BinaryFormatter: please don't do that - it will hurt you; as for the exception: fundamentally, use a debugger and step through the code. In particular, if you say that the exception is coming from:
dataPlayer = (DataPlayerSave)storage.Load(new DataPlayerSave());
then we can assume that Load is not returning a DataPlayerSave. So: what is it? We can't tell you, but: you can find out:
var obj = storage.Load(new DataPlayerSave());
var type = obj.GetType(); // put a break-point here
Log(type.FullName); // or just log it
dataPlayer = (DataPlayerSave)obj;
and investigate what exactly obj is. Note that BinaryFormatter is very brittle as you change types (rename, move, refactor, etc) - but that isn't even the top reason not to use it.

BizTalk SMTP send mail with attachments (no orchestration, custom send pipeline component)

I hope someone can help me out with this since when I drop a JSON test file in my rcv-folder the only result is that in the BizTalk console (in 'Running instances'), the message only states 'Queued (Awaiting processing)'. I am not sure where my issue is in the code.
I am supposed to receive a JSON which will contain some info and possibly multiple attachments (in Base64 format) and then send the mail out (with attachments in correct format, i.e. PDF, txt, xls) to a certain email-address. One of the requirements is to not use an orchestration. But I am stuck and have no idea what I am doing anymore. What makes this question different from others is that I have no orchestration in my solution. Everything will be processed in a custom send pipeline component.
The sendpipeline (in encode stage) does contain my custom component and also the MIME/SMIME encoder. I am using the SMTP-adapter.
I have created the custom pipeline component with a wizard and my initial plan was in the form below:
Receive the file (in JSON-format. It will be transformed into XML for further processing). This will all be taken care of in my rcv-pipeline and this step already works for me.
Pick out all the necessary variables from XML needed to send in the e-mail. These variables are the ones I want to show in the E-mail text. Not sure how to explain it better.
Pick out all the attachments in base64, loop through them, convert to 'regular files' and then attach them to the mail (with correct filenames, extension etc.)
The XML looks like below:
ArchiveobjectsListErrands
- ArchiveobjectErrand
* UUID (Here are the variables I need to show in the E-mail. The plain text, so to say)
- ArchiveobjectListPaper
- Attachments
* Name
* Extension
* Size
* Base64String (Base64 string which will be needed to be fetched (in GetAttachments) and then processed in (ProcessAttachments))
The code I have is below:
public Microsoft.BizTalk.Message.Interop.IBaseMessage Execute(Microsoft.BizTalk.Component.Interop.IPipelineContext pContext, Microsoft.BizTalk.Message.Interop.IBaseMessage pInMsg)
{
// 1) Read file with XPathNavigator (https://learn.microsoft.com/en-us/dotnet/standard/data/xml/extract-xml-data-using-xpathnavigator)
XPathNavigator nav = ReadXmlFromMsgBox(pInMsg);
var outMsg = pContext.GetMessageFactory().CreateMessage();
outMsg.Context = PipelineUtil.CloneMessageContext(pInMsg.Context);
// 2) Pick out the necessary vars that the registrator requires
GetRegistratorProperties(nav, pContext, outMsg);
// 3) Read attachments
var attachments = GetAttachments(pInMsg, nav);
// 4) Processa attachments
ProcessAttachments(pContext, outMsg, attachments);
// 5) Send message along for further processing in the send pipeline
return outMsg;
}
private void GetRegistratorProperties(XPathNavigator _nav, IPipelineContext _pContext, IBaseMessage _msg)
{
var bodyPart = _pContext.GetMessageFactory().CreateMessagePart();
bodyPart.ContentType = "text/application";
bodyPart.PartProperties.Write("EmailBodyText", "http://schemas.microsoft.com/BizTalk/2003/smtp-properties", "EmailBodyText.");
bodyPart.PartProperties.Write("Subject", "http://schemas.microsoft.com/BizTalk/2003/smtp-properties", "Registratorsubject - Create errand");
_msg.AddPart("Body", bodyPart, true); // True for body but false for attachments
}
private void ProcessAttachments(IPipelineContext _pContext, IBaseMessage _msg, IList<Attachment> _attachments)
{
var msgPart = _pContext.GetMessageFactory().CreateMessagePart();
//outMsg.Context = PipelineUtil.CloneMessageContext(_msg.Context);
int i = 0;
foreach (var item in _attachments)
{
msgPart.PartProperties.Write("FileName", "http://schemas.microsoft.com/BizTalk/2003/mime-properties", item.filnamn+item.extension);
msgPart.PartProperties.Write("ContentDescription", "http://schemas.microsoft.com/BizTalk/2003/mime-properties", item.contentType);
msgPart.Data = new MemoryStream(BytesFromBase64String(item.base64));
//bodyPart.Charset = "utf-8";
msgPart.ContentType = item.contentType;
//_pInMsg.AddPart("Attachment part " + i.ToString(), bodyPart, false);
_msg.AddPart("Attachment part " + i.ToString(), msgPart, false);
i++;
}
}
private IList<Attachment> GetAttachments(IBaseMessage pInMsg, XPathNavigator _nav)
{
XPathNodeIterator iterator = _nav.Select("Path to attachments in xml");
IList<Attachment> myList = new List<Attachment>();
while (iterator.MoveNext())
{
XPathNavigator node = iterator.Current;
Attachment atttachments = new Attachment();
atttachments.filenamne = node.SelectSingleNode("Name").Value;
atttachments.extension = node.SelectSingleNode("Extension").Value;
atttachments.contentType = node.SelectSingleNode("Mimetype").Value;
atttachments.base64 = node.SelectSingleNode("Base64String").Value;
myList.Add(atttachments);
}
return myList;
}
private XPathNavigator ReadXmlFromMsgBox(IBaseMessage pInMsg)
{
// Using XPathNavigator to avoid creating a XMLDoc in memory
Stream originalMessage = pInMsg.BodyPart.GetOriginalDataStream();
XPathNavigator _navigator = new XPathDocument(originalMessage).CreateNavigator();
return _navigator;
}
[Serializable]
private class FileStreamFactory : IStreamFactory
{
byte[] _data;
public FileStreamFactory(byte[] data)
{
_data = data;
}
public Stream CreateStream()
{
return new MemoryStream(_data);
}
}
private static byte[] BytesFromBase64String(string msg)
{
return Convert.FromBase64String(msg);
}
#endregion
}
I can show some example file of the XML if deemed necessary. I avoided it due to brevity and also since it is quite large.
I would greatly appreciate if anyone could help out with how the code is supposed to look to achieve what is needed, a mail with some text and attachments named correctly regarding filename and extension.

Body text and Attachments are just different segments in a MIME encoded email, usually a plain text one is the first one.
If it has a status of 'Queued (Awaiting processing)', then it sounds like either
the host instance that is expecting to process it is not in a running state. Fix: Start the host instance.
the send port is not in a Started sate. Fix: set the send port to Started
or the send port has a service window set on it. Fix: disable the service window.

The endpoint for my Azure APIs (Translator Text, Computer Vision) is not accesible (error code 404)

I followed the Hololens Tutorials provided by Microsoft and encountered some problems regarding Azure services in 301 (https://learn.microsoft.com/en-us/windows/mixed-reality/mr-azure-301#chapter-4--setup-debug-canvas) and 302 (https://learn.microsoft.com/en-us/windows/mixed-reality/mr-azure-302). I think the problem has something to do with the endpoint, because the error code 404 appears when checking https://westeurope.api.cognitive.microsoft.com/.
I use the Hololens first generation, Unity 2017.4.27f1, HoloToolkit-Unity-Examples-2017.4.3.0 and Visual Studio Community 2017. I've tried some of the Hololens tutorials, which all worked fine with these versions. My code is the one provided by Microsoft for the Tutorials 301 and 302. I've changed the location in Azure from
'global' to 'Western Europe' by creating a new ressource group and have now https://westeurope.api.cognitive.microsoft.com/ as my endpoint, I also created new authentification keys for the APIs and updated them in my code.
Instead of myKey I inserted the first key of the respective API (Translator Text, Computer Vision). The code below is from Tutorial 302 and the class 'VisionManager.cs' (which seems to be the responsible one for the communication with the services of the Computer Vision API).
public static VisionManager instance;
// you must insert your service key here!
private string authorizationKey = "myKey";
private const string ocpApimSubscriptionKeyHeader = "Ocp-Apim-Subscription-Key";
private string visionAnalysisEndpoint = "https://westcentralus.api.cognitive.microsoft.com/vision/v2.0"; // This is where you need to update your endpoint, if you set your location to something other than west-us.
internal byte[] imageBytes;
internal string imagePath;
/// <summary>
/// Call the Computer Vision Service to submit the image.
/// </summary>
public IEnumerator AnalyseLastImageCaptured()
{
WWWForm webForm = new WWWForm();
using (UnityWebRequest unityWebRequest = UnityWebRequest.Post(visionAnalysisEndpoint, webForm))
{
// gets a byte array out of the saved image
imageBytes = GetImageAsByteArray(imagePath);
unityWebRequest.SetRequestHeader("Content-Type", "application/octet-stream");
unityWebRequest.SetRequestHeader(ocpApimSubscriptionKeyHeader, authorizationKey);
// the download handler will help receiving the analysis from Azure
unityWebRequest.downloadHandler = new DownloadHandlerBuffer();
// the upload handler will help uploading the byte array with the request
unityWebRequest.uploadHandler = new UploadHandlerRaw(imageBytes);
unityWebRequest.uploadHandler.contentType = "application/octet-stream";
yield return unityWebRequest.SendWebRequest();
long responseCode = unityWebRequest.responseCode;
try
{
string jsonResponse = null;
jsonResponse = unityWebRequest.downloadHandler.text;
// The response will be in Json format
// therefore it needs to be deserialized into the classes AnalysedObject and TagData
AnalysedObject analysedObject = new AnalysedObject();
analysedObject = JsonUtility.FromJson<AnalysedObject>(jsonResponse);
if (analysedObject.tags == null)
{
Debug.Log("analysedObject.tagData is null");
}
else
{
Dictionary<string, float> tagsDictionary = new Dictionary<string, float>();
foreach (TagData td in analysedObject.tags)
{
TagData tag = td as TagData;
tagsDictionary.Add(tag.name, tag.confidence);
}
ResultsLabel.instance.SetTagsToLastLabel(tagsDictionary);
}
}
catch (Exception exception)
{
Debug.Log("Json exception.Message: " + exception.Message);
}
yield return null;
}
}
[System.Serializable]
public class TagData
{
public string name;
public float confidence;
}
[System.Serializable]
public class AnalysedObject
{
public TagData[] tags;
public string requestId;
public object metadata;
}
// Use this for initialization
void Start () {
}
private void Awake()
{
// allows this instance to behave like a singleton
instance = this;
}
// Update is called once per frame
void Update () {
}
/// <summary>
/// Returns the contents of the specified file as a byte array.
/// </summary>
private static byte[] GetImageAsByteArray(string imageFilePath)
{
FileStream fileStream = new FileStream(imageFilePath, FileMode.Open, FileAccess.Read);
BinaryReader binaryReader = new BinaryReader(fileStream);
return binaryReader.ReadBytes((int)fileStream.Length);
}
I expected the output to be like the examples in the tutorials, but instead I got the same GUI results without any updates. In 301 I could see the same screen, but the text field 'You just said' and the 'Translation' never updated when I spoke. In 302 I could airtap and make a screenshot and could see at first 'Analyzing' and later on 'I see:' but without any results of detected objects. So it seems to me, that the Azure APIs services can't be contacted and I don't get any analysis data in return (probably you notice, that I'm quite new to the whole topic, so maybe my assumption is wrong).

C# Sockets send/receive problems and questions

Good morning all,
First of all apologies for the somewhat generic title. If in the course of this message I can come up with something a little more detailed, I will certainly change it.
I am working on a project that will contain 3 programs. The purpose is to be able to send a notification to all clients connected. For this there is a server, client, and console program.
The message itself will be an RTF file, but the notification also requires a sending department (string), and a display timer (TimeSpan).
Most of the project is done. The logic is mostly complete, it is multithreaded, all the classes are ready, and most tests work.
The problem I am having is that the server seems to not receive the data from the console in the right order. Thereby causing all sorts of problems.
The process from console to server is as follows:
Console first selects all relevant information for the notification:
Users (List, gathered from LDAP)
Expiration date and time (DateTime)
Time before close is allowed (long, amount of ticks)
Department (string)
RTF file
The server is already active, and has a separate thread for the console connections
Console connects to server
Server creates a separate thread to handle the console connection
Console sends the username of the logged on user to the server
Server checks whether user is allowed to create notification (for now it will always return true)
Console sends all the relevant information in the following order:
MessageID (string)
RecipientCount (number of users, used to loop appropriately from server side)
Recipients (foreach loop of List)
Department (string)
VisibleTime (long)
Expiration (DateTime)
and finally RTF file
I have used System.Diagnostics.Trace to check whether all information is sent correctly, and in the right order. This all checks out. But the thing is that roughly 75% of the time the server side seems to receive the RTF file at the moment it should be receiving the visibleTime.
The code is as follows:
private void SendMessage()
{
SendToServer(Environment.UserName);
if (bool.Parse(ReadFromServer()))
{
// User is allowed, continue
string messageID = DateTime.Now.ToUniversalTime().Ticks.ToString();
SendToServer(messageID); // MessageID
string recipientCount = lvRecipients.Items.Count.ToString();
SendToServer(lvRecipients.Items.Count.ToString()); // Amount of recipients
foreach (string item in lvRecipients.Items) // Loop to send each recipient
{
SendToServer(item);
}
string department = TB_Department.Text;
SendToServer(department); // Send department string
string visibleTime = TimeSpan.FromSeconds(SLIDER_VisibleTime.Value).Ticks.ToString();
SendToServer(visibleTime); // Send message visibility time
string expiration = DateTime.Now.ToUniversalTime().AddMinutes(2).ToString();
SendToServer(expiration); //TODO add UI control for this
SendRTFToServer(); // Send RTF file
MessageBox.Show(
"Your designated MessageID is: " + messageID + Environment.NewLine +
"Message upload is succesful.",
"Complete",
MessageBoxButton.OK);
}
else
{
// User is not allowed. Report to user. Disconnect (will be managed by the finally block)
MessageBox.Show(
"You are not allowed to upload messages to the server.",
"Access denied",
MessageBoxButton.OK,
MessageBoxImage.Stop);
return;
}
}
private void SendToServer(string toSend)
{
StreamWriter writer = new StreamWriter(server.GetStream());
writer.WriteLine(toSend);
writer.Flush();
}
private void SendRTFToServer()
{
StreamReader rtfFile = new StreamReader(File.Open(RTFLocation, FileMode.Open, FileAccess.Read));
StreamWriter sw = new StreamWriter(server.GetStream());
sw.Write(rtfFile.ReadToEnd());
sw.Flush();
server.GetStream().Flush();
}
private string ReadFromServer()
{
server.GetStream().Flush();
StreamReader reader = new StreamReader(server.GetStream());
return reader.ReadLine();
}
And from the server:
private void Connect()
{
string username = ReadFromConsole();
if (IsUserAllowed(username)) // Receive username
SendToConsole(bool.TrueString); // Send confirmation
else
{
SendToConsole(bool.FalseString); // Send denial
console.Close();
return;
}
string messageID = ReadFromConsole(); // Receive MessageID
string recipientCount = ReadFromConsole();
int numOfRecipients = int.Parse(recipientCount); // Receive and parse number of recipients
List<string> recipients = new List<string>();
for (int i = 0; i < numOfRecipients; i++)
{
string recipient = ReadFromConsole();
recipients.Add(recipient); // Receive recipient, add to list (required for Message)
}
string department = ReadFromConsole(); // Receive department string
string visibleTime = ReadFromConsole();
string expiration = ReadFromConsole();
StoreRTF(messageID); // Receive and store RTF file
console.Close(); // Connection is done, close
Message message = new Message(messageID, department, recipients, visibleTime, expiration);
}
private void SendToConsole(string toSend)
{
// Open client stream, and write information to it.
StreamWriter writer = new StreamWriter(console.GetStream());
writer.WriteLine(toSend);
writer.Flush();
}
private string ReadFromConsole()
{
// Read information from client stream
StreamReader reader = new StreamReader(console.GetStream());
return reader.ReadLine();
}
private void StoreRTF(string messageID)
{
// Check/create folder for Message storage
string messageFolder = Environment.GetFolderPath(Environment.SpecialFolder.ApplicationData) + #"\BMNotify\";
if (!Directory.Exists(messageFolder))
Directory.CreateDirectory(messageFolder);
// Create file to store message in
Stream rtfFile = File.Create(messageFolder + messageID + ".rtf");
// Store information from stream, and close resources
console.GetStream().CopyTo(rtfFile);
rtfFile.Close();
rtfFile.Dispose();
}
And the message class:
public class Message
{
internal string messageID;
internal string department;
internal List<string> recipients;
internal TimeSpan visibleAtLeast;
internal DateTime messageExpiration;
private static List<Message> allMessages; // Will hold te collection of Message's
public Message(string _messageID, string _department, List<string> _recipients, string visibleTime, string expiration)
{
messageID = _messageID;
recipients = _recipients;
department = _department;
visibleAtLeast = TimeSpan.FromTicks(long.Parse(visibleTime));
messageExpiration = DateTime.Parse(expiration);
if (allMessages == null)
allMessages = new List<Message>(); // Initialize if required
allMessages.Add(this);
}
internal Stream GetRTF()
{
return File.Open
(Environment.GetFolderPath(Environment.SpecialFolder.ApplicationData) + #"\BMNotify\" + messageID + ".rtf",
FileMode.Open,
FileAccess.Read,
FileShare.Read);
}
static public List<Message> AllMessages()
{
if (allMessages == null)
allMessages = new List<Message>(); // Initialize if required
return allMessages;
}
static public void RemoveMessage(Message message)
{
allMessages.Remove(message);
}
}
If anyone could shed any light on this, or tell me what I should change.. Or basically anything that could get me going again, I would be very grateful!

Your problem may stem from the fact that you're using a StreamReader to read data from the connection. Internally StreamReader buffers data it reads from the underlying Stream. You are creating a new StreamReader every time you attempt to read from the connection, and then discarding it once you've read a single line. In doing so, you're also discarding any data read from the connection that was buffered by the StreamReader (which may have constituted all or part of the following fields).
You should probably attempt to create a single StreamReader on the network stream, and use that same instance for all reads.

code contracts usage with exceptions

void ReadContent(string path)
{
Contract.Requires(path!=null);
string contentofileasstring = filehelperobj.GetContent(path);
if(String.IsNullOrEmpty(contentofileasstring ))
{
throw new FileContentException(path + "No content found");
}
m_xmlobj = contentofileasstring ;
}
Is my assumption of the usage of code contracts and exceptions right in this case. Do you think it is logical to replace the exception with a code contract(or vice versa)?
code not tested.Just an example scenario

I would probably go for an implementation which looks like the following:
private void ReadContent(string path)
{
Contract.Requires<FileMissingException>(File.Exists(path));
string content = filehelperobj.GetContent(path);
m_xmlobj = content;
}
Post Edit
As it's the content you want to validate, I would put a Contract.Ensures(!String.IsNullOrEmpty(Contract.Result<string>())); inside the filehelperobj.GetContent(string) method. Then if the content being read was null or empty, I would throw an exception. e.g.
public string GetContent(string path)
{
Contract.Requires<FileMissingException>(File.Exists(path));
Contract.Ensures(!String.IsNullOrEmpty(Contract.Result<string>()));
using(var reader = new StreamReader(File.OpenRead(path)))
{
var content = reader.ReadToEnd();
if(String.IsNullOrEmpty(content))
throw new FileContentException("No content found at file: " + path);
return content;
}
}

Well assuming you had the lines the wrong way round (ie, test the path for null before trying to use it) then yes, it is a valid pre-condition and therefore should be a code contract.