Search Results

Search found 68 results on 3 pages for 'ulong'.

Page 3/3 | < Previous Page | 1 2 3 

• Clone existing structs with different alignment in Visual C++

  - by Crend King

  Is there a way to clone an existing struct with different member alignment in Visual C++? Here is the background: I use an 3rd-party library, which uses several structs. To fill up the structs, I pass the address of the struct instances to some functions. Unfortunately, the functions only returns unaligned buffer, so that data of some members are always wrong. /Zp is out of choice, since it breaks the other parts of the program. I know #pragma pack modifies the alignment of the following struct, but I would like to avoid copying the structs into my code, for the definitions in the library might change in the future. Sample code: test.h: struct am_aligned { BYTE data1[10]; ULONG data2; }; test.cpp: #include "test.h" // typedef alignment(1) struct am_aligned am_unaligned; int APIENTRY wWinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPTSTR lpCmdLine, int nCmdShow) { char buffer[20] = {}; for (int i = 0; i < sizeof(unaligned_struct); i++) { buffer[i] = i; } am_aligned instance = *(am_aligned*) buffer; return 0; } Consider am_aligned is defined in the library header file. am_unaligned is my custom declaration, and only effective in test.cpp. The commented line does not work of course. instance.data2 is 0x0f0e0d0c, while 0x0d0c0b0a is desired. Thanks for help!

  Read the article

• C# struct with object as data member

  - by source-energy

  As we know, in C# structs are passed by value, not by reference. So if I have a struct with the following data members: private struct MessageBox { // data members private DateTime dm_DateTimeStamp; // a struct type private TimeSpan dm_TimeSpanInterval; // also a struct private ulong dm_MessageID; // System.Int64 type, struct private String dm_strMessage; // an object (hence a reference is stored here) // more methods, properties, etc ... } So when a MessageBox is passed as a parameter, a COPY is made on the stack, right? What does that mean in terms of how the data members are copied? The first two are struct types, so copies should be made of DateTime and TimeSpan. The third type is a primitive, so it's also copied. But what about the dm_strMessage, which is a reference to an object? When it's copied, another reference to the same String is created, right? The object itself resides in the heap, and is NOT copied (there is only one instance of it on the heap.) So now we have to references to the same object of type String. If the two references are accessed from different threads, it's conceivable that the String object could be corrupted by being modified from two different directions simultaneously. The MSDN documentation says that System.String is thread safe. Does that mean that the String class has a built-in mechanism to prevent an object being corrupted in exactly the type of situation described here? I'm trying to figure out if my MessageBox struct has any potential flaws / pitfalls being a structure vs. a class. Thanks for any input. Source.Energy.

  Read the article

• Modify existing struct alignment in Visual C++

  - by Crend King

  Is there a way to modify the member alignment of an existing struct in Visual C++? Here is the background: I use an 3rd-party library, which uses several structs. To fill up the structs, I pass the address of the struct instance to some functions. Unfortunately, the functions only returns unaligned buffer, so that data of some members are always wrong. /Zp is out of choice, since it breaks the other parts of the program. I know #pragma pack modifies the alignment of the following struct, but I do not want to copy the structs into my code, for the definitions in the library might change in the future. Sample code: test.h: struct am_aligned { BYTE data1[10]; ULONG data2; }; test.cpp: include "test.h" // typedef alignment(1) struct am_aligned am_unaligned int APIENTRY wWinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPTSTR lpCmdLine, int nCmdShow) { char buffer[20] = {}; for (int i = 0; i < sizeof(unaligned_struct); i++) { buffer[i] = i; } am_aligned instance = *(am_aligned*) buffer; return 0; } instance.data2 is 0x0f0e0d0c, while 0x0d0c0b0a is desired. The commented line does not work of course. Thanks for help!

  Read the article

• ERROR 2019 Linker Error Visual Studio

  - by Corrie Duck

  Hey I hope someone can tell me know to fix this issue I am having i keep getting an error 2019 from Visual studio for the following file. Now most of the functions have been removed so excuse the empty varriables etc. Error error LNK2019: unresolved external symbol "void * __cdecl OpenOneDevice(void *,struct _SP_DEVICE_INTERFACE_DATA *,char *)" (?OpenOneDevice@@YAPAXPAXPAU_SP_DEVICE_INTERFACE_DATA@@PAD@Z) referenced in function _wmain c:\Users\K\documents\visual studio 2010\Projects\test2\test2\test2.obj test2 #include "stdafx.h" #include <windows.h> #include <setupapi.h> SP_DEVICE_INTERFACE_DATA deviceInfoData; HDEVINFO hwDeviceInfo; HANDLE hOut; char *devName; // HANDLE OpenOneDevice(IN HDEVINFO hwDeviceInfo,IN PSP_DEVICE_INTERFACE_DATA DeviceInfoData,IN char *devName); // HANDLE OpenOneDevice(IN HDEVINFO HardwareDeviceInfo,IN PSP_DEVICE_INTERFACE_DATA DeviceInfoData,IN char *devName) { PSP_DEVICE_INTERFACE_DETAIL_DATA functionClassDeviceData = NULL; ULONG predictedLength = 0, requiredLength = 0; HANDLE hOut = INVALID_HANDLE_VALUE; SetupDiGetDeviceInterfaceDetail(HardwareDeviceInfo, DeviceInfoData, NULL, 0, &requiredLength, NULL); predictedLength = requiredLength; functionClassDeviceData = (PSP_DEVICE_INTERFACE_DETAIL_DATA)malloc(predictedLength); if(NULL == functionClassDeviceData) { return hOut; } functionClassDeviceData->cbSize = sizeof (SP_DEVICE_INTERFACE_DETAIL_DATA); if (!SetupDiGetDeviceInterfaceDetail(HardwareDeviceInfo, DeviceInfoData, functionClassDeviceData, predictedLength, &requiredLength, NULL)) { free( functionClassDeviceData ); return hOut; } //strcpy(devName,functionClassDeviceData->DevicePath) ; hOut = CreateFile(functionClassDeviceData->DevicePath, GENERIC_READ | GENERIC_WRITE, FILE_SHARE_READ | FILE_SHARE_WRITE, NULL, OPEN_EXISTING, 0, NULL); free(functionClassDeviceData); return hOut; } // int _tmain(int argc, _TCHAR* argv[]) { hOut = OpenOneDevice (hwDeviceInfo, &deviceInfoData, devName); if(hOut != INVALID_HANDLE_VALUE) { // error report } return 0; } Been driving me mad for hours. Any help appreciated. SOLVED THANKS TO CHRIS :-) Add #pragma comment (lib, "Setupapi.lib") Thanks

  Read the article

• Getting the full-name of the current user, returns an empty string (C#/C++)

  - by Nir

  I try to get the full-name of the current log-in user (Fullname, not username). The following code C#, C++ works fine but on XP computers not connected to the Net, I get empty string as result if I run it ~20 minutes after login (It runs OK whithin the first ~20 minutes after login) A Win32 API (GetUserNameEx) is used rather that PrincipalContext since it PrincipalContext may takes up to 15 seconds when working offline. Any Help why am I getting an empty string as result though a user full name is specified??? - C# Code public static string CurrentUserFullName { get { const int EXTENDED_NAME_FORMAT_NAME_DISPLAY = 3; StringBuilder userName = new StringBuilder(256); uint length = (uint) userName.Capacity; string ret; if (GetUserNameEx(EXTENDED_NAME_FORMAT_NAME_DISPLAY, userName, ref length)) { ret = userName.ToString(); } else { int errorCode = Marshal.GetLastWin32Error(); throw new Win32Exception("GetUserNameEx Failed. Error code - " + errorCode); } return ret; } } [DllImport("Secur32.dll", CharSet = CharSet.Auto, SetLastError = true)] private static extern bool GetUserNameEx(int nameFormat, StringBuilder lpNameBuffer, ref uint lpnSize); - Code in C++ #include "stdafx.h" #include <windows.h> #define SECURITY_WIN32 #include <Security.h> #pragma comment( lib, "Secur32.lib" ) int _tmain(int argc, _TCHAR* argv[]) { char szName[100]; ULONG nChars = sizeof( szName ); if ( GetUserNameEx( NameDisplay, szName, &nChars ) ) { printf( "Name: %s\n", szName); } else { printf( "Failed to GetUserNameEx\n" ); printf( "%d\n", GetLastError() ); } return 0; }

  Read the article

• What's the deal with the hidden Throw when catching a ThreadAbortException?

  - by priehl

  I'm going through a book of general c# development, and I've come to the thread abort section. The book says something along the lines that when you call Thread.Abort() on another thread, that thread will throw a ThreadAbortException, and even if you tried to supress it it would automatically rethrow it, unless you did some bs that's generally frowned upon. Here's the simple example offered. using System; using System.Threading; public class EntryPoint { private static void ThreadFunc() { ulong counter = 0; while (true) { try { Console.WriteLine("{0}", counter++); } catch (ThreadAbortException) { // Attempt to swallow the exception and continue. Console.WriteLine("Abort!"); } } } static void Main() { try { Thread newThread = new Thread(new ThreadStart(EntryPoint.ThreadFunc)); newThread.Start(); Thread.Sleep(2000); // Abort the thread. newThread.Abort(); // Wait for thread to finish. newThread.Join(); } catch (Exception e) { Console.WriteLine(e.ToString()); } } } The book says: When your thread finishes processing the abort exception, the runtime implicitly rethrows it at the end of your exception handler. It’s the same as if you had rethrown the exception yourself. Therefore, any outer exception handlers or finally blocks will still execute normally. In the example, the call to Join won’t be waiting forever as initially expected. So i wrapped a try catch around the Thread.Abort() call and set a break point, expecting it to hit this, considering the text says "any outer exception handlers or finally blocks will still execute normally". BUT IT DOES NOT. I'm racking my brain to figure out why. Anyone have any thoughts on why this isn't the case? Is the book wrong? Thanks in advance.

  Read the article

• ISAPI filter with LDAP over SSL only works as administrator

  - by Zac

  I have created an ISAPI filter for IIS 6.0 that tries to authenticate against Active directory using LDAP. The filter works fine when authenticating regularly over port 389, but when I try to use SSL, I always get the 0x51 Server Down error at the ldap_connect() call. Even skipping the connect call and using ldap_simple_bind_s() results in the same error. The weird thing is that if I change the app pool identity to the local admin account, then the filter works fine and LDAP over SSL is successful. I created an exe with the same code below and ran it on the server as admin and it works. Using the default NETWORK SERVICE identity for the site's app pool is what seems to be the problem. Any thoughts as to what is happening? I want to use the default identity since I don't want the website to have elevated admin privileges. The server is in a DMZ outside the network and domain where our DCs are that run AD. We have a valid certificate on our DCs for AD as well. Code: // Initialize LDAP connection LDAP * ldap = ldap_sslinit(servers, LDAP_SSL_PORT, 1); ULONG version = LDAP_VERSION3; if (ldap == NULL) { strcpy(error_msg, ldap_err2string(LdapGetLastError())); valid_user = false; } else { // Set LDAP options ldap_set_option(ldap, LDAP_OPT_PROTOCOL_VERSION, (void *) &version); ldap_set_option(ldap, LDAP_OPT_SSL, LDAP_OPT_ON); // Make the connection ldap_response = ldap_connect(ldap, NULL); // <-- Error occurs here! // Bind and continue... } UPDATE: I created a new user without admin privileges and ran the test exe as the new user and I got the same Server Down error. I added the user to the Administrators group and got the same error as well. The only user that seems to work with LDAP over SSL authentication on this particular server is administrator. The web server with the ISAPI filter (and where I've been running the test exe) is running Windows Server 2003. The DCs with AD on them are running 2008 R2. Also worth mentioning, we have a WordPress site on the same server that authenticates against LDAP over SSL using PHP (OpenLDAP) and there's no problem there. I have an ldap.conf file that specifies TLS_REQCERT never and the user running the PHP code is IUSR.

  Read the article

• Creating packages in code – Execute SQL Task

  The Execute SQL Task is for obvious reasons very well used, so I thought if you are building packages in code the chances are you will be using it. Using the task basic features of the task are quite straightforward, add the task and set some properties, just like any other. When you start interacting with variables though it can be a little harder to grasp so these samples should see you through. Some of these more advanced features are explained in much more detail in our ever popular post The Execute SQL Task, here I’ll just be showing you how to implement them in code. The abbreviated code blocks below demonstrate the different features of the task. The complete code has been encapsulated into a sample class which you can download (ExecSqlPackage.cs). Each feature described has its own method in the sample class which is mentioned after the code block. This first sample just shows adding the task, setting the basic properties for a connection and of course an SQL statement. Package package = new Package(); // Add the SQL OLE-DB connection ConnectionManager sqlConnection = AddSqlConnection(package, "localhost", "master"); // Add the SQL Task package.Executables.Add("STOCK:SQLTask"); // Get the task host wrapper TaskHost taskHost = package.Executables[0] as TaskHost; // Set required properties taskHost.Properties["Connection"].SetValue(taskHost, sqlConnection.ID); taskHost.Properties["SqlStatementSource"].SetValue(taskHost, "SELECT * FROM sysobjects"); For the full version of this code, see the CreatePackage method in the sample class. The AddSqlConnection method is a helper method that adds an OLE-DB connection to the package, it is of course in the sample class file too. Returning a single value with a Result Set The following sample takes a different approach, getting a reference to the ExecuteSQLTask object task itself, rather than just using the non-specific TaskHost as above. Whilst it means we need to add an extra reference to our project (Microsoft.SqlServer.SQLTask) it makes coding much easier as we have compile time validation of any property and types we use. For the more complex properties that is very valuable and saves a lot of time during development. The query has also been changed to return a single value, one row and one column. The sample shows how we can return that value into a variable, which we also add to our package in the code. To do this manually you would set the Result Set property on the General page to Single Row and map the variable on the Result Set page in the editor. Package package = new Package(); // Add the SQL OLE-DB connection ConnectionManager sqlConnection = AddSqlConnection(package, "localhost", "master"); // Add the SQL Task package.Executables.Add("STOCK:SQLTask"); // Get the task host wrapper TaskHost taskHost = package.Executables[0] as TaskHost; // Add variable to hold result value package.Variables.Add("Variable", false, "User", 0); // Get the task object ExecuteSQLTask task = taskHost.InnerObject as ExecuteSQLTask; // Set core properties task.Connection = sqlConnection.Name; task.SqlStatementSource = "SELECT id FROM sysobjects WHERE name = 'sysrowsets'"; // Set single row result set task.ResultSetType = ResultSetType.ResultSetType_SingleRow; // Add result set binding, map the id column to variable task.ResultSetBindings.Add(); IDTSResultBinding resultBinding = task.ResultSetBindings.GetBinding(0); resultBinding.ResultName = "id"; resultBinding.DtsVariableName = "User::Variable"; For the full version of this code, see the CreatePackageResultVariable method in the sample class. The other types of Result Set behaviour are just a variation on this theme, set the property and map the result binding as required. Parameter Mapping for SQL Statements This final example uses a parameterised SQL statement, with the coming from a variable. The syntax varies slightly between connection types, as explained in the Working with Parameters and Return Codes in the Execute SQL Taskhelp topic, but OLE-DB is the most commonly used, for which a question mark is the parameter value placeholder. Package package = new Package(); // Add the SQL OLE-DB connection ConnectionManager sqlConnection = AddSqlConnection(package, ".", "master"); // Add the SQL Task package.Executables.Add("STOCK:SQLTask"); // Get the task host wrapper TaskHost taskHost = package.Executables[0] as TaskHost; // Get the task object ExecuteSQLTask task = taskHost.InnerObject as ExecuteSQLTask; // Set core properties task.Connection = sqlConnection.Name; task.SqlStatementSource = "SELECT id FROM sysobjects WHERE name = ?"; // Add variable to hold parameter value package.Variables.Add("Variable", false, "User", "sysrowsets"); // Add input parameter binding task.ParameterBindings.Add(); IDTSParameterBinding parameterBinding = task.ParameterBindings.GetBinding(0); parameterBinding.DtsVariableName = "User::Variable"; parameterBinding.ParameterDirection = ParameterDirections.Input; parameterBinding.DataType = (int)OleDBDataTypes.VARCHAR; parameterBinding.ParameterName = "0"; parameterBinding.ParameterSize = 255; For the full version of this code, see the CreatePackageParameterVariable method in the sample class. You’ll notice the data type has to be specified for the parameter IDTSParameterBinding .DataType Property, and these type codes are connection specific too. My enumeration I wrote several years ago is shown below was probably done by reverse engineering a package and also the API header file, but I recently found a very handy post that covers more connections as well for exactly this, Setting the DataType of IDTSParameterBinding objects (Execute SQL Task). /// <summary> /// Enumeration of OLE-DB types, used when mapping OLE-DB parameters. /// </summary> private enum OleDBDataTypes { BYTE = 0x11, CURRENCY = 6, DATE = 7, DB_VARNUMERIC = 0x8b, DBDATE = 0x85, DBTIME = 0x86, DBTIMESTAMP = 0x87, DECIMAL = 14, DOUBLE = 5, FILETIME = 0x40, FLOAT = 4, GUID = 0x48, LARGE_INTEGER = 20, LONG = 3, NULL = 1, NUMERIC = 0x83, NVARCHAR = 130, SHORT = 2, SIGNEDCHAR = 0x10, ULARGE_INTEGER = 0x15, ULONG = 0x13, USHORT = 0x12, VARCHAR = 0x81, VARIANT_BOOL = 11 } Download Sample code ExecSqlPackage.cs (10KB)

  Read the article

• Creating packages in code – Execute SQL Task

  The Execute SQL Task is for obvious reasons very well used, so I thought if you are building packages in code the chances are you will be using it. Using the task basic features of the task are quite straightforward, add the task and set some properties, just like any other. When you start interacting with variables though it can be a little harder to grasp so these samples should see you through. Some of these more advanced features are explained in much more detail in our ever popular post The Execute SQL Task, here I’ll just be showing you how to implement them in code. The abbreviated code blocks below demonstrate the different features of the task. The complete code has been encapsulated into a sample class which you can download (ExecSqlPackage.cs). Each feature described has its own method in the sample class which is mentioned after the code block. This first sample just shows adding the task, setting the basic properties for a connection and of course an SQL statement. Package package = new Package(); // Add the SQL OLE-DB connection ConnectionManager sqlConnection = AddSqlConnection(package, "localhost", "master"); // Add the SQL Task package.Executables.Add("STOCK:SQLTask"); // Get the task host wrapper TaskHost taskHost = package.Executables[0] as TaskHost; // Set required properties taskHost.Properties["Connection"].SetValue(taskHost, sqlConnection.ID); taskHost.Properties["SqlStatementSource"].SetValue(taskHost, "SELECT * FROM sysobjects"); For the full version of this code, see the CreatePackage method in the sample class. The AddSqlConnection method is a helper method that adds an OLE-DB connection to the package, it is of course in the sample class file too. Returning a single value with a Result Set The following sample takes a different approach, getting a reference to the ExecuteSQLTask object task itself, rather than just using the non-specific TaskHost as above. Whilst it means we need to add an extra reference to our project (Microsoft.SqlServer.SQLTask) it makes coding much easier as we have compile time validation of any property and types we use. For the more complex properties that is very valuable and saves a lot of time during development. The query has also been changed to return a single value, one row and one column. The sample shows how we can return that value into a variable, which we also add to our package in the code. To do this manually you would set the Result Set property on the General page to Single Row and map the variable on the Result Set page in the editor. Package package = new Package(); // Add the SQL OLE-DB connection ConnectionManager sqlConnection = AddSqlConnection(package, "localhost", "master"); // Add the SQL Task package.Executables.Add("STOCK:SQLTask"); // Get the task host wrapper TaskHost taskHost = package.Executables[0] as TaskHost; // Add variable to hold result value package.Variables.Add("Variable", false, "User", 0); // Get the task object ExecuteSQLTask task = taskHost.InnerObject as ExecuteSQLTask; // Set core properties task.Connection = sqlConnection.Name; task.SqlStatementSource = "SELECT id FROM sysobjects WHERE name = 'sysrowsets'"; // Set single row result set task.ResultSetType = ResultSetType.ResultSetType_SingleRow; // Add result set binding, map the id column to variable task.ResultSetBindings.Add(); IDTSResultBinding resultBinding = task.ResultSetBindings.GetBinding(0); resultBinding.ResultName = "id"; resultBinding.DtsVariableName = "User::Variable"; For the full version of this code, see the CreatePackageResultVariable method in the sample class. The other types of Result Set behaviour are just a variation on this theme, set the property and map the result binding as required. Parameter Mapping for SQL Statements This final example uses a parameterised SQL statement, with the coming from a variable. The syntax varies slightly between connection types, as explained in the Working with Parameters and Return Codes in the Execute SQL Taskhelp topic, but OLE-DB is the most commonly used, for which a question mark is the parameter value placeholder. Package package = new Package(); // Add the SQL OLE-DB connection ConnectionManager sqlConnection = AddSqlConnection(package, ".", "master"); // Add the SQL Task package.Executables.Add("STOCK:SQLTask"); // Get the task host wrapper TaskHost taskHost = package.Executables[0] as TaskHost; // Get the task object ExecuteSQLTask task = taskHost.InnerObject as ExecuteSQLTask; // Set core properties task.Connection = sqlConnection.Name; task.SqlStatementSource = "SELECT id FROM sysobjects WHERE name = ?"; // Add variable to hold parameter value package.Variables.Add("Variable", false, "User", "sysrowsets"); // Add input parameter binding task.ParameterBindings.Add(); IDTSParameterBinding parameterBinding = task.ParameterBindings.GetBinding(0); parameterBinding.DtsVariableName = "User::Variable"; parameterBinding.ParameterDirection = ParameterDirections.Input; parameterBinding.DataType = (int)OleDBDataTypes.VARCHAR; parameterBinding.ParameterName = "0"; parameterBinding.ParameterSize = 255; For the full version of this code, see the CreatePackageParameterVariable method in the sample class. You’ll notice the data type has to be specified for the parameter IDTSParameterBinding .DataType Property, and these type codes are connection specific too. My enumeration I wrote several years ago is shown below was probably done by reverse engineering a package and also the API header file, but I recently found a very handy post that covers more connections as well for exactly this, Setting the DataType of IDTSParameterBinding objects (Execute SQL Task). /// <summary> /// Enumeration of OLE-DB types, used when mapping OLE-DB parameters. /// </summary> private enum OleDBDataTypes { BYTE = 0x11, CURRENCY = 6, DATE = 7, DB_VARNUMERIC = 0x8b, DBDATE = 0x85, DBTIME = 0x86, DBTIMESTAMP = 0x87, DECIMAL = 14, DOUBLE = 5, FILETIME = 0x40, FLOAT = 4, GUID = 0x48, LARGE_INTEGER = 20, LONG = 3, NULL = 1, NUMERIC = 0x83, NVARCHAR = 130, SHORT = 2, SIGNEDCHAR = 0x10, ULARGE_INTEGER = 0x15, ULONG = 0x13, USHORT = 0x12, VARCHAR = 0x81, VARIANT_BOOL = 11 } Download Sample code ExecSqlPackage.cs (10KB)

  Read the article

• List of Commonly Used Value Types in XNA Games

  - by Michael B. McLaughlin

  Most XNA programmers are concerned about generating garbage. More specifically about allocating GC-managed memory (GC stands for “garbage collector” and is both the name of the class that provides access to the garbage collector and an acronym for the garbage collector (as a concept) itself). Two of the major target platforms for XNA (Windows Phone 7 and Xbox 360) use variants of the .NET Compact Framework. On both variants, the GC runs under various circumstances (Windows Phone 7 and Xbox 360). Of concern to XNA programmers is the fact that it runs automatically after a fixed amount of GC-managed memory has been allocated (currently 1MB on both systems). Many beginning XNA programmers are unaware of what constitutes GC-managed memory, though. So here’s a quick overview. In .NET, there are two different “types” of types: value types and reference types. Only reference types are managed by the garbage collector. Value types are not managed by the garbage collector and are instead managed in other ways that are implementation dependent. For purposes of XNA programming, the important point is that they are not managed by the GC and thus do not, by themselves, increment that internal 1 MB allocation counter. (n.b. Structs are value types. If you have a struct that has a reference type as a member, then that reference type, when instantiated, will still be allocated in the GC-managed memory and will thus count against the 1 MB allocation counter. Putting it in a struct doesn’t change the fact that it gets allocated on the GC heap, but the struct itself is created outside of the GC’s purview). Both value types and reference types use the keyword ‘new’ to allocate a new instance of them. Sometimes this keyword is hidden by a method which creates new instances for you, e.g. XmlReader.Create. But the important thing to determine is whether or not you are dealing with a value types or a reference type. If it’s a value type, you can use the ‘new’ keyword to allocate new instances of that type without incrementing the GC allocation counter (except as above where it’s a struct with a reference type in it that is allocated by the constructor, but there are no .NET Framework or XNA Framework value types that do this so it would have to be a struct you created or that was in some third-party library you were using for that to even become an issue). The following is a list of most all of value types you are likely to use in a generic XNA game: AudioCategory (used with XACT; not available on WP7) AvatarExpression (Xbox 360 only, but exposed on Windows to ease Xbox development) bool BoundingBox BoundingSphere byte char Color DateTime decimal double any enum (System.Enum itself is a class, but all enums are value types such that there are no GC allocations for enums) float GamePadButtons GamePadCapabilities GamePadDPad GamePadState GamePadThumbSticks GamePadTriggers GestureSample int IntPtr (rarely but occasionally used in XNA) KeyboardState long Matrix MouseState nullable structs (anytime you see, e.g. int? something, that ‘?’ denotes a nullable struct, also called a nullable type) Plane Point Quaternion Ray Rectangle RenderTargetBinding sbyte (though I’ve never seen it used since most people would just use a short) short TimeSpan TouchCollection TouchLocation TouchPanelCapabilities uint ulong ushort Vector2 Vector3 Vector4 VertexBufferBinding VertexElement VertexPositionColor VertexPositionColorTexture VertexPositionNormalTexture VertexPositionTexture Viewport So there you have it. That’s not quite a complete list, mind you. For example: There are various structs in the .NET framework you might make use of. I left out everything from the Microsoft.Xna.Framework.Graphics.PackedVector namespace, since everything in there ventures into the realm of advanced XNA programming anyway (n.b. every single instantiable thing in that namespace is a struct and thus a value type; there are also two interfaces but interfaces cannot be instantiated at all and thus don’t figure in to this discussion). There are so many enums you’re likely to use (PlayerIndex, SpriteSortMode, SpriteEffects, SurfaceFormat, etc.) that including them would’ve flooded the list and reduced its utility. So I went with “any enum” and trust that you can figure out what the enums are (and it’s rare to use ‘new’ with an enum anyway). That list also doesn’t include any of the pre-defined static instances of some of the classes (e.g. BlendState.AlphaBlend, BlendState.Opaque, etc.) which are already allocated such that using them doesn’t cause any new allocations and therefore doesn’t increase that 1 MB counter. That list also has a few misleading things. VertexElement, VertexPositionColor, and all the other vertex types are structs. But you’re only likely to ever use them as an array (for use with VertexBuffer or DynamicVertexBuffer), and all arrays are reference types (even arrays of value types such as VertexPositionColor[ ] or int[ ]). * So that’s it for now. The note below may be a bit confusing (it deals with how the GC works and how arrays are managed in .NET). If so, you can probably safely ignore it for now but feel free to ask any questions regardless. * Arrays of value types (where the value type doesn’t contain any reference type members) are much faster for the GC to examine than arrays of reference types, so there is a definite benefit to using arrays of value types where it makes sense. But creating arrays of value types does cause the GC’s allocation counter to increase. Indeed, allocating a large array of a value type is one of the quickest ways to increment the allocation counter since a .NET array is a sequential block of memory. An array of reference types is just a sequential block of references (typically 4 bytes each) while an array of value types is a sequential block of instances of that type. So for an array of Vector3s it would be 12 bytes each since each float is 4 bytes and there are 3 in a Vector3; for an array of VertexPositionNormalTexture structs it would typically be 32 bytes each since it has two Vector3s and a Vector2. (Note that there are a few additional bytes taken up in the creation of an array, typically 12 but sometimes 16 or possibly even more, which depend on the implementation details of the array type on the particular platform the code is running on).

  Read the article

• Synchronization requirements for FileStream.(Begin/End)(Read/Write)

  - by Doug McClean

  Is the following pattern of multi-threaded calls acceptable to a .Net FileStream? Several threads calling a method like this: ulong offset = whatever; // different for each thread byte[] buffer = new byte[8192]; object state = someState; // unique for each call, hence also for each thread lock(theFile) { theFile.Seek(whatever, SeekOrigin.Begin); IAsyncResult result = theFile.BeginRead(buffer, 0, 8192, AcceptResults, state); } if(result.CompletedSynchronously) { // is it required for us to call AcceptResults ourselves in this case? // or did BeginRead already call it for us, on this thread or another? } Where AcceptResults is: void AcceptResults(IAsyncResult result) { lock(theFile) { int bytesRead = theFile.EndRead(result); // if we guarantee that the offset of the original call was at least 8192 bytes from // the end of the file, and thus all 8192 bytes exist, can the FileStream read still // actually read fewer bytes than that? // either: if(bytesRead != 8192) { Panic("Page read borked"); } // or: // issue a new call to begin read, moving the offsets into the FileStream and // the buffer, and decreasing the requested size of the read to whatever remains of the buffer } } I'm confused because the documentation seems unclear to me. For example, the FileStream class says: Any public static members of this type are thread safe. Any instance members are not guaranteed to be thread safe. But the documentation for BeginRead seems to contemplate having multiple read requests in flight: Multiple simultaneous asynchronous requests render the request completion order uncertain. Are multiple reads permitted to be in flight or not? Writes? Is this the appropriate way to secure the location of the Position of the stream between the call to Seek and the call to BeginRead? Or does that lock need to be held all the way to EndRead, hence only one read or write in flight at a time? I understand that the callback will occur on a different thread, and my handling of state, buffer handle that in a way that would permit multiple in flight reads. Further, does anyone know where in the documentation to find the answers to these questions? Or an article written by someone in the know? I've been searching and can't find anything. Relevant documentation: FileStream class Seek method BeginRead method EndRead IAsyncResult interface

  Read the article

• How do I use IImgCtx to get load an image with an alpha channel?

  - by fret

  I have some working code that uses IImgCtx to load images, but I can't work out how to get at the alpha channel. For images like .gif's and .png's there are transparent pixels, but using anything other than a 24-bit bitmap as a drawing surface doesn't work. For reference on the interface: http://www.codeproject.com/KB/graphics/JianImgCtxDecoder.aspx My code looks like this: IImgCtx *Ctx = 0; HRESULT hr = CoCreateInstance(CLSID_IImgCtx, NULL, CLSCTX_INPROC_SERVER, IID_IImgCtx, (LPVOID*)&Ctx); if (SUCCEEDED(hr)) { GVariant Fn = Name; hr = Ctx->Load(Fn.WStr(), 0); if (SUCCEEDED(hr)) { SIZE Size = { -1, -1 }; ULONG State = 0; while (true) { hr = Ctx->GetStateInfo(&State, &Size, false); if (SUCCEEDED(hr)) { if ((State & IMGLOAD_COMPLETE) || (State & IMGLOAD_STOPPED) || (State & IMGLOAD_ERROR)) { break; } else { LgiSleep(1); } } else break; } if (Size.cx > 0 && Size.cy > 0 && pDC.Reset(new GMemDC)) { if (pDC->Create(Size.cx, Size.cy, 32)) { HDC hDC = pDC->StartDC(); if (hDC) { RECT rc = { 0, 0, pDC->X(), pDC->Y() }; Ctx->Draw(hDC, &rc); pDC->EndDC(); } } else pDC.Reset(); } } Ctx->Release(); Where "StartDC" basically wraps CreateCompatibleDC(NULL) and "EndDC" wraps DeleteDC, with appropriate SelectObjects for the HBITMAPS etc. And pDC-Create(x, y, bit_depth) calls CreateDIBSection(...DIB_RGB_COLORS...). So it works if I create a 24 bits/pixel bitmap but has no alpha to speak of, and it leaves the 32 bits/pixel bitmap blank. Now this interface apparently is used by Internet Explorer to load images, and obviously THAT supports transparency, so I believe that it's possible to get some level of alpha out of the interface. The question is how? (I also have fall back code that will call libpng/libjpeg/my .gif loader etc)

  Read the article

• How to dispose of a NET COM interop object on Release()

  - by mhenry1384

  I have a COM object written in managed code (C++/CLI). I am using that object in standard C++. How do I force my COM object's destructor to be called immediately when the COM object is released? If that's not possible, call I have Release() call a MyDispose() method on my COM object? My code to declare the object (C++/CLI): [Guid("57ED5388-blahblah")] [InterfaceType(ComInterfaceType::InterfaceIsIDispatch)] [ComVisible(true)] public interface class IFoo { void Doit(); }; [Guid("417E5293-blahblah")] [ClassInterface(ClassInterfaceType::None)] [ComVisible(true)] public ref class Foo : IFoo { public: void MyDispose(); ~Foo() {MyDispose();} // This is never called !Foo() {MyDispose();} // This is called by the garbage collector. virtual ULONG Release() {MyDispose();} // This is never called virtual void Doit(); }; My code to use the object (native C++): #import "..\\Debug\\Foo.tlb" ... Bar::IFoo setup(__uuidof(Bar::Foo)); // This object comes from the .tlb. setup.Doit(); setup-Release(); // explicit release, not really necessary since Bar::IFoo's destructor will call Release(). If I put a destructor method on my COM object, it is never called. If I put a finalizer method, it is called when the garbage collector gets around to it. If I explicitly call my Release() override it is never called. I would really like it so that when my native Bar::IFoo object goes out of scope it automatically calls my .NET object's dispose code. I would think I could do it by overriding the Release(), and if the object count = 0 then call MyDispose(). But apparently I'm not overriding Release() correctly because my Release() method is never called. Obviously, I can make this happen by putting my MyDispose() method in the interface and requiring the people using my object to call MyDispose() before Release(), but it would be slicker if Release() just cleaned up the object. Is it possible to force the .NET COM object's destructor, or some other method, to be called immediately when a COM object is released? Googling on this issue gets me a lot of hits telling me to call System.Runtime.InteropServices.Marshal.ReleaseComObject(), but of course, that's how you tell .NET to release a COM object. I want COM Release() to Dispose of a .NET object.

  Read the article

• Strange behavior with gcc inline assembly

  - by Chris

  When inlining assembly in gcc, I find myself regularly having to add empty asm blocks in order to keep variables alive in earlier blocks, for example: asm("rcr $1,%[borrow];" "movq 0(%[b_],%[i],8),%%rax;" "adcq %%rax,0(%[r_top],%[i],8);" "rcl $1,%[borrow];" : [borrow]"+r"(borrow) : [i]"r"(i),[b_]"r"(b_.data),[r_top]"r"(r_top.data) : "%rax","%rdx"); asm("" : : "r"(borrow) : ); // work-around to keep borrow alive ... Another example of weirdness is that the code below works great without optimizations, but with -O3 it seg-faults: ulong carry = 0,hi = 0,qh = s.data[1],ql = s.data[0]; asm("movq 0(%[b]),%%rax;" "mulq %[ql];" "movq %%rax,0(%[sb]);" "movq %%rdx,%[hi];" : [hi]"=r"(hi) : [ql]"r"(ql),[b]"r"(b.data),[sb]"r"(sb.data) : "%rax","%rdx","memory"); for (long i = 1; i < b.size; i++) { asm("movq 0(%[b],%[i],8),%%rax;" "mulq %[ql];" "xorq %%r10,%%r10;" "addq %%rax,%[hi];" "adcq %%rdx,%[carry];" "adcq $0,%%r10;" "movq -8(%[b],%[i],8),%%rax;" "mulq %[qh];" "addq %%rax,%[hi];" "adcq %%rdx,%[carry];" "adcq $0,%%r10;" "movq %[hi],0(%[sb],%[i],8);" "movq %[carry],%[hi];" "movq %%r10,%[carry];" : [carry]"+r"(carry),[hi]"+r"(hi) : [i]"r"(i),[ql]"r"(ql),[qh]"r"(qh),[b]"r"(b.data),[sb]"r"(sb.data) : "%rax","%rdx","%r10","memory"); } asm("movq -8(%[b],%[i],8),%%rax;" "mulq %[qh];" "addq %%rax,%[hi];" "adcq %%rdx,%[carry];" "movq %[hi],0(%[sb],%[i],8);" "movq %[carry],8(%[sb],%[i],8);" : [hi]"+r"(hi),[carry]"+r"(carry) : [i]"r"(long(b.size)),[qh]"r"(qh),[b]"r"(b.data),[sb]"r"(sb.data) : "%rax","%rdx","memory"); I think it has to do with the fact that it's using so many registers. Is there something I'm missing here or is the register allocation just really buggy with gcc inline assembly?

  Read the article

• Writing a managed wrapper for unmanaged (C++) code - custom types/structs

  - by Bobby

  faacEncConfigurationPtr FAACAPI faacEncGetCurrentConfiguration( faacEncHandle hEncoder); I'm trying to come up with a simple wrapper for this C++ library; I've never done more than very simple p/invoke interop before - like one function call with primitive arguments. So, given the above C++ function, for example, what should I do to deal with the return type, and parameter? FAACAPI is defined as: #define FAACAPI __stdcall faacEncConfigurationPtr is defined: typedef struct faacEncConfiguration { int version; char *name; char *copyright; unsigned int mpegVersion; unsigned long bitRate; unsigned int inputFormat; int shortctl; psymodellist_t *psymodellist; int channel_map[64]; } faacEncConfiguration, *faacEncConfigurationPtr; AFAIK this means that the return type of the function is a reference to this struct? And faacEncHandle is: typedef struct { unsigned int numChannels; unsigned long sampleRate; ... SR_INFO *srInfo; double *sampleBuff[MAX_CHANNELS]; ... double *freqBuff[MAX_CHANNELS]; double *overlapBuff[MAX_CHANNELS]; double *msSpectrum[MAX_CHANNELS]; CoderInfo coderInfo[MAX_CHANNELS]; ChannelInfo channelInfo[MAX_CHANNELS]; PsyInfo psyInfo[MAX_CHANNELS]; GlobalPsyInfo gpsyInfo; faacEncConfiguration config; psymodel_t *psymodel; /* quantizer specific config */ AACQuantCfg aacquantCfg; /* FFT Tables */ FFT_Tables fft_tables; int bitDiff; } faacEncStruct, *faacEncHandle; So within that struct we see a lot of other types... hmm. Essentially, I'm trying to figure out how to deal with these types in my managed wrapper? Do I need to create versions of these types/structs, in C#? Something like this: [StructLayout(LayoutKind.Sequential)] struct faacEncConfiguration { uint useTns; ulong bitRate; ... } If so then can the runtime automatically "map" these objects onto eachother? And, would I have to create these "mapped" types for all the types in these return types/parameter type hierarchies, all the way down until I get to all primitives? I know this is a broad topic, any advice on getting up-to-speed quickly on what I need to learn to make this happen would be very much appreciated! Thanks!

  Read the article

• HELP: MS Virtual Disk Service to Access Volumes and Discs on Local Machine.

  - by Jibran Ahmed

  Hi, here it is my code through which I am successfully initialize the VDS service and get the Packs but When I call QueryVolumes on IVdsPack Object, I am able to get IEnumVdsObjects but unable to get IUnknown* array through IEnumVdsObject::Next method, it reutrns S_FALSE with IUnkown* = NULL. So this IUnknown* cant be used to QueryInterface for IVdsVolume Below is my code HRESULT hResult; IVdsService* pService = NULL; IVdsServiceLoader *pLoader = NULL; //Launch the VDS Service hResult = CoInitialize(NULL); if( SUCCEEDED(hResult) ) { hResult = CoCreateInstance( CLSID_VdsLoader, NULL, CLSCTX_LOCAL_SERVER, IID_IVdsServiceLoader, (void**) &pLoader ); //if succeeded load VDS on local machine if( SUCCEEDED(hResult) ) pLoader->LoadService(NULL, &pService); //Done with Loader now release VDS Loader interface _SafeRelease(pLoader); if( SUCCEEDED(hResult) ) { hResult = pService->WaitForServiceReady(); if ( SUCCEEDED(hResult) ) { AfxMessageBox(L"VDS Service Loaded"); IEnumVdsObject* pEnumVdsObject = NULL; hResult = pService->QueryProviders(VDS_QUERY_SOFTWARE_PROVIDERS, &pEnumVdsObject); IUnknown* ppObjUnk ; IVdsSwProvider* pVdsSwProvider = NULL; IVdsPack* pVdsPack = NULL; IVdsVolume* pVdsVolume = NULL; ULONG ulFetched = 0; hResult = E_INVALIDARG; while(!SUCCEEDED(hResult)) { hResult = pEnumVdsObject->Next(1, &ppObjUnk, &ulFetched); hResult = ppObjUnk->QueryInterface(IID_IVdsSwProvider, (void**)&pVdsSwProvider); if(!SUCCEEDED(hResult)) _SafeRelease(ppObjUnk); } _SafeRelease(pEnumVdsObject); _SafeRelease(ppObjUnk); hResult = pVdsSwProvider->QueryPacks(&pEnumVdsObject); hResult = E_INVALIDARG; while(!SUCCEEDED(hResult)) { hResult = pEnumVdsObject->Next(1, &ppObjUnk, &ulFetched); hResult = ppObjUnk->QueryInterface(IID_IVdsPack, (void**)&pVdsPack); if(!SUCCEEDED(hResult)) _SafeRelease(ppObjUnk); } _SafeRelease(pEnumVdsObject); _SafeRelease(ppObjUnk); hResult = pVdsPack->QueryVolumes(&pEnumVdsObject); pEnumVdsObject->Reset(); hResult = E_INVALIDARG; ulFetched = 0; BOOL bDone = FALSE; while(!SUCCEEDED(hResult)) { hResult = pEnumVdsObject->Next(1, &ppObjUnk, &ulFetched); //hResult = ppObjUnk->QueryInterface(IID_IVdsVolume, (void**)&pVdsVolume); if(!SUCCEEDED(hResult)) _SafeRelease(ppObjUnk); } _SafeRelease(pEnumVdsObject); _SafeRelease(ppObjUnk); _SafeRelease(pVdsPack); _SafeRelease(pVdsSwProvider); // hResult = pVdsVolume-AddAccessPath(TEXT("G:\")); if(SUCCEEDED(hResult)) AfxMessageBox(L"Add Access Path Successfully"); else AfxMessageBox(L"Unable to Add access path"); //UUID of IVdsVolumeMF {EE2D5DED-6236-4169-931D-B9778CE03DC6} static const GUID GUID_IVdsVolumeMF = {0xEE2D5DED, 0x6236, 4169,{0x93, 0x1D, 0xB9, 0x77, 0x8C, 0xE0, 0x3D, 0XC6} }; hResult = pService->GetObject(GUID_IVdsVolumeMF, VDS_OT_VOLUME, &ppObjUnk); if(hResult == VDS_E_OBJECT_NOT_FOUND) AfxMessageBox(L"Object Not found"); if(hResult == VDS_E_INITIALIZED_FAILED) AfxMessageBox(L"Initialization failed"); // pVdsVolume = reinterpret_cast(ppObjUnk); if(SUCCEEDED(hResult)) { // hResult = pVdsVolume-AddAccessPath(TEXT("G:\")); if(SUCCEEDED(hResult)) { IVdsAsync* ppVdsSync; AfxMessageBox(L"Formatting is about to Start......"); // hResult = pVdsVolume-Format(VDS_FST_UDF, TEXT("UDF_FORMAT_TEST"), 2048, TRUE, FALSE, FALSE, &ppVdsSync); if(SUCCEEDED(hResult)) AfxMessageBox(L"Formatting Started......."); else AfxMessageBox(L"Formatting Failed"); } else AfxMessageBox(L"Unable to Add Access Path"); } _SafeRelease(pVdsVolume); } else { AfxMessageBox(L"VDS Service Cannot be Loaded"); } } } _SafeRelease(pService);

  Read the article

• How do I programmatically send email w/attachment to a known recipient using MAPI in C++? MAPISendM

  - by Tim

  This question is similar, but does not show how to add a recipient. How do I do both? We'd like the widest support possible for as many Windows platforms as possible (from XP and greater) We're using visual studio 2008 Essentially we want to send an email with: pre-filled destination address file attachment subject line from our program and give the user the ability to add any information or cancel it. EDIT I am trying to use MAPISendMail() I copied much of the code from the questions linked near the top, but I get no email dlg box and the error return I get from the call is: 0x000f - "The system cannot find the drive specified" If I comment out the lines to set the recipient, it works fine (of course then I have no recipient pre-filled in) Here is the code: #include <tchar.h> #include <windows.h> #include <mapi.h> #include <mapix.h> int _tmain( int argc, wchar_t *argv[] ) { HMODULE hMapiModule = LoadLibrary( _T( "mapi32.dll" ) ); if ( hMapiModule != NULL ) { LPMAPIINITIALIZE lpfnMAPIInitialize = NULL; LPMAPIUNINITIALIZE lpfnMAPIUninitialize = NULL; LPMAPILOGONEX lpfnMAPILogonEx = NULL; LPMAPISENDDOCUMENTS lpfnMAPISendDocuments = NULL; LPMAPISESSION lplhSession = NULL; LPMAPISENDMAIL lpfnMAPISendMail = NULL; lpfnMAPIInitialize = (LPMAPIINITIALIZE)GetProcAddress( hMapiModule, "MAPIInitialize" ); lpfnMAPIUninitialize = (LPMAPIUNINITIALIZE)GetProcAddress( hMapiModule, "MAPIUninitialize" ); lpfnMAPILogonEx = (LPMAPILOGONEX)GetProcAddress( hMapiModule, "MAPILogonEx" ); lpfnMAPISendDocuments = (LPMAPISENDDOCUMENTS)GetProcAddress( hMapiModule, "MAPISendDocuments" ); lpfnMAPISendMail = (LPMAPISENDMAIL)GetProcAddress( hMapiModule, "MAPISendMail" ); if ( lpfnMAPIInitialize && lpfnMAPIUninitialize && lpfnMAPILogonEx && lpfnMAPISendDocuments ) { HRESULT hr = (*lpfnMAPIInitialize)( NULL ); if ( SUCCEEDED( hr ) ) { hr = (*lpfnMAPILogonEx)( 0, NULL, NULL, MAPI_EXTENDED | MAPI_USE_DEFAULT, &lplhSession ); if ( SUCCEEDED( hr ) ) { // this opens the email client // create the msg. We need to add recipients AND subject AND the dmp file // file attachment MapiFileDesc filedesc; ::ZeroMemory(&filedesc, sizeof(filedesc)); filedesc.nPosition = (ULONG)-1; filedesc.lpszPathName = "E:\\Development\\Open\\testmail\\testmail.cpp"; // recipient(s) MapiRecipDesc recip; ::ZeroMemory(&recip, sizeof(recip)); recip.lpszName = "QA email"; recip.lpszAddress = "[email protected]"; // the message MapiMessage msg; ::ZeroMemory(&msg, sizeof(msg)); msg.lpszSubject = "Test"; msg.nRecipCount = 1; // if I comment out this line it works fine... msg.lpRecips = &recip; msg.nFileCount = 1; msg.lpFiles = &filedesc; hr = (*lpfnMAPISendMail)(0, NULL, &msg, MAPI_LOGON_UI|MAPI_DIALOG, 0); if ( SUCCEEDED( hr ) ) { hr = lplhSession->Logoff( 0, 0, 0 ); hr = lplhSession->Release(); lplhSession = NULL; } } } (*lpfnMAPIUninitialize)(); } FreeLibrary( hMapiModule ); } return 0; }

  Read the article

• How to capture live camera frames in RGB with DirectShow

  - by Jonny Boy

  I'm implementing live video capture through DirectShow for live processing and display. (Augmented Reality app). I can access the pixels easily enough, but it seems I can't get the SampleGrabber to provide RGB data. The device (an iSight -- running VC++ Express in VMWare) only reports MEDIASUBTYPE_YUY2. After extensive Googling, I still can't figure out whether DirectShow is supposed to provide built-in color space conversion for this sort of thing. Some sites report that there is no YUV<-RGB conversion built in, others report that you just have to call SetMediaType on your ISampleGrabber with an RGB subtype. Any advice is greatly appreciated, I'm going nuts on this one. Code provided below. Please note that The code works, except that it doesn't provide RGB data I'm aware that I can implement my own conversion filter, but this is not feasible because I'd have to anticipate every possible device format, and this is a relatively small project // Playback IGraphBuilder *pGraphBuilder = NULL; ICaptureGraphBuilder2 *pCaptureGraphBuilder2 = NULL; IMediaControl *pMediaControl = NULL; IBaseFilter *pDeviceFilter = NULL; IAMStreamConfig *pStreamConfig = NULL; BYTE *videoCaps = NULL; AM_MEDIA_TYPE **mediaTypeArray = NULL; // Device selection ICreateDevEnum *pCreateDevEnum = NULL; IEnumMoniker *pEnumMoniker = NULL; IMoniker *pMoniker = NULL; ULONG nFetched = 0; HRESULT hr = CoInitializeEx(NULL, COINIT_MULTITHREADED); // Create CreateDevEnum to list device hr = CoCreateInstance(CLSID_SystemDeviceEnum, NULL, CLSCTX_INPROC_SERVER, IID_ICreateDevEnum, (PVOID *)&pCreateDevEnum); if (FAILED(hr)) goto ReleaseDataAndFail; // Create EnumMoniker to list devices hr = pCreateDevEnum->CreateClassEnumerator(CLSID_VideoInputDeviceCategory, &pEnumMoniker, 0); if (FAILED(hr)) goto ReleaseDataAndFail; pEnumMoniker->Reset(); // Find desired device while (pEnumMoniker->Next(1, &pMoniker, &nFetched) == S_OK) { IPropertyBag *pPropertyBag; TCHAR devname[256]; // bind to IPropertyBag hr = pMoniker-&gt;BindToStorage(0, 0, IID_IPropertyBag, (void **)&amp;pPropertyBag); if (FAILED(hr)) { pMoniker-&gt;Release(); continue; } VARIANT varName; VariantInit(&amp;varName); HRESULT hr = pPropertyBag-&gt;Read(L"DevicePath", &amp;varName, 0); if (FAILED(hr)) { pMoniker-&gt;Release(); pPropertyBag-&gt;Release(); continue; } char devicePath[DeviceInfo::STRING_LENGTH_MAX] = ""; wcstombs(devicePath, varName.bstrVal, DeviceInfo::STRING_LENGTH_MAX); if (strcmp(devicePath, deviceId) == 0) { // Bind Moniker to Filter pMoniker-&gt;BindToObject(0, 0, IID_IBaseFilter, (void**)&amp;pDeviceFilter); break; } pMoniker-&gt;Release(); pPropertyBag-&gt;Release(); } if (pDeviceFilter == NULL) goto ReleaseDataAndFail; // Create sample grabber IBaseFilter *pGrabberF = NULL; hr = CoCreateInstance(CLSID_SampleGrabber, NULL, CLSCTX_INPROC_SERVER, IID_IBaseFilter, (void**)&pGrabberF); if (FAILED(hr)) goto ReleaseDataAndFail; hr = pGrabberF->QueryInterface(IID_ISampleGrabber, (void**)&pGrabber); if (FAILED(hr)) goto ReleaseDataAndFail; // Create FilterGraph hr = CoCreateInstance(CLSID_FilterGraph, NULL, CLSCTX_INPROC, IID_IGraphBuilder, (LPVOID *)&pGraphBuilder); if (FAILED(hr)) goto ReleaseDataAndFail; // create CaptureGraphBuilder2 hr = CoCreateInstance(CLSID_CaptureGraphBuilder2, NULL, CLSCTX_INPROC, IID_ICaptureGraphBuilder2, (LPVOID *)&pCaptureGraphBuilder2); if (FAILED(hr)) goto ReleaseDataAndFail; // set FilterGraph hr = pCaptureGraphBuilder2->SetFiltergraph(pGraphBuilder); if (FAILED(hr)) goto ReleaseDataAndFail; // get MediaControl interface hr = pGraphBuilder->QueryInterface(IID_IMediaControl, (LPVOID *)&pMediaControl); if (FAILED(hr)) goto ReleaseDataAndFail; // Add filters hr = pGraphBuilder->AddFilter(pDeviceFilter, L"Device Filter"); if (FAILED(hr)) goto ReleaseDataAndFail; hr = pGraphBuilder->AddFilter(pGrabberF, L"Sample Grabber"); if (FAILED(hr)) goto ReleaseDataAndFail; // Set sampe grabber options AM_MEDIA_TYPE mt; ZeroMemory(&mt, sizeof(AM_MEDIA_TYPE)); mt.majortype = MEDIATYPE_Video; mt.subtype = MEDIASUBTYPE_RGB32; hr = pGrabber->SetMediaType(&mt); if (FAILED(hr)) goto ReleaseDataAndFail; hr = pGrabber->SetOneShot(FALSE); if (FAILED(hr)) goto ReleaseDataAndFail; hr = pGrabber->SetBufferSamples(TRUE); if (FAILED(hr)) goto ReleaseDataAndFail; // Get stream config interface hr = pCaptureGraphBuilder2->FindInterface(NULL, &MEDIATYPE_Video, pDeviceFilter, IID_IAMStreamConfig, (void **)&pStreamConfig); if (FAILED(hr)) goto ReleaseDataAndFail; int streamCapsCount = 0, capsSize, bestFit = -1, bestFitPixelDiff = 1000000000, desiredPixelCount = _width * _height, bestFitWidth = 0, bestFitHeight = 0; float desiredAspectRatio = (float)_width / (float)_height; hr = pStreamConfig->GetNumberOfCapabilities(&streamCapsCount, &capsSize); if (FAILED(hr)) goto ReleaseDataAndFail; videoCaps = (BYTE *)malloc(capsSize * streamCapsCount); mediaTypeArray = (AM_MEDIA_TYPE **)malloc(sizeof(AM_MEDIA_TYPE *) * streamCapsCount); for (int i = 0; i < streamCapsCount; i++) { hr = pStreamConfig->GetStreamCaps(i, &mediaTypeArray[i], videoCaps + capsSize * i); if (FAILED(hr)) continue; VIDEO_STREAM_CONFIG_CAPS *currentVideoCaps = (VIDEO_STREAM_CONFIG_CAPS *)(videoCaps + capsSize * i); int closestWidth = MAX(currentVideoCaps-&gt;MinOutputSize.cx, MIN(currentVideoCaps-&gt;MaxOutputSize.cx, width)); int closestHeight = MAX(currentVideoCaps-&gt;MinOutputSize.cy, MIN(currentVideoCaps-&gt;MaxOutputSize.cy, height)); int pixelDiff = ABS(desiredPixelCount - closestWidth * closestHeight); if (pixelDiff &lt; bestFitPixelDiff &amp;&amp; ABS(desiredAspectRatio - (float)closestWidth / (float)closestHeight) &lt; 0.1f) { bestFit = i; bestFitPixelDiff = pixelDiff; bestFitWidth = closestWidth; bestFitHeight = closestHeight; } } if (bestFit == -1) goto ReleaseDataAndFail; AM_MEDIA_TYPE *mediaType; hr = pStreamConfig->GetFormat(&mediaType); if (FAILED(hr)) goto ReleaseDataAndFail; VIDEOINFOHEADER *videoInfoHeader = (VIDEOINFOHEADER *)mediaType->pbFormat; videoInfoHeader->bmiHeader.biWidth = bestFitWidth; videoInfoHeader->bmiHeader.biHeight = bestFitHeight; //mediaType->subtype = MEDIASUBTYPE_RGB32; hr = pStreamConfig->SetFormat(mediaType); if (FAILED(hr)) goto ReleaseDataAndFail; pStreamConfig->Release(); pStreamConfig = NULL; free(videoCaps); videoCaps = NULL; free(mediaTypeArray); mediaTypeArray = NULL; // Connect pins IPin *pDeviceOut = NULL, *pGrabberIn = NULL; if (FindPin(pDeviceFilter, PINDIR_OUTPUT, 0, &pDeviceOut) && FindPin(pGrabberF, PINDIR_INPUT, 0, &pGrabberIn)) { hr = pGraphBuilder->Connect(pDeviceOut, pGrabberIn); if (FAILED(hr)) goto ReleaseDataAndFail; } else { goto ReleaseDataAndFail; } // start playing hr = pMediaControl->Run(); if (FAILED(hr)) goto ReleaseDataAndFail; hr = pGrabber->GetConnectedMediaType(&mt); // Set dimensions width = bestFitWidth; height = bestFitHeight; _width = bestFitWidth; _height = bestFitHeight; // Allocate pixel buffer pPixelBuffer = (unsigned *)malloc(width * height * 4); // Release objects pGraphBuilder->Release(); pGraphBuilder = NULL; pEnumMoniker->Release(); pEnumMoniker = NULL; pCreateDevEnum->Release(); pCreateDevEnum = NULL; return true;

  Read the article

< Previous Page | 1 2 3