Search Results

Search found 32397 results on 1296 pages for 'reference program'.

Page 298/1296 | < Previous Page | 294 295 296 297 298 299 300 301 302 303 304 305  | Next Page >

  • Silverlight project builds in VS2008 but fails when using MSBuild

    - by Tom
    Hi, We have 2 Silverlight projects in the same solution; SLGlobalResource and SLData. SLData references SLGlobalResource (using references add reference projects). When we build it in debug within VS2008, everything builds fine and all is good. But when we build it using: msbuild TheSolution.sln /p:Configuration=Debug /t:rebuild SLData fails with the following error: ViewModels\ImportViewModel.cs : error CS0246: The type of name space "SLGlobalResource" could not be found (are you missing a using directive or an assembly reference?) This also happens in TeamCity (I guess because the TeamCity vs2008 runner uses MSBuild) Any ideas? Thanks Edit: There are actually 33 projects in total in the solution. I didn't think this was relevant before but now I'm thinking it could be - could this be a build order thing?

    Read the article

  • Custom DataType in DataTemplate breaks WPF designer

    - by PRINCESS FLUFF
    Why does the DataTemplate line break the WPF designer in Visual Studio 2008? The program compiles and runs properly. The DataTemplate is applied as it should. However the entire DataTemplate block of code is underlined in red, and when I simply "build" the program without running, I get the error "Type reference cannot find public type named 'Character'" How come it can't find it in the designer yet the program applies the template properly? <UserControl x:Class="WPF_Tests.Tests.TwoCollecViews.TwoViews" xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation" xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml" xmlns:DetailsPane="clr-namespace:WPF_Tests.Tests.DetailsPane" > <UserControl.Resources> <DataTemplate DataType="{x:Type DetailsPane:Character}"> <StackPanel Orientation="Horizontal"> <TextBlock Text="{Binding Path=Name}"></TextBlock> </StackPanel> </DataTemplate> </UserControl.Resources> <Grid> <ListBox ItemsSource="{Binding Path=Characters}" /> </Grid> </UserControl> EDIT: I am being told that this may be a bug in Visual Studio 2008, as it worked correctly in 2010. You can download the code here: http://www.mediafire.com/?z1myytvwm4n - The Test/TwoCollec xaml file's designer will break with this code.

    Read the article

  • handling java exception

    - by Noona
    This questions is related to java exception, why are there some cases that when an exception thrown the program exits even though the exception was caught and there was no exit() statement? my code looks something like this void bindProxySocket(DefaultHttpClientConnection proxyConnection, String hostName, HttpParams params) { if (!proxyConnection.isOpen()) { Socket socket = null; try { socket = new Socket(hostName, 80); } catch (UnknownHostException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } try { proxyConnection.bind(socket, params); } catch(IOException e) { System.err.println ("couldn't bind socket"); e.printStackTrace(); } } } and then I call this method like this: bindProxySocket(proxyConn, hostName, params1); but, the program exits, although I want to handle the exception by doing something else, can it be because I didn't enclose the method call within a try catch clause? what happens if I catch the exception again even though it's already in the method? and what should I do if i want to clean resources only if an exception occurs and otherwise I want to continue with the program? I am guessing in this case I have to include the whole piece of code until I can clean the resources with in a try statement or can I do it in the handle exception statement? some of these questions are on this specific case, but I would like to get a thorough answer to all my questions for future reference. thanks

    Read the article

  • Solved::MessageQueue.BeginReceive() null ref error - c#

    - by ltech
    Have a windows service that listens to a msmq. In the OnStart method is have this protected override void OnStart(string[] args) { try { _queue = new MessageQueue(_qPath);//this part works as i had logging before and afer this call //Add MSMQ Event _queue.ReceiveCompleted += new ReceiveCompletedEventHandler(queue_ReceiveCompleted);//this part works as i had logging before and afer this call _queue.BeginReceive();//This is where it is failing - get a null reference exception } catch(Exception ex) { EventLogger.LogEvent(EventSource, EventLogType, "OnStart" + _lineFeed + ex.InnerException.ToString() + _lineFeed + ex.Message.ToString()); } } where private MessageQueue _queue = null; This works on my machine but when deployed to a windows 2003 server and running as Network service account, it fails Exception recvd: Service cannot be started. System.NullReferenceException: Object reference not set to an instance of an object. at MYService.Service.OnStart(String[] args) at System.ServiceProcess.ServiceBase.ServiceQueuedMainCallback(Object state) Solved: Turned out that the Q that i set up, I had to explicitly add Network Service account to it under security tab

    Read the article

  • Get current PHP executable from within script?

    - by benizi
    I want to run a PHP cli program from within PHP cli. On some machines where this will run, both php4 and php5 are installed. If I run the outer program as php5 outer.php I want the inner script to be run with the same php version. In Perl, I would use $^X to get the perl executable. It appears there's no such variable in PHP? Right now, I'm using $_SERVER['_'], because bash (and zsh) set the environment variable $_ to the last-run program. But, I'd rather not rely on a shell-specific idiom. UPDATE: Version differences are but one problem. If PHP isn't in PATH, for example, or isn't the first version found in PATH, the suggestions to find the version information won't help. Additionally, csh and variants appear to not set the $_ environment variable for their processes, so the workaround isn't applicable there. UPDATE 2: I was using $_SERVER['_'], until I discovered that it doesn't do the right thing under xargs (which makes sense... zsh sets it to the command it ran, which is xargs, not php5, and xargs doesn't change the variable). Falling back to using: $version = explode('.', phpversion()); $phpcli = "php{$version[0]}";

    Read the article

  • problem with trying to create ssms add-in

    - by Karl
    I'm trying to create an add-in for SSMS 2008 and/or 2008 R2 but I've run into a problem straight away. I can get my add-in to work and on SSMS start-up get it to simply show a message box. However, after downloading various code-samples, when trying to reference Microsoft.SqlServer.Management.UI.VSIntegration.ServiceCache I get a null reference exception: Commands2 commands = (Commands2)ServiceCache.ExtensibilityModel.Commands; I get this problem when using SSMS 2008 or SSMS 2008 R2. I'm working on Visual Studio 2010. It's a bit frustrating because I'm keen to learn more about SSMS add-ins but can't seem to get past the few samples out there. Any advice/tips appreciated. Thanks

    Read the article

  • gdb reverse debugging error

    - by Werner
    Hi, i started to try reverse debugging with gdb 7, followin the tutorial: http://www.sourceware.org/gdb/wiki/ProcessRecord/Tutorial and I thought, great! Then I started to debug a real program which gives an error at the end. So I run it with gdb, and I put a breakpoint just before the place I think the error appears. Then I type "record" in order to start to recrd actions for future reverse-debugging. But after some steps I get Process record doesn't support instruction 0xf0d at address 0x2aaaab4c4b4e. Process record: failed to record execution log. Program received signal SIGTRAP, Trace/breakpoint trap. 0x00002aaaab4c4b4e in memcpy () from /lib64/libc.so.6 (gdb) n Single stepping until exit from function memcpy, which has no line number information. Process record doesn't support instruction 0xf0d at address 0x2aaaab4c4b4e. Process record: failed to record execution log. Program received signal SIGABRT, Aborted. 0x00002aaaab4c4b4e in memcpy () from /lib64/libc.so.6 Before I look at in in detail, I wonder if this feature is still buggy, or if I should start to record from the beginning. Where this "record" error happens, just an object is created as a copy of other: Thanks

    Read the article

  • Progressing Alfresco workflows through web script

    - by Domchi
    I have an Alfresco document reference; what I'm looking for is a way to access workflow attached to that document and finish it (or progress it to the next transition) through Javascript. Almost every example on the web shows how to start workflow, and from the dashlet I could call task command processor (/alfresco/command/task/end/[/transition]) if I knew the task ID, but how do I do the same thing from server-side web script starting only from the document reference? There must be a way to access workflows from document and manage them programatically.

    Read the article

  • Web Service URL change is not recognized

    - by ila
    I deployed in a production environment a .net solution that consumes a web service added as "Service reference" in visual studio. Today the endpoint URL has changed, and I modified that URL in web.config. But when I run the solution I get the error: System.ServiceModel.FaultException: Server did not recognize the value of HTTP Header SOAPAction: http://93.62.150.200/LogEvent. Now that sound strange to me. As you can see, the IP is 93.62.... but the new value is different. The web.config entry (with the new value or the URL) is: <client> <endpoint address="http://213.92.50.215/sawfc/WS_SAWFC_Int.asmx" binding="basicHttpBinding" bindingConfiguration="wsSAWFCInterfaceSoap" contract="it.datasphere.ws.wsSAWFCInterfaceSoap" name="wsSAWFCInterfaceSoap" /> </client> I cannot reach the new web service from dev environment, so I cannot modify the Service reference there. Any idea? Thanks a lot!

    Read the article

  • Turing Machine & Modern Computer

    - by smwikipedia
    I heard a lot that modern computers are based on Turing machine. I'd like to share my understanding and hear your comments. I think the computer is a big general-purpose Turing machine. Each program we write is a small specific-purpose Turing machine. The classical Turing machine do its job based on the input and its current state inside and so do our programs. Let's take a running program (a process) as an example. We know that in the process's address space, there's areas for stack, heap, and code. A classical Turing machine doesn't have the ability to remember many things, so we borrow the concept of stack from the push-down automaton. The heap and stack areas contains the state of our specific-purpose Turing machine (our program). The code area represents the logic of this small Turing machine. And various I/O devices supply input to this Turing machine. The above is my naive understanding about the working paradigm of modern computer. I couln't wait to hear your comments. Thanks very much.

    Read the article

  • Address Book callback not called

    - by Oliver
    I have an iPhone app that makes use of the AddressBook.framework and uses Core Data to store these contacts. In order to make sure I update my own database when the Address Book is updated (whether via MobileMe or editing within my own app), I am subscribing to the notification as to when the Address Book is updating. I call this on startup: ABAddressBookRef book = ABAddressBookCreate(); ABAddressBookRegisterExternalChangeCallback(book, addressBookChanged, self); Which (supposedly) calls this on any editing. I have an ABPersonViewController which allows editing, and addressBookChanged never seems to get called. void addressBookChanged(ABAddressBookRef reference, CFDictionaryRef dictionary, void *context) { // The contacts controller we need to call ContactsController *contacts = (ContactsController *)context; // Sync with the Address Book [contacts synchronizeWithAddressBook:reference]; } Is there any reason for it to not be called?

    Read the article

  • Mixing static and dynamic endpoints in app.yaml file

    - by Greg
    I'm trying to describe endpoints in my App Engine app and am having difficulty for directory structures that mix static and dynamic content. But my yaml rules are conflicting with one another. Before I change my directory structure, does anyone have a recommendation? The goal is to create a directory that contains both documentation (static html files) and implementations. /api - /v1 - getitdone.py - doc.html - index.html What I think I should be doing with my application yaml... - url: /api/v1/getitdone script: api/v1/getitdone.py - url: /api/ static_files: api/index.html upload: api/index.html - url: /api static_dir: api But this causes the dynamic endpoints to fail. I'm assuming the static_dir reference is breaking it. How can I do this without describing every script and static file reference (I have many more than are listed here)?

    Read the article

  • getchar does not stop when using scanf

    - by Oz123
    I am have a difficulty understanding getchar(). In the following program getchar works as expected: #include <stdio.h> int main() { printf("Type Enter to continue..."); getchar(); return 0; } However, in the following program, getchar does not create a delay and the program ends: #include <stdio.h> int main() { char command[100]; scanf("%s", command ); printf("Type Enter to continue..."); getchar(); return 0; } I have the following weired workaround, which works, but I understand why: #include <stdio.h> int main() { char command[100]; int i; scanf("%s", command ); printf("Type Enter to continue..."); while ( getchar() != '\n') { i=0; } getchar(); return 0; } So my questions are: 1. What is scanf doing? Why does scanf do this ? 2. Why is my work around working? 3. What is a good way to emulate the following Python code: raw_input("Type Enter to continue")

    Read the article

  • R: disentangling scopes

    - by rescdsk
    Hi, Right now, in my R project, I have functions1.R with doFoo() and doBar(), functions2.R with other functions, and main.R with the main program in it, which first does source('functions1.R'); source('functions2.R'), and then calls the other functions. I've been starting the program from the R GUI in Mac OS X, with source('main.R'). This is fine the first time, but after that, the variables that were defined the first time through the program are defined for the second time functions*.R are sourced, and so the functions get a whole bunch of extra variables defined. I don't want that! I want an "undefined variable" error when my function uses a variable it shouldn't! Twice this has given me very late nights of debugging! So how do other people deal with this sort of problem? Is there something like source(), but that makes an independent namespace that doesn't fall through to the main one? Making a package seems like one solution, but it seems like a big pain in the butt compared to e.g. Python, where a source file is automatically a separate namespace. Any tips? Thank you!

    Read the article

  • is it possible to turn off vdso on glibc side?

    - by heroxbd
    I am aware that passing vdso=0 to kernel can turn this feature off, and that the dynamic linker in glibc can automatic detect and use vdso feature from kernel. Here I met with this problem. There is a RHEL 5.6 box (kernel 2.6.18-238.el5) in my institution where I only have a normal user access, probably suffering from RHEL bug 673616. As I compile a toolchain of linux-headers-3.9/gcc-4.7.2/glibc-2.17/binutils-2.23 on top of it, gcc bootstrap fails in cc1 in stage2 cannnot be run Program received signal SIGSEGV, Segmentation fault. 0x00002aaaaaaca6eb in ?? () (gdb) info sharedlibrary From To Syms Read Shared Object Library 0x00002aaaaaaabba0 0x00002aaaaaac3249 Yes (*) /home/benda/gnto/lib64/ld-linux-x86-64.so.2 0x00002aaaaacd29b0 0x00002aaaaace2480 Yes (*) /home/benda/gnto/usr/lib/libmpc.so.3 0x00002aaaaaef2cd0 0x00002aaaaaf36c08 Yes (*) /home/benda/gnto/usr/lib/libmpfr.so.4 0x00002aaaab14f280 0x00002aaaab19b658 Yes (*) /home/benda/gnto/usr/lib/libgmp.so.10 0x00002aaaab3b3060 0x00002aaaab3b3b50 Yes (*) /home/benda/gnto/lib/libdl.so.2 0x00002aaaab5b87b0 0x00002aaaab5c4bb0 Yes (*) /home/benda/gnto/usr/lib/libz.so.1 0x00002aaaab7d0e70 0x00002aaaab80f62c Yes (*) /home/benda/gnto/lib/libm.so.6 0x00002aaaaba70d40 0x00002aaaabb81aec Yes (*) /home/benda/gnto/lib/libc.so.6 (*): Shared library is missing debugging information. and a simple program #include <sys/time.h> #include <stdio.h> int main () { struct timeval tim; gettimeofday(&tim, NULL); return 0; } get segment fault in the same way if compiled against glibc-2.17 and xgcc from stage1. Both cc1 and the test program can be run on another running RHEL 5.5 (kernel 2.6.18-194.26.1.el5) with gcc-4.7.2/glibc-2.17/binutils-2.23 as normal user. I cannot simply upgrade the box to a newer RHEL version, nor could I turn VDSO off via sysctl or proc. The question is, is there a way to compile glibc so that it turns off VDSO unconditionally?

    Read the article

  • Prevent deferred creation of controls.

    - by Scott Chamberlain
    Here is a test framework to show what I am doing, just create a new project add a tabbed control, on tab 1 put a button on tab 2 put a check box (default names) and paste this code for its code public partial class Form1 : Form { private List<bool> boolList = new List<bool>(); BindingSource bs = new BindingSource(); public Form1() { InitializeComponent(); boolList.Add(false); bs.DataSource = boolList; checkBox1.DataBindings.Add("Checked", bs, ""); } bool updating = false; private void button1_Click(object sender, EventArgs e) { updating = true; boolList[0] = true; bs.ResetBindings(false); Application.DoEvents(); updating = false; } private void checkBox1_CheckedChanged(object sender, EventArgs e) { if (!updating) MessageBox.Show("CheckChanged fired outside of updating"); } } The issue is if you run the program and look at tab 2 then press the button on tab 1 the program works as expected, however if you press the button on tab 1 then look at tab 2 the event for the checkbox will not fire untill you look at tab 2. The reason for this is the controll on tab 2 is not in the "created" state, so its binding to change the checkbox from unchecked to checked does not happen until after the control has been "Created". checkbox1.CreateControl() does not do anything because according to MSDN CreateControl does not create a control handle if the control's Visible property is false. You can either call the CreateHandle method or access the Handle property to create the control's handle regardless of the control's visibility, but in this case, no window handles are created for the control's children. I tried getting the value of Handle(there is no CreateHandle for Button) but still the same result. Any suggestions other than have the program quickly flash all of my tabs that have data-bound check boxes when it first loads?

    Read the article

  • Calling a method in a view controller from a view

    - by Lakshmie
    I have to invoke a method present in a view controller who's reference is available in the view. When I try to call the method like any other method, for some reason, iPhone just ignores the call. Can somebody explain as to why this happens and also how can I go about invoking this method? In the view I have this method: -(void) touchesBegan :(NSSet *) touches withEvent:(UIEvent *)event{ NSArray* mySubViews = [self subviews]; for (UITouch *touch in touches) { int i = 0; for(; i<[mySubViews count]; i++){ if(CGRectContainsPoint([[mySubViews objectAtIndex:i] frame], [touch locationInView:self])){ break; } } if(i<[mySubViews count]){ // viewController is the reference to the View Controller. [viewController pointToSummary:[touch locationInView:self].y]; NSLog(@"Helloooooo"); break; } } } Whenever the touches event is triggered, Hellooooo gets printed in the console but the method before that is simply ignored

    Read the article

  • .NET 4.0 build issues on CI Server

    - by DMcKenna
    Anybody manage to get .net 4.0 applications compiling on a CI server without installing visual studio 2010 on a CI server. No SDK exists for .net 4.0. Have installed .NET 4.0 on CI Server. Msbuild.exe works for simple projects and give the following warning (GetReferenceAssemblyPaths target) - C:\Windows\Microsoft.NET\Framework\v4.0.30319\Microsoft.Common.targets(847,9): warning MSB3644: The reference assemblies for framework ".NETFramework,Version=v4.0" were not found. To resolve this, install the SDK or Targeting Pack for this framework version or retarget your application to a version of the framework for which you have the SDK or Targeting Pack installed. Note that assemblies will be resolved from the Global Assembly Cache (GAC) and will be used in placeof reference assemblies. Therefore your assembly may not be correctly targeted for the framework you intend.

    Read the article

  • Problem with signals and slots

    - by Jen
    I am creating a class with custom slots in Qt: class CustomEdit : public QTextEdit { Q_OBJECT public: CustomEdit(QWidget* parent); public slots: void onTextChanged (); }; However, I'm getting thise linker error: undefined reference to 'vtable for CustomEdit' The documentation says: if you get compiler errors along the lines of "undefined reference to vtable for LcdNumber", you have probably forgotten to run the moc or to include the moc output in the link command. ... but it is not obvious what that means. Is there something I need to add to my class, or to the .pro file?

    Read the article

  • Perl : package using problem.

    - by pavun_cool
    Hi All.. Normally when I am writing the perl program . I used to include following package . use strict ; use warnings ; use Data::Dumper ; Now , I want like this, I will not include all this package for every program . for that I will have these all package in my own package. like following. my_packages.pm package my_packages ; { use strict ; use warnings ; use Data::Dumper; } 1; So, that if I add my_packages.pm in perl program , it needs have all above packages . Actually I have done this experimentation . But I am not able get this things . which means when I am using my_packages . I am not able get the functionality of "use strict, use warnings , use Data::Dumper ". Someone help me out of this problem.....

    Read the article

  • Validate xsl fo in xslt styleesheet

    - by Biegal
    Hi, i have a little problem with validating xml, xslt in details. I have a xslt stylesheet that, transforms xml data source to xsl:fo document. Something like this: <?xml version="1.0" encoding="utf-8"?> <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <xsl:template match="/"> <fo:root xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns="http://www.w3.org/1999/xhtml"> <fo:layout-master-set> <fo:simple-page-master margin-top="25mm" margin-bottom="25mm" margin-left="25mm" margin-right="25mm" page-width="210mm" page-height="297mm" master-name="simplePageLayout"> <fo:region-body region-name="xsl-region-body" column-gap="0.25in" /> <fo:region-before region-name="xsl-region-before" display-align="after" extent="0.1mm" padding-top="0pt" padding-left="0.4in" padding-right="0.4in" padding-bottom="0pt" /> <fo:region-after region-name="xsl-region-after" display-align="before" extent="0.4in" padding-top="4pt" padding-left="0.4in" padding-right="0.4in" padding-bottom="0pt" /> </fo:simple-page-master> <fo:page-sequence-master master-name="default-sequence"> <fo:repeatable-page-master-reference master-reference="simplePageLayout" /> </fo:page-sequence-master> </fo:layout-master-set> <fo:page-sequence master-reference="default-sequence"> <fo:flow flow-name="xsl-region-body"> <fo:block font-family="Segoe UI" color="#000000" font-size="9pt" /> </fo:flow> </fo:page-sequence> </fo:root> </xsl:template> What I want to do, is to validate written xsl:fo elements, ignoring xsl tags. Is it posible? For now I use dtd validation (I have xsd schema too) for validating Fo, but it give me an error on each xsl tag. Summary. Is it posiible to validate only fo elements against the schema, ignoring xsl tags, and how should I do it? Maybe a code snnippet in C#, or a hint how to modify documents? Thanks in advance!

    Read the article

  • unable to implement HTTP Tunneling correctly in order to enable Java rmi calls over internet(and und

    - by Lokesh Kumar
    in my previous question :-How to Setup RMI Server under(NAT/ISP) Now,i m able to start my RMI server by Installing apache Tomcat 6.0 server. i have also installed servlet programs into apache Tomcat server in order to enable HTTP tunneling. my servlet codes:- (1) [SimplifiedServletHandler.java][2] (2) [ServletForwardCommand.java][3] these servlets resides inside :- C:\Program Files\Apache Software Foundation\Tomcat 6.0\webapps\examples\WEB-INF\classes\ one more thing that i hv added to my CalcultorClient.java program:- try { RMISocketFactory. setSocketFactory(new sun.rmi.transport.proxy .RMIHttpToCGISocketFactory( )); }catch (IOException ignored) { System.out.println("Error :- ignored.getMessage()"); } But,when i try to make client connect with server(under ISP/NAT) i get the following Exception :- RemoteException java.rmi.UnmarshalException: Error unmarshaling return header; nested exception is: java.io.IOException: HTTP request failed i don't know the correct reason behind this Exception.. but,i think that i haven't installed or invoke my servlet programs properly on server side. so,can anybody tell me the correct reason behind this error/Exception????? and if u think that it is servlet problem then tell me the correct procedure to run my serlvet program inside tomcat server.

    Read the article

  • Global find object references in NHibernate

    - by Miral
    Is it possible to perform a global reversed-find on NHibernate-managed objects? Specifically, I have a persistent class called "Io". There are a huge number of fields across multiple tables which can potentially contain an object of that type. Is there a way (given a specific instance of an Io object), to retrieve a list of objects (of any type) that actually do reference that specific object? (Bonus points if it can identify which specific fields actually contain the reference, but that's not critical.) Since the NHibernate mappings define all the links (and the underlying database has corresponding foreign key links), there ought to be some way to do it.

    Read the article

  • Unit finalization order for application, compiled with run-time packages?

    - by Alexander
    I need to execute my code after finalization of SysUtils unit. I've placed my code in separate unit and included it first in uses clause of dpr-file, like this: project Project1; uses MyUnit, // <- my separate unit SysUtils, Classes, SomeOtherUnits; procedure Test; begin // end; begin SetProc(Test); end. MyUnit looks like this: unit MyUnit; interface procedure SetProc(AProc: TProcedure); implementation var Test: TProcedure; procedure SetProc(AProc: TProcedure); begin Test := AProc; end; initialization finalization Test; end. Note that MyUnit doesn't have any uses. This is usual Windows exe, no console, without forms and compiled with default run-time packages. MyUnit is not part of any package (but I've tried to use it from package too). I expect that finalization section of MyUnit will be executed after finalization section of SysUtils. This is what Delphi's help tells me. However, this is not always the case. I have 2 test apps, which differs a bit by code in Test routine/dpr-file and units, listed in uses. MyUnit, however, is listed first in all cases. One application is run as expected: Halt0 - FinalizeUnits - ...other units... - SysUtils's finalization - MyUnit's finalization - ...other units... But the second is not. MyUnit's finalization is invoked before SysUtils's finalization. The actual call chain looks like this: Halt0 - FinalizeUnits - ...other units... - SysUtils's finalization (skipped) - MyUnit's finalization - ...other units... - SysUtils's finalization (executed) Both projects have very similar settings. I tried a lot to remove/minimize their differences, but I still do not see a reason for this behaviour. I've tried to debug this and found out that: it seems that every unit have some kind of reference counting. And it seems that InitTable contains multiply references to the same unit. When SysUtils's finalization section is called first time - it change reference counter and do nothing. Then MyUnit's finalization is executed. And then SysUtils is called again, but this time ref-count reaches zero and finalization section is executed: Finalization: // SysUtils' finalization 5003B3F0 55 push ebp // here and below is some form of stub 5003B3F1 8BEC mov ebp,esp 5003B3F3 33C0 xor eax,eax 5003B3F5 55 push ebp 5003B3F6 688EB50350 push $5003b58e 5003B3FB 64FF30 push dword ptr fs:[eax] 5003B3FE 648920 mov fs:[eax],esp 5003B401 FF05DCAD1150 inc dword ptr [$5011addc] // here: some sort of reference counter 5003B407 0F8573010000 jnz $5003b580 // <- this jump skips execution of finalization for first call 5003B40D B8CC4D0350 mov eax,$50034dcc // here and below is actual SysUtils' finalization section ... Can anyone can shred light on this issue? Am I missing something?

    Read the article

< Previous Page | 294 295 296 297 298 299 300 301 302 303 304 305  | Next Page >