Search Results

Search found 102 results on 5 pages for 'russ bradberry'.

Page 3/5 | < Previous Page | 1 2 3 4 5  | Next Page >

  • Create a new domain with the same name of a trusted domain

    - by Russ
    I have a domain blah.com that was aquired a while back by my company foo.com. I set up a two way trust between the two domains, but now I want to move their servers into our forest, while keeping the domain name of blah.com. Is this possible? What things do I need to consider when doing this? I know I can't move the domain from its forest into our forest. blah.com is a 2003 native domain/forest foo.com is a 2008R2 domain/2003 forest.

    Read the article

  • Altering the Windows 7 grouping thumbnail preview

    - by russ
    I like the windows 7 grouping behaviour for multiple similar windows, except when I have multiple Explorer windows open. In this case, showing thumbnails of each open explorer window is next to useless, and I would like to know how to either: Disable thumbnail display completely (but preserve the grouping behaviour) Disable thumbnail display for a subset of programs (all explorer.exe instances) Change the number of similar programs that are open before the grouping behaviour falls back on a more sensible text preview when hovering over the taskbar icon (currently 14 or over) I don't particularly want to disable any other features of windows to get this behaviour, so disabling the theme compleletely isn't what I want to do. I definitely want to keep the grouping behaviour. Its just those thumbnail previews I would like to alter. Here are some screen shots of what I don't want, and what I do want: http://dl.dropbox.com/u/13184372/StackOverflow/BadGrouping.png http://dl.dropbox.com/u/13184372/StackOverflow/GoodGrouping.png

    Read the article

  • How can I find the names of AD Group policies that a user/pc is using?

    - by Russ
    I am having trouble locating some settings in group policy so I can make changes due to the convoluted nature of our policies. What I would like to be able to do is go to a specific PC and see what group policies are being applied, so I can focus on those policies. My goal would be to clean up the GP's a bit, while allowing me to "walk the tree" to see what people have implemented and what is worthless. Thanks. EDIT: In this specific case, I am looking to find which GP maped drives are configured in. (User Configuration -- Preferences -- Windows Settings -- Drive Maps)

    Read the article

  • Auto-response on a mail non universal group

    - by russ
    Four domain users are part of global security group [email protected] running on Windows Server 2008 Standard and Exchange 2007 as a mail non universal group. Our goal is to create an auto-responder rule, or template that fires at sender. I have created a mail-enabled public folder, with owner permissions for each group member. I have also set up a rule template (response) and I am a little unclear as to the send as permission set, or where to set that up. Can someone please tell me if I am on the right track? Will I have to do each Outlook client machine (whole group!) the same?

    Read the article

  • Domain controller policies having administrative issues with program installations

    - by russ
    I have an issue with our domain and policies where a user needs administrative rights to install a program and someone provides the credentials for the 'run as admin' or 'run as other user' it will always show an error: Cannot find specified file on the drive. So we cannot install programs for our employees. How I fix it is I add the user to administrators, force a policy update, have them reboot, run as administrator and install it that way. Then remove them after and update the policy. Another way was to right click the install file (.exe/.msi) and add the user and administrator as an owner with full control. Only works when we have the installation file. I have no idea how to fix this and I can't find a relevant thread or QA about this problem. I can only find users having issues on their home computers not on a domain. Any help and threads to read would be great!

    Read the article

  • How Do I Parse a String?

    - by Russ
    I am new to bash, and I am creating a script that loops through the files in a directory and based on part of the filename, does something with the file, so far I have this: #!/bin/bash DIR="/Users/me/Documents/import/*" for f in "$DIR" do $t=?????? echo "Loading $f int $t..." done so $f will output something like this: /Users/me/Documents/import/time_dim-1272037430173 out of this, I want time_dim, the directory can be variable length and -1272037430173 is a fixed length (it's the unix timestamp btw). What is the best way to go about this?

    Read the article

  • Windows Workflow Foundation 4 (WF4) Bookmark

    - by Russ Clark
    I'm working with Visual Studio 2010 and have a workflow where I'm trying to resume a bookmark. The bookmark appears to be getting set just fine, but when I try to run the code below to resume it, I get an entry in the InstancesTable in the persistence database with the SurrogateLockOwnerID field set to an integer, usually this field is null, and I can no longer do anything with the workflow. I think this is indicating that I have some kind of blocking lock going on, but I can't find any references anywhere to the SurrogateLockOwnerID. Does anyone know what is causing this, and how to fix it? WorkflowApplication workflowApp = new WorkflowApplication(new WebCARSWorkflow()); workflowApp.Extensions.Add(new SqlTrackingParticipant(ConnString)); InstanceStore instanceStore = new SqlWorkflowInstanceStore(ConnString); workflowApp.InstanceStore = instanceStore; workflowApp.Load(WorkflowInstanceId); workflowApp.PersistableIdle = (waie) => PersistableIdleAction.Unload; Approver actualApprover = new Approver(ProcessingStatus.Awaiting, my_RDA_Role, "RDAGetPlayersFromExternalRoleURI", 14); actualApprover.ActionDate = DateTime.Now; Player actualPlayer = (Player)GetPlayer(login); actualApprover.ActualPlayer = actualPlayer; actualApprover.ApproverAction = action; workflowApp.ResumeBookmark("ErrorOccurredBookmark", actualApprover);

    Read the article

  • Windows 7 - Enable Network DTC Access

    - by Russ Clark
    I have a Visual Studio 2010 Windows Forms application in which I start a transaction using the TransactionScope class. I then Receive a message from a Sql Server Broker Services message queue, which works fine. I next try to call a stored procedure from the same database with a call to my data access layer which is a Visual Studio dataset (xsd file). When I make this second call to the database I get the following error message: The MSDTC transaction manager was unable to pull the transaction from the source transaction manager due to communication problems. Possible causes are: a firewall is present and it doesn't have an exception for the MSDTC process, the two machines cannot find each other by their NetBIOS names, or the support for network transactions is not enabled for one of the two transaction managers. (Exception from HRESULT: 0x8004D02B). I've seen several posts on the web that talk about Enabling DTC access through dcomcnfg.exe, and allowing DTC to communicate through Windows Firewall. I've done those things, and am still having this problem. I know our remote database server is setup to Enable DTC access, because we are using similar transactions in other projects built with Visual Studio 2008 on Windows XP and Vista. I think there is something specific about Windows 7 and Visual Studio 2010 causing this problem, but haven't been able to find out what it is. Can anyone help with this problem?

    Read the article

  • ASP.NET MVC 2 and authentication using WIF (Windows Identity Foundation)

    - by Russ Cam
    Are there any decent examples of the following available: Looking through the WIF SDK, there are examples of using WIF in conjunction with ASP.NET using the WSFederationAuthenticationModule (FAM) to redirect to an ASP.NET site thin skin on top of a Security Token Service (STS) that user uses to authenticate (via supplying a username and password). If I understand WIF and claims-based access correctly, I would like my application to provide its own login screen where users provide their username and password and let this delegate to an STS for authentication, sending the login details to an endpoint via a security standard (WS-*), and expecting a SAML token to be returned. Ideally, the SessionAuthenticationModule would work as per the examples using FAM in conjunction with SessionAuthenticationModule i.e. be responsible for reconstructing the IClaimsPrincipal from the session security chunked cookie and redirecting to my application login page when the security session expires. Is what I describe possible using FAM and SessionAuthenticationModule with appropriate web.config settings, or do I need to think about writing a HttpModule myself to handle this? Alternatively, is redirecting to a thin web site STS where users log in the de facto approach in a passive requestor scenario?

    Read the article

  • Windows Workflow Foundation 4 (WF4) ReHosting

    - by Russ Clark
    I've been looking at the possibility of ReHosting a WF4 Workflow to be used to debug running Workflows. All the posts and samples I've seen regarding WF4 Rehosting are using a WPF application to initially Host the Workflow, and then use the WorkflowDesigner in ReHosting it. Is there any way to Rehost a Workflow that was hosted in a non WPF application, like ASP.Net MVC?

    Read the article

  • Windows Workflow Foundation 4 (WF4) Error Handling

    - by Russ Clark
    What is the best way to get error messages from a WF4 workflow back to a hosting ASP.NET MVC application? I need the workflow to not terminate, but continue to be active, and then pass a message back to the hosting app regarding the error, so the user can take an alternative action, but I'm not sure how to do that.

    Read the article

  • How can I use a custom TabItem control when databinding a TabControl in WPF?

    - by Russ
    I have a custom control that is derived from TabItem, and I want to databind that custom TabItem to a stock TabControl. I would rather avoid creating a new TabControl just for this rare case. This is what I have and I'm not having any luck getting the correct control to be loaded. In this case I want to use my ClosableTabItem control instead of the stock TabItem control. <TabControl x:Name="tabCases" IsSynchronizedWithCurrentItem="True" Controls:ClosableTabItem.TabClose="TabClosed" > <TabControl.ItemTemplate> <DataTemplate DataType="{x:Type Controls:ClosableTabItem}" > <TextBlock Text="{Binding Path=Id}" /> </DataTemplate> </TabControl.ItemTemplate> <TabControl.ContentTemplate> <DataTemplate DataType="{x:Type Entities:Case}"> <CallLog:CaseReadOnlyDisplay DataContext="{Binding}" /> </DataTemplate> </TabControl.ContentTemplate> </TabControl> EDIT: This is what I ended up with, rather than trying to bind a custom control. The "CloseCommand" im getting from a previous question. <Style TargetType="{x:Type TabItem}" BasedOn="{StaticResource {x:Type TabItem}}" > <Setter Property="Template"> <Setter.Value> <ControlTemplate TargetType="{x:Type TabItem}"> <Border Name="Border" Background="LightGray" BorderBrush="Black" BorderThickness="1" CornerRadius="25,0,0,0" SnapsToDevicePixels="True"> <StackPanel Orientation="Horizontal"> <ContentPresenter x:Name="ContentSite" VerticalAlignment="Center" HorizontalAlignment="Center" ContentSource="Header" Margin="20,1,5,1"/> <Button Command="{Binding Path=CloseCommand}" Cursor="Hand" DockPanel.Dock="Right" Focusable="False" Margin="1,1,5,1" Background="Transparent" BorderThickness="0"> <Image Source="/Russound.Windows;component/Resources/Delete.png" Height="10" /> </Button> </StackPanel> </Border> <ControlTemplate.Triggers> <Trigger Property="IsSelected" Value="True"> <Setter Property="FontWeight" Value="Bold" /> <Setter TargetName="Border" Property="Background" Value="LightBlue" /> <Setter TargetName="Border" Property="BorderThickness" Value="1,1,1,0" /> <Setter TargetName="Border" Property="BorderBrush" Value="DarkBlue" /> </Trigger> </ControlTemplate.Triggers> </ControlTemplate> </Setter.Value> </Setter> </Style>

    Read the article

  • Windows Workflow Foundation 4 (WF4) Delay

    - by Russ Clark
    I'm working with the the Release Candidate of Visual Studio 2010 using Wf4 to write a new workflow for approving resource requests. In my workflow, I would like for a request to expire after a few days if no approval has been made for the request. We did this in WF 3.5 (Visual Studio 2008) by adding a Delay timer into an EventDrivenActivity parallel to the EventDrivenActivity that was awaiting an approver to come and approve the request. If the Delay expired before an approval was made, the EventDrivenActivity would terminate the request. Does anyone know if there is a similar mechanism for doing this in WF4?

    Read the article

  • iPhone "slide to unlock" animation

    - by Russ
    Any ideas as to how Apple implemented the "slide to unlock" (also, "slide to power off" is another identical example) animation? I thought about some sort of animating mask - but masking is not available on the iPhone OS for performance reasons. Is there some private API effect (like SuckEffect) that they might have used? A spotlight type of effect? Some Core Animation thing? Edit: It's definitely not a series of stills. I've seen examples of being edit a plist value or something and customize the string on jailbroken iphones.

    Read the article

  • DataContractSerializer and XSLT not Serializing Class Properties

    - by Russ Clark
    I've written a simple Employee class that I'm trying to serialize to an XDocument and then use XSLT to transform the document to a page that displays both the properties (Name and ID) from the Employee class, and an html form with 2 radio buttons (Approve and Reject) and a submit button. Here is the Employee class: [Serializable, DataContract(Namespace="XSLT_MVC.Controllers/")] public class Employee { [DataMember] public string Name { get; set; } [DataMember] public int ID { get; set; } public Employee() { } public Employee(string name, int id) { Name = name; ID = id; } public XDocument GetDoc() { XDocument doc = new XDocument(); var serializer = new DataContractSerializer(typeof(Employee)); using (var writer = doc.CreateWriter()) { serializer.WriteObject(writer, this); writer.Close(); } return doc; } } And here is the XSLT file: <?xml version="1.0" encoding="utf-8"?> <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform" > <xsl:output method="html" indent="yes"/> <xsl:template match="/"> <html> <body> <xsl:value-of select="Employee/Name"/> <br /> <xsl:value-of select="Employee/ID"/> <br /> <form method="post" action="/Home/ProcessRequest?id={Employee/ID}"> <input id="Action" name="Action" type="radio" value="Approved"></input> Approved <br /> <input id="Action" name="Action" type="radio" value="Rejected"></input> Rejected <br /> <input type="submit" value="Submit"></input> </form> </body> </html> </xsl:template> </xsl:stylesheet> When I run this, all I get is the html form with the 2 radio buttons and the submit button, but not the properties from the Employee class. I saw a separate StackOverflow post that said I need to change the <xsl:template match="/"> to match on the namespace of my Employee class like this: <xsl:template match="/XSLT_MVC.Controllers"> but when I do that, now all I get are the Employee properties, and not the html form with the 2 radio buttons and the submit button. Does anyone know what needs to be done so that my transform will select and display both the Employee properties and the html form?

    Read the article

  • How do you create a non-Thread-based Guice custom Scope?

    - by Russ
    It seems that all Guice's out-of-the-box Scope implementations are inherently Thread-based (or ignore Threads entirely): Scopes.SINGLETON and Scopes.NO_SCOPE ignore Threads and are the edge cases: global scope and no scope. ServletScopes.REQUEST and ServletScopes.SESSION ultimately depend on retrieving scoped objects from a ThreadLocal<Context>. The retrieved Context holds a reference to the HttpServletRequest that holds a reference to the scoped objects stored as named attributes (where name is derived from com.google.inject.Key). Class SimpleScope from the custom scope Guice wiki also provides a per-Thread implementation using a ThreadLocal<Map<Key<?>, Object>> member variable. With that preamble, my question is this: how does one go about creating a non-Thread-based Scope? It seems that something that I can use to look up a Map<Key<?>, Object> is missing, as the only things passed in to Scope.scope() are a Key<T> and a Provider<T>. Thanks in advance for your time.

    Read the article

  • JQuery Autocomplete plugin not working with JQuery 1.4.1

    - by Russ Clark
    I've been using the JQuery Autocomplete plugin with JQuery version 1.3.2, and it has been working great. I recently updated JQuery in my project to version 1.4.2, and the Autocomplete plugin is now broken. The JQuery code to add items to a textbox on my web page doesn't seem to be getting called at all. Does anyone know if the JQuery Autocomplete plugin is incompatible with JQuery version 1.4.2, and if there is a fix for this problem? Here is some sample code I've built in an ASP.Net web site (which works fine if I change the JQuery file to jquery-1.3.2.js, but nothing happens using jquery-1.4.2.js): <%@ Page Language="C#" AutoEventWireup="true" CodeFile="Default.aspx.cs" Inherits="_Default" %> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head runat="server"> <title>Untitled Page</title> <script type="text/javascript" src="js/jquery-1.4.2.js" ></script> <script type="text/javascript" src="js/jquery.autocomplete.js" ></script> <script type="text/javascript"> $(document).ready(function() { var data = "Core Selectors Attributes Traversing Manipulation CSS Events Effects Ajax Utilities".split(" "); $(':input:text:id$=sapleUser').autocomplete(data); }); </script> </head> <body> <form id="form1" runat="server"> API Reference: <input id="sapleUser" autocomplete="off" type="text" runat="server" /> (try "C" or "E") </form> </body> </html>

    Read the article

  • Can I spead out a long running stored proc accross multiple CPU's?

    - by Russ
    [Also on SuperUser - http://superuser.com/questions/116600/can-i-spead-out-a-long-running-stored-proc-accross-multiple-cpus] I have a stored procedure in SQL server the gets, and decrypts a block of data. ( Credit cards in this case. ) Most of the time, the performance is tolerable, but there are a couple customers where the process is painfully slow, taking literally 1 minute to complete. ( Well, 59377ms to return from SQL Server to be exact, but it can vary by a few hundred ms based on load ) When I watch the process, I see that SQL is only using a single proc to perform the whole process, and typically only proc 0. Is there a way I can change my stored proc so that SQL can multi-thread the process? Is it even feasible to cheat and to break the calls in half, ( top 50%, bottom 50% ), and spread the load, as a gross hack? ( just spit-balling here ) My stored proc: USE [Commerce] GO /****** Object: StoredProcedure [dbo].[GetAllCreditCardsByCustomerId] Script Date: 03/05/2010 11:50:14 ******/ SET ANSI_NULLS ON GO SET QUOTED_IDENTIFIER ON GO ALTER PROCEDURE [dbo].[GetAllCreditCardsByCustomerId] @companyId UNIQUEIDENTIFIER, @DecryptionKey NVARCHAR (MAX) AS SET NoCount ON DECLARE @cardId uniqueidentifier DECLARE @tmpdecryptedCardData VarChar(MAX); DECLARE @decryptedCardData VarChar(MAX); DECLARE @tmpTable as Table ( CardId uniqueidentifier, DecryptedCard NVarChar(Max) ) DECLARE creditCards CURSOR FAST_FORWARD READ_ONLY FOR Select cardId from CreditCards where companyId = @companyId and Active=1 order by addedBy desc --2 OPEN creditCards --3 FETCH creditCards INTO @cardId -- prime the cursor WHILE @@Fetch_Status = 0 BEGIN --OPEN creditCards DECLARE creditCardData CURSOR FAST_FORWARD READ_ONLY FOR select convert(nvarchar(max), DecryptByCert(Cert_Id('Oh-Nay-Nay'), EncryptedCard, @DecryptionKey)) FROM CreditCardData where cardid = @cardId order by valueOrder OPEN creditCardData FETCH creditCardData INTO @tmpdecryptedCardData -- prime the cursor WHILE @@Fetch_Status = 0 BEGIN print 'CreditCardData' print @tmpdecryptedCardData set @decryptedCardData = ISNULL(@decryptedCardData, '') + @tmpdecryptedCardData print '@decryptedCardData' print @decryptedCardData; FETCH NEXT FROM creditCardData INTO @tmpdecryptedCardData -- fetch next END CLOSE creditCardData DEALLOCATE creditCardData insert into @tmpTable (CardId, DecryptedCard) values ( @cardId, @decryptedCardData ) set @decryptedCardData = '' FETCH NEXT FROM creditCards INTO @cardId -- fetch next END select CardId, DecryptedCard FROM @tmpTable CLOSE creditCards DEALLOCATE creditCards

    Read the article

  • Can I spread out a long running stored proc accross multiple CPU's?

    - by Russ
    [Also on SuperUser - http://superuser.com/questions/116600/can-i-spead-out-a-long-running-stored-proc-accross-multiple-cpus] I have a stored procedure in SQL server the gets, and decrypts a block of data. ( Credit cards in this case. ) Most of the time, the performance is tolerable, but there are a couple customers where the process is painfully slow, taking literally 1 minute to complete. ( Well, 59377ms to return from SQL Server to be exact, but it can vary by a few hundred ms based on load ) When I watch the process, I see that SQL is only using a single proc to perform the whole process, and typically only proc 0. Is there a way I can change my stored proc so that SQL can multi-thread the process? Is it even feasible to cheat and to break the calls in half, ( top 50%, bottom 50% ), and spread the load, as a gross hack? ( just spit-balling here ) My stored proc: USE [Commerce] GO /****** Object: StoredProcedure [dbo].[GetAllCreditCardsByCustomerId] Script Date: 03/05/2010 11:50:14 ******/ SET ANSI_NULLS ON GO SET QUOTED_IDENTIFIER ON GO ALTER PROCEDURE [dbo].[GetAllCreditCardsByCustomerId] @companyId UNIQUEIDENTIFIER, @DecryptionKey NVARCHAR (MAX) AS SET NoCount ON DECLARE @cardId uniqueidentifier DECLARE @tmpdecryptedCardData VarChar(MAX); DECLARE @decryptedCardData VarChar(MAX); DECLARE @tmpTable as Table ( CardId uniqueidentifier, DecryptedCard NVarChar(Max) ) DECLARE creditCards CURSOR FAST_FORWARD READ_ONLY FOR Select cardId from CreditCards where companyId = @companyId and Active=1 order by addedBy desc --2 OPEN creditCards --3 FETCH creditCards INTO @cardId -- prime the cursor WHILE @@Fetch_Status = 0 BEGIN --OPEN creditCards DECLARE creditCardData CURSOR FAST_FORWARD READ_ONLY FOR select convert(nvarchar(max), DecryptByCert(Cert_Id('Oh-Nay-Nay'), EncryptedCard, @DecryptionKey)) FROM CreditCardData where cardid = @cardId order by valueOrder OPEN creditCardData FETCH creditCardData INTO @tmpdecryptedCardData -- prime the cursor WHILE @@Fetch_Status = 0 BEGIN print 'CreditCardData' print @tmpdecryptedCardData set @decryptedCardData = ISNULL(@decryptedCardData, '') + @tmpdecryptedCardData print '@decryptedCardData' print @decryptedCardData; FETCH NEXT FROM creditCardData INTO @tmpdecryptedCardData -- fetch next END CLOSE creditCardData DEALLOCATE creditCardData insert into @tmpTable (CardId, DecryptedCard) values ( @cardId, @decryptedCardData ) set @decryptedCardData = '' FETCH NEXT FROM creditCards INTO @cardId -- fetch next END select CardId, DecryptedCard FROM @tmpTable CLOSE creditCards DEALLOCATE creditCards

    Read the article

  • Can I spread out a long running stored proc accross multiple CPU's?

    - by Russ
    [Also on SuperUser - http://superuser.com/questions/116600/can-i-spead-out-a-long-running-stored-proc-accross-multiple-cpus] I have a stored procedure in SQL server the gets, and decrypts a block of data. ( Credit cards in this case. ) Most of the time, the performance is tolerable, but there are a couple customers where the process is painfully slow, taking literally 1 minute to complete. ( Well, 59377ms to return from SQL Server to be exact, but it can vary by a few hundred ms based on load ) When I watch the process, I see that SQL is only using a single proc to perform the whole process, and typically only proc 0. Is there a way I can change my stored proc so that SQL can multi-thread the process? Is it even feasible to cheat and to break the calls in half, ( top 50%, bottom 50% ), and spread the load, as a gross hack? ( just spit-balling here ) My stored proc: USE [Commerce] GO /****** Object: StoredProcedure [dbo].[GetAllCreditCardsByCustomerId] Script Date: 03/05/2010 11:50:14 ******/ SET ANSI_NULLS ON GO SET QUOTED_IDENTIFIER ON GO ALTER PROCEDURE [dbo].[GetAllCreditCardsByCustomerId] @companyId UNIQUEIDENTIFIER, @DecryptionKey NVARCHAR (MAX) AS SET NoCount ON DECLARE @cardId uniqueidentifier DECLARE @tmpdecryptedCardData VarChar(MAX); DECLARE @decryptedCardData VarChar(MAX); DECLARE @tmpTable as Table ( CardId uniqueidentifier, DecryptedCard NVarChar(Max) ) DECLARE creditCards CURSOR FAST_FORWARD READ_ONLY FOR Select cardId from CreditCards where companyId = @companyId and Active=1 order by addedBy desc --2 OPEN creditCards --3 FETCH creditCards INTO @cardId -- prime the cursor WHILE @@Fetch_Status = 0 BEGIN --OPEN creditCards DECLARE creditCardData CURSOR FAST_FORWARD READ_ONLY FOR select convert(nvarchar(max), DecryptByCert(Cert_Id('Oh-Nay-Nay'), EncryptedCard, @DecryptionKey)) FROM CreditCardData where cardid = @cardId order by valueOrder OPEN creditCardData FETCH creditCardData INTO @tmpdecryptedCardData -- prime the cursor WHILE @@Fetch_Status = 0 BEGIN print 'CreditCardData' print @tmpdecryptedCardData set @decryptedCardData = ISNULL(@decryptedCardData, '') + @tmpdecryptedCardData print '@decryptedCardData' print @decryptedCardData; FETCH NEXT FROM creditCardData INTO @tmpdecryptedCardData -- fetch next END CLOSE creditCardData DEALLOCATE creditCardData insert into @tmpTable (CardId, DecryptedCard) values ( @cardId, @decryptedCardData ) set @decryptedCardData = '' FETCH NEXT FROM creditCards INTO @cardId -- fetch next END select CardId, DecryptedCard FROM @tmpTable CLOSE creditCards DEALLOCATE creditCards

    Read the article

  • How to make a mapped field inherited from a superclass transient in JPA?

    - by Russ Hayward
    I have a legacy schema that cannot be changed. I am using a base class for the common features and it contains an embedded object. There is a field that is normally mapped in the embedded object that needs to be in the persistence id for only one (of many) subclasses. I have made a new id class that includes it but then I get the error that the field is mapped twice. Here is some example code that is much simplified to maintain the sanity of the reader: @MappedSuperclass class BaseClass { @Embedded private Data data; } @Entity class SubClass extends BaseClass { @EmbeddedId private SubClassId id; } @Embeddable class Data { private int location; private String name; } @Embeddable class SubClassId { private int thingy; private int location; } I have tried @AttributeOverride but I can only get it to rename the field. I have tried to set it to updatable = false, insertable = false but this did not seem to work when used in the @AttributeOverride annotation. See answer below for the solution to this issue. I realise I could change the base class but I really do not want to split up the embedded object to separate the shared field as it would make the surrounding code more complex and require some ugly wrapping code. I could also redesign the whole system for this corner case but I would really rather not. I am using Hibernate as my JPA provider.

    Read the article

  • How can I find installed programs dependent on any version of the .NET framework?

    - by Russ
    I have an application that uses .NET 3.5SP1, but I have been having a LOT of random crashes with it. I am starting to narrow the fields of possible causes to the framework itself, where I suspect some other app is installing a lower patch version. Is there any apps in the wild, or anything I can slap together that can tell me what apps that are installed that depend on .NET to run? Their minimum required version would be nice to know also, but not necessary.

    Read the article

  • Python - Number of Significant Digits in results of division

    - by russ
    Newbie here. I have the following code: myADC = 128 maxVoltage = 5.0 maxADC = 255.0 VoltsPerADC = maxVoltage/maxADC myVolts = myADC * VoltsPerADC print "myADC = {0: >3}".format(myADC) print "VoltsPerADC = {0: >7}".format(VoltsPerADC) print VoltsPerADC print "myVolts = {0: >7}".format(myVolts) print myVolts This outputs the following: myADC = 128 VoltsPerADC = 0.0196078 0.0196078431373 myVolts = 2.5098 2.50980392157 I have been searching for an explanation of how the number of significant digits is determined by default, but have had trouble locating an explanation that makes sense to me. This link link text suggests that by default the "print" statement prints numbers to 10 significant figures, but that does not seem to be the case in my results. How are the number of significant digits/precision determined? Can someone shed some light on this for me. Thanks in advance for your time and patience.

    Read the article

  • Auto accept outlook VBA

    - by Russ
    Is there a VB macro or some sort of add-on out there that will allow me to auto accept invitations in outlook by sender or by folder? I was thinking about doing a VB script for this but I don't want to re-invent the wheel?

    Read the article

  • WinDev user standpoint

    - by Russ T
    I am not developer but I found this board while researching windev. I am the VP of a start up company that is developing a web-based application for a retail point of sale system. If anyone out there is familiar enough with this product and especially it's limitations, I would appreciate any heads up info on any of the final product(s) that have been developed that I might be able to compare/contrast to our enterprise.

    Read the article

< Previous Page | 1 2 3 4 5  | Next Page >