I read this question: Command Line Parser for .NET.
I thought that was what I was looking for, but the library Command Line Parser Library is not Compact framework friendly...
I REALLY don't want to write a CL parser and I have been drifting away from the real purpose of my little app because of this unfortunate trial.
Does someone know of a library that fits the compact-framework? (preferably with simplicity and functionality like the one mentioned above)
Does not matter whether version 2 or 3.5
I developed this framework, maybe it helps:
The SysCommand is a powerful cross-platform framework, to develop Console Applications in .NET. Is simple, type-safe, and with great influences of the MVC pattern.
https://github.com/juniorgasparotto/SysCommand
namespace Example.Initialization.Simple
{
using SysCommand.ConsoleApp;
public class Program
{
public static int Main(string[] args)
{
return App.RunApplication();
}
}
// Classes inheriting from `Command` will be automatically found by the system
// and its public properties and methods will be available for use.
public class MyCommand : Command
{
public void Main(string arg1, int? arg2 = null)
{
if (arg1 != null)
this.App.Console.Write(string.Format("Main arg1='{0}'", arg1));
if (arg2 != null)
this.App.Console.Write(string.Format("Main arg2='{0}'", arg2));
}
public void MyAction(bool a)
{
this.App.Console.Write(string.Format("MyAction a='{0}'", a));
}
}
}
Tests:
// auto-generate help
$ my-app.exe help
// method "Main" typed
$ my-app.exe --arg1 value --arg2 1000
// or without "--arg2"
$ my-app.exe --arg1 value
// actions support
$ my-app.exe my-action -a
This is what I'm using. I borrowed it from somewhere, but not sure where:
using System.Collections.Specialized;
using System.Text.RegularExpressions;
/// <summary>
/// Parses the command line arguments into a name/value collection
/// </summary>
public class CommandLineArgumentParser
{
#region Fields
private StringDictionary parameters;
#endregion
#region Constructors
/// <summary>
/// Initializes a new instance of the <see cref="CommandLineArgumentParser"/> class.
/// </summary>
/// <param name="args">command-line arguments
/// </param>
public CommandLineArgumentParser(string[] args)
{
this.parameters = new StringDictionary();
Regex spliter = new Regex(#"^-{1,2}|^/|=|:", RegexOptions.IgnoreCase | RegexOptions.Compiled);
Regex remover = new Regex(#"^['""]?(.*?)['""]?$", RegexOptions.IgnoreCase | RegexOptions.Compiled);
string parameter = null;
string[] parts;
// Valid parameters forms:
// {-,/,--}param{ ,=,:}((",')value(",'))
// Examples:
// -param1 value1 --param2 /param3:"Test-:-work"
// /param4=happy -param5 '--=nice=--'
foreach (string txt in args)
{
// Look for new parameters (-,/ or --) and a
// possible enclosed value (=,:)
parts = spliter.Split(txt, 3);
switch (parts.Length)
{
// Found a value (for the last parameter
// found (space separator))
case 1:
if (parameter != null)
{
if (!this.parameters.ContainsKey(parameter))
{
parts[0] = remover.Replace(parts[0], "$1");
this.parameters.Add(parameter, parts[0]);
}
parameter = null;
}
// else Error: no parameter waiting for a value (skipped)
break;
// Found just a parameter
case 2:
// The last parameter is still waiting.
// With no value, set it to true.
if (parameter != null)
{
if (!this.parameters.ContainsKey(parameter))
{
this.parameters.Add(parameter, "true");
}
}
parameter = parts[1];
break;
// Parameter with enclosed value
case 3:
// The last parameter is still waiting.
// With no value, set it to true.
if (parameter != null)
{
if (!this.parameters.ContainsKey(parameter))
{
this.parameters.Add(parameter, "true");
}
}
parameter = parts[1];
// Remove possible enclosing characters (",')
if (!this.parameters.ContainsKey(parameter))
{
parts[2] = remover.Replace(parts[2], "$1");
this.parameters.Add(parameter, parts[2]);
}
parameter = null;
break;
}
}
// In case a parameter is still waiting
if (parameter != null)
{
if (!this.parameters.ContainsKey(parameter))
{
this.parameters.Add(parameter, "true");
}
}
}
#endregion
#region Properties
/// <summary>
/// Gets a count of command line arguments
/// </summary>
public int Count
{
get
{
return this.parameters.Count;
}
}
/// <summary>
/// Gets the value with the given parameter name
/// </summary>
/// <param name="param">name of the parameter</param>
/// <returns>the value of the parameter</returns>
public string this[string param]
{
get
{
return this.parameters[param];
}
}
#endregion
}
http://commandline.codeplex.com/ I've used this so many times I've lost count. Maybe it works for CE. If not, it'll provide a fantastic starting point.
Related
Can anyone explain in simple terms what the 'forwarding' concept means in AspNetCore authentictation ? In the source code, comments on AuthenticationSchemeOptions class explain that the ForwardAuthenticate property is "If set, this specifies the target scheme that this scheme should forward AuthenticateAsync calls to"... What exactly does that mean ? In what secnario would this happen ?
According to your source codes, you could find, it just remove the previous scheme and use the Forward scheme to authenticate the user.
/// </summary>
/// <param name="scheme">The scheme to forward. One of ForwardAuthenticate, ForwardChallenge, ForwardForbid, ForwardSignIn, or ForwardSignOut.</param>
/// <returns>The forwarded scheme or <see langword="null"/>.</returns>
protected virtual string? ResolveTarget(string? scheme)
{
var target = scheme ?? Options.ForwardDefaultSelector?.Invoke(Context) ?? Options.ForwardDefault;
// Prevent self targetting
return string.Equals(target, Scheme.Name, StringComparison.Ordinal)
? null
: target;
}
/// <inheritdoc />
public async Task<AuthenticateResult> AuthenticateAsync()
{
var target = ResolveTarget(Options.ForwardAuthenticate);
if (target != null)
{
return await Context.AuthenticateAsync(target);
}
// Calling Authenticate more than once should always return the original value.
var result = await HandleAuthenticateOnceAsync() ?? AuthenticateResult.NoResult();
if (result.Failure == null)
{
var ticket = result.Ticket;
if (ticket?.Principal != null)
{
Logger.AuthenticationSchemeAuthenticated(Scheme.Name);
}
else
{
Logger.AuthenticationSchemeNotAuthenticated(Scheme.Name);
}
}
else
{
Logger.AuthenticationSchemeNotAuthenticatedWithFailure(Scheme.Name, result.Failure.Message);
}
return result;
}
The usage for this is, if one of the old existing scheme is should be replaced by a new one, by using this option could avoid a lot of code changing.
I'm using EPPlus as a calculations server. Here is my code:
using (var xlp = new ExcelPackage(stream))
{
OfficeOpenXml.ExcelWorksheet Sheet = xlp.Workbook.Worksheets["sheet1"];
//Some code for feeding user data to excel sheet
//...
//We first invoke calculate method to let contraints of data validation get updated.
xlp.Workbook.Calculate();
var v = Sheet.DataValidations["A1"];
if (v != null)
{
switch (v.ValidationType.Type)
{
case OfficeOpenXml.DataValidation.eDataValidationType.DateTime:
OfficeOpenXml.DataValidation.ExcelDataValidationDateTime V1 = (OfficeOpenXml.DataValidation.ExcelDataValidationDateTime)v;
try
{
//this line doesn't do any thing
V1.Validate();
}
catch
{
}
break;
case ...
}
}
}
I had read somewhere that Validate() method throws exception for invalid data. It doesn't.
My question: How to use the Validate() method?
That would depend on what the content of the cell and the settings of the validator's Operator:
http://epplus.codeplex.com/SourceControl/latest#EPPlus/DataValidation/ExcelDataValidationOperator.cs
/// <summary>
/// Operator for comparison between Formula and Formula2 in a validation.
/// </summary>
public enum ExcelDataValidationOperator
{
any,
equal,
notEqual,
lessThan,
lessThanOrEqual,
greaterThan,
greaterThanOrEqual,
between,
notBetween
}
The ExcelDataValidationDateTime (eventually) derives from ExcelDataValidationWithFormula<IExcelDataValidationFormulaDateTime> which contains the implemenation of Validate():
http://epplus.codeplex.com/SourceControl/latest#EPPlus/DataValidation/ExcelDataValidationWithFormula.cs
public override void Validate()
{
base.Validate();
if (Operator == ExcelDataValidationOperator.between || Operator == ExcelDataValidationOperator.notBetween)
{
if (string.IsNullOrEmpty(Formula2Internal))
{
throw new InvalidOperationException("Validation of " + Address.Address + " failed: Formula2 must be set if operator is 'between' or 'notBetween'");
}
}
}
So it will throw an exception (invalidate) when the validation operation is either ExcelDataValidationOperator.between or ExcelDataValidationOperator.notBetween and Forumla2 is not set (not to be confused with the primary Formula). In other words, it considers the validator invalid when you are using an operation which requires TWO values/formulas to compare but only one is set.
I am able to use log4net to send logging information to an email address using the smtpappender and a Gmail account in a VB solution (Visual Studio 2010). The recipient is configured in the log4net config file, however I would like to be able to change the recipient email address dynamically.
Is it possible without having to write a custom smtpappender?
Wether the answer is yes or no, please give me an example, preferably in VB.
It's not possible, the current SmtpAppender won't allow it. But you're lucky, the SendBuffer in the SmtpAppender can be overridden, so you can easily add some behavior to it. I think your best bet is to use the LoggingEvent properties to set the recipient:
public class MySmtpAppender : SmtpAppender
{
protected override void SendBuffer(log4net.Core.LoggingEvent[] events)
{
var Recipients = events
.Where(e => e.Properties.Contains("recipient"))
.Select(e => e.Properties["recipient"])
.Distinct();
var RecipientsAsASingleLine = string.Join(";", Recipients.ToArray()); // or whatever the separator is
var PreviousTo = To;
To = RecipientsAsASingleLine;
base.SendBuffer(events);
To = PreviousTo;
}
}
You may want to change the way to select recipients, your call.
edit The tool recommended by stuartd works quite well (well, it is quite a simple class, but still):
Public Class MySmtpAppender
Inherits SmtpAppender
Protected Overrides Sub SendBuffer(events As log4net.Core.LoggingEvent())
Dim Recipients = events.Where(Function(e) e.Properties.Contains("recipient")).[Select](Function(e) e.Properties("recipient")).Distinct()
Dim RecipientsAsASingleLine = String.Join(";", Recipients.ToArray())
' or whatever the separator is
Dim PreviousTo = [To]
[To] = RecipientsAsASingleLine
MyBase.SendBuffer(events)
[To] = PreviousTo
End Sub
End Class
it is possible. see my answer in this question - copied code below
using System;
using System.IO;
using System.Web.Mail;
using log4net.Layout;
using log4net.Core;
using log4net.Appender;
namespace SampleAppendersApp.Appender
{
/// <summary>
/// Simple mail appender that sends individual messages
/// </summary>
/// <remarks>
/// This SimpleSmtpAppender sends each LoggingEvent received as a
/// separate mail message.
/// The mail subject line can be specified using a pattern layout.
/// </remarks>
public class SimpleSmtpAppender : AppenderSkeleton
{
public SimpleSmtpAppender()
{
}
public string To
{
get { return m_to; }
set { m_to = value; }
}
public string From
{
get { return m_from; }
set { m_from = value; }
}
public PatternLayout Subject
{
get { return m_subjectLayout; }
set { m_subjectLayout = value; }
}
public string SmtpHost
{
get { return m_smtpHost; }
set { m_smtpHost = value; }
}
#region Override implementation of AppenderSkeleton
override protected void Append(LoggingEvent loggingEvent)
{
try
{
StringWriter writer = new StringWriter(System.Globalization.CultureInfo.InvariantCulture);
string t = Layout.Header;
if (t != null)
{
writer.Write(t);
}
// Render the event and append the text to the buffer
RenderLoggingEvent(writer, loggingEvent);
t = Layout.Footer;
if (t != null)
{
writer.Write(t);
}
MailMessage mailMessage = new MailMessage();
mailMessage.Body = writer.ToString();
mailMessage.From = m_from;
mailMessage.To = m_to;
if (m_subjectLayout == null)
{
mailMessage.Subject = "Missing Subject Layout";
}
else
{
StringWriter subjectWriter = new StringWriter(System.Globalization.CultureInfo.InvariantCulture);
m_subjectLayout.Format(subjectWriter, loggingEvent);
mailMessage.Subject = subjectWriter.ToString();
}
if (m_smtpHost != null && m_smtpHost.Length > 0)
{
SmtpMail.SmtpServer = m_smtpHost;
}
SmtpMail.Send(mailMessage);
}
catch(Exception e)
{
ErrorHandler.Error("Error occurred while sending e-mail notification.", e);
}
}
override protected bool RequiresLayout
{
get { return true; }
}
#endregion // Override implementation of AppenderSkeleton
private string m_to;
private string m_from;
private PatternLayout m_subjectLayout;
private string m_smtpHost;
}
}
You can use log4Net.GlobalContext class.
code:
App.config
<appender name="SmtpLogAppender" type="log4net.Appender.SmtpAppender">
<to type="log4net.Util.PatternString" value="%property{SenderList}"/>
C# Code
GlobalContext.Properties["SenderList"] = "abc#xyz.com, def#xyz.com";
log4net.Config.XmlConfigurator.Configure();
It is possible, to a certain degree, to get dynamic recipient.
In the SMTP appender the replacements for To, CC, From etc is done during configuration. Not ideal (would be best if it were to calculate the values at every send) but still workable.
Re-configuring the logging is not free but is doable programatically. You can set your To field as such :
<to type="log4net.Util.PatternString" value="SomeAccountThatReceivesAll#yourCorp.com%property{MailRecipient}" />
then in your code you can set a comma separated list of recipient like this :
log4net.GlobalContext.Properties["MailRecipient"] = "SomeOtherAccount#yourCorp.com,YourCorpSupportForThisApp#yourCorp.com";
the important bit is that you force a re-configuration AFTER you set these values. The exact syntax will depend on your config strategy, we use a central config file for all the logging so in C# it would look like this :
log4net.Config.XmlConfigurator.ConfigureAndWatch("PathToYourCentralFile.xml");
and voila ! Dynamic recipient without any custom appenders !
Personally I would prefer the custom appender, less hackish as it does not require constant re-configuring if you need to change them often. But if you only need the 10 minute fix and the config for recipient does not change once started then I found this to be good enough.
Does anyone know how to get access to the VSTO Application object of multiple instances of Visio 2010 using C#? Marshal.GetActiveObject() only returns the active instance.
There are forum posts on how to get multiple Application objects for all instances of Excel - where the author is using the technique to iterate through each Excel instance by process, get the native object model from a child window of the Excel instance, and finally to get the Application object.
This technique works well for for Excel, but I am unable to get a valid native OM (IDispatch or IAccessible) from the call to AccessibleObjectFromWindow() for a Visio window handle that would allow me to reference the parent Application object.
Here's my code snippet:
using VISIO = Microsoft.Office.Interop.Visio;
Process[] processes = Process.GetProcessesByName(Visio.APP_PROCESS_NAME);
foreach (Process process in processes)
{
int hwnd = (int)process.MainWindowHandle;
// We need to enumerate the child windows to find one that
// supports accessibility. To do this, instantiate the
// delegate and wrap the callback method in it, then call
// EnumChildWindows, passing the delegate as the 2nd arg.
if (hwnd != 0)
{
int hwndChild = 0;
cb = new EnumChildCallback(EnumChildProc);
EnumChildWindows(hwnd, cb, ref hwndChild);
// If we found an accessible child window, call
// AccessibleObjectFromWindow, passing the constant
// OBJID_NATIVEOM (defined in winuser.h) and
// IID_IDispatch - we want an IDispatch pointer
// into the native object model.
if (hwndChild != 0)
{
const uint OBJID_NATIVEOM = 0xFFFFFFF0;
Guid IID_IDispatch = new Guid("{00020400-0000-0000-C000-000000000046}");
EXL.Window ptr = null;
int hr = AccessibleObjectFromWindow(
hwndChild, OBJID_NATIVEOM,
IID_IDispatch.ToByteArray(), ref ptr);
if (hr >= 0)
{
// If we successfully got a native OM
// IDispatch pointer, we can QI this for
// an Viso Application (using the implicit
// cast operator supplied in the PIA).
VISIO.Application app = (VISIO.Application)ptr.Application;
allInstances.Add(app);
}
}
}
}
I found a solution that iterates through the Runtime Object Table (ROT) to get all monikers corresponding to Visio documents (".vdx") and get their corresponding VSO.Document object (using VSO=Microsoft.Office.Interop.Visio); with the VSO.Document object, I can get the VSO.Application object from the Document.Application property.
From this article by Andrew Baker: http://www.vbusers.com/codecsharp/codeget.asp?ThreadID=69&PostID=1, I created a ROTUtil utility/helper class:
/// <summary>
/// The COM running object table utility class.
/// </summary>
public class ROTUtil
{
#region APIs
[DllImport("ole32.dll")]
private static extern int GetRunningObjectTable(int reserved,
out UCOMIRunningObjectTable prot);
[DllImport("ole32.dll")]
private static extern int CreateBindCtx(int reserved,
out UCOMIBindCtx ppbc);
[DllImport("ole32.dll", PreserveSig = false)]
private static extern void CLSIDFromProgIDEx([MarshalAs(UnmanagedType.LPWStr)] string progId, out Guid clsid);
[DllImport("ole32.dll", PreserveSig = false)]
private static extern void CLSIDFromProgID([MarshalAs(UnmanagedType.LPWStr)] string progId, out Guid clsid);
[DllImport("ole32.dll")]
private static extern int ProgIDFromCLSID([In()]ref Guid clsid, [MarshalAs(UnmanagedType.LPWStr)]out string lplpszProgID);
#endregion
#region Public Methods
/// <summary>
/// Converts a COM class ID into a prog id.
/// </summary>
/// <param name="progID">The prog id to convert to a class id.</param>
/// <returns>Returns the matching class id or the prog id if it wasn't found.</returns>
public static string ConvertProgIdToClassId(string progID)
{
Guid testGuid;
try
{
CLSIDFromProgIDEx(progID, out testGuid);
}
catch
{
try
{
CLSIDFromProgID(progID, out testGuid);
}
catch
{
return progID;
}
}
return testGuid.ToString().ToUpper();
}
/// <summary>
/// Converts a COM class ID into a prog id.
/// </summary>
/// <param name="classID">The class id to convert to a prog id.</param>
/// <returns>Returns the matching class id or null if it wasn't found.</returns>
public static string ConvertClassIdToProgId(string classID)
{
Guid testGuid = new Guid(classID.Replace("!", ""));
string progId = null;
try
{
ProgIDFromCLSID(ref testGuid, out progId);
}
catch (Exception)
{
return null;
}
return progId;
}
/// <summary>
/// Get a snapshot of the running object table (ROT).
/// </summary>
/// <returns>A hashtable mapping the name of the object in the ROT to the corresponding object
/// <param name="filter">The filter to apply to the list (nullable).</param>
/// <returns>A hashtable of the matching entries in the ROT</returns>
public static Hashtable GetActiveObjectList(string filter)
{
Hashtable result = new Hashtable();
int numFetched;
UCOMIRunningObjectTable runningObjectTable;
UCOMIEnumMoniker monikerEnumerator;
UCOMIMoniker[] monikers = new UCOMIMoniker[1];
GetRunningObjectTable(0, out runningObjectTable);
runningObjectTable.EnumRunning(out monikerEnumerator);
monikerEnumerator.Reset();
while (monikerEnumerator.Next(1, monikers, out numFetched) == 0)
{
UCOMIBindCtx ctx;
CreateBindCtx(0, out ctx);
string runningObjectName;
monikers[0].GetDisplayName(ctx, null, out runningObjectName);
if (filter == null || filter.Length == 0 || runningObjectName.IndexOf(filter) != -1)
{
object runningObjectVal;
runningObjectTable.GetObject(monikers[0], out runningObjectVal);
result[runningObjectName] = runningObjectVal;
}
}
return result;
}
/// <summary>
/// Returns an object from the ROT, given a prog Id.
/// </summary>
/// <param name="progId">The prog id of the object to return.</param>
/// <returns>The requested object, or null if the object is not found.</returns>
public static object GetActiveObject(string progId)
{
// Convert the prog id into a class id
string classId = ConvertProgIdToClassId(progId);
UCOMIRunningObjectTable prot = null;
UCOMIEnumMoniker pMonkEnum = null;
try
{
int Fetched = 0;
// Open the running objects table.
GetRunningObjectTable(0, out prot);
prot.EnumRunning(out pMonkEnum);
pMonkEnum.Reset();
UCOMIMoniker[] pmon = new UCOMIMoniker[1];
// Iterate through the results
while (pMonkEnum.Next(1, pmon, out Fetched) == 0)
{
UCOMIBindCtx pCtx;
CreateBindCtx(0, out pCtx);
string displayName;
pmon[0].GetDisplayName(pCtx, null, out displayName);
Marshal.ReleaseComObject(pCtx);
if (displayName.IndexOf(classId) != -1)
{
// Return the matching object
object objReturnObject;
prot.GetObject(pmon[0], out objReturnObject);
return objReturnObject;
}
}
return null;
}
finally
{
// Free resources
if (prot != null)
Marshal.ReleaseComObject(prot);
if (pMonkEnum != null)
Marshal.ReleaseComObject(pMonkEnum);
}
}
Then here's the method in my class to get the instances:
/// <summary>
/// This strategy of getting the VSO.Application instances uses the appropriate file monikers
/// from the Runtime Object Table to get the VSO.Document object and hence the VSO.Document.Application
/// property.
/// </summary>
/// <param name="allInstances"></param>
protected void GetInstancesROTStrategy(IList<VSO.Application> allInstances)
{
// Iterate through all the objects in the ROT
string filter = ".vdx";
Hashtable runningObjects = ROTUtil.GetActiveObjectList(filter);
Hashtable appInstances = new Hashtable();
// Display the object ids
foreach (DictionaryEntry de in runningObjects)
{
// get visio document file monikers
string progId = de.Key.ToString();
object getObj = ROTUtil.GetActiveObject(progId);
if (getObj != null)
{
// try to cast the object to a VSO.Document
VSO.Document doc = getObj as VSO.Document;
if (doc != null)
{
VSO.Application app = doc.Application;
// only add to results IList if not duplicate
if (!appInstances.ContainsKey(app.WindowHandle32))
{
appInstances.Add(app.WindowHandle32, app);
allInstances.Add(app);
}
}
}
}
}
I am developing a windows app that performs some common TFS tasks using the 2010 Beta 2 API (like creating new team projects, new work items, selective build, etc. ).
In the process of editing existing work items, I should be able to automatically set the 'Reason' field's values according to state change of the WI (mimic-ing Visual Studio). (eg)- When I edit a bug, when state changes from Active to Resolved, the default Reason is 'Fixed' and similarly the default Reason='Deferred' when state goes from Active to Closed. (As defined in the work item type definition xml file. ) This transition is easy to capture and implement inside a simple event handler on the form, since the initial state will be Active when the Bug is edited for the first time.
I want to know how to implement the remaining transitions like Resolved to Closed (Reason=Fixed), Resolved to Active (Reason=Test failed/Not fixed) or Closed to Active (Reason=Reactivated/Regression).
I know there is a method called WorkItem.GetNextState(current_state,action), but this doesn't help as it requires a specific action.
What I have done so far is shown below:
void cmbBugState_SelectedIndexChanged(object sender, EventArgs e)
{
//private enum bugWorkFlows{"Fixed","Deferred","Duplicate","As Designed","Cannot Reproduce","Obsolete","Test Failed","Not Fixed","Reactivated","Regression"}
string[] activeToResolvedReasons = { "Fixed", "Deferred", "Duplicate", "As Designed", "Cannot Reproduce", "Obsolete" };
string[] resolvedToActiveReasons = { "Test Failed", "Not fixed" };
string[] resolvedToClosedReasons = activeToResolvedReasons;
string[] closedToActiveReasons = { "Reactivated", "Regression" };
string[] activeToClosedReasons = activeToResolvedReasons;
cmbBugReason.Items.AddRange(activeToResolvedReasons);
// Set the default reason according to change of state of the work item.
if (cmbBugState.SelectedItem.ToString() == "Resolved")
{
cmbBugReason.Enabled = true;
cmbBugReason.SelectedItem = activeToResolvedReasons[0];
}
if (cmbBugState.SelectedItem.ToString() == "Closed")
{
cmbBugReason.Enabled = true;
cmbBugReason.SelectedItem = activeToResolvedReasons[1];
}
}
Can anyone show how to handle these events on the form?
Thanks,
Tara.
I tried GetNextState. It was never reliable enough for what I needed.
So I "rolled my own" state transition code that has worked very well for me when I am moving from State "A" to State "B". It is a bit long, but it should have what you are looking for in it.
As a side note: Because this does not use the GetNextState method it has to get the next state somehow. The way it does this is by downloading the XML of the work item type in question. It parses that out and uses that to make a Transition list (_allTransistions).
The permissions levels in TFS 2010 needed to do this are: Team Foundation Administrators or Project Administrators. (As a side note, in TFS 2008 and 2005 all valid users could do this.)
The full code that uses this can be found in the WorkItemHelpers.cs file in the TFS Aggregator project on codeplex.
public static void TransitionToState(this WorkItem workItem, string state, string commentPrefix)
{
// Set the sourceWorkItem's state so that it is clear that it has been moved.
string originalState = (string)workItem.Fields["State"].Value;
// Try to set the state of the source work item to the "Deleted/Moved" state (whatever is defined in the file).
// We need an open work item to set the state
workItem.TryOpen();
// See if we can go directly to the planned state.
workItem.Fields["State"].Value = state;
if (workItem.Fields["State"].Status != FieldStatus.Valid)
{
// Revert back to the orginal value and start searching for a way to our "MovedState"
workItem.Fields["State"].Value = workItem.Fields["State"].OriginalValue;
// If we can't then try to go from the current state to another state. Saving each time till we get to where we are going.
foreach (string curState in workItem.Type.FindNextState((string)workItem.Fields["State"].Value, state))
{
string comment;
if (curState == state)
comment = commentPrefix + Environment.NewLine + " State changed to " + state;
else
comment = commentPrefix + Environment.NewLine + " State changed to " + curState + " as part of move toward a state of " + state;
bool success = ChangeWorkItemState(workItem, originalState, curState, comment);
// If we could not do the incremental state change then we are done. We will have to go back to the orginal...
if (!success)
break;
}
}
else
{
// Just save it off if we can.
string comment = commentPrefix + "\n State changed to " + state;
ChangeWorkItemState(workItem, originalState, state, comment);
}
}
private static bool ChangeWorkItemState(this WorkItem workItem, string orginalSourceState, string destState, String comment)
{
// Try to save the new state. If that fails then we also go back to the orginal state.
try
{
workItem.TryOpen();
workItem.Fields["State"].Value = destState;
workItem.History = comment;
workItem.Save();
return true;
}
catch (Exception)
{
// Revert back to the original value.
workItem.Fields["State"].Value = orginalSourceState;
return false;
}
}
/// <summary>
/// Used to find the next state on our way to a destination state.
/// (Meaning if we are going from a "Not-Started" to a "Done" state,
/// we usually have to hit a "in progress" state first.
/// </summary>
/// <param name="wiType"></param>
/// <param name="fromState"></param>
/// <param name="toState"></param>
/// <returns></returns>
public static IEnumerable<string> FindNextState(this WorkItemType wiType, string fromState, string toState)
{
var map = new Dictionary<string, string>();
var edges = wiType.GetTransitions().ToDictionary(i => i.From, i => i.To);
var q = new Queue<string>();
map.Add(fromState, null);
q.Enqueue(fromState);
while (q.Count > 0)
{
var current = q.Dequeue();
foreach (var s in edges[current])
{
if (!map.ContainsKey(s))
{
map.Add(s, current);
if (s == toState)
{
var result = new Stack<string>();
var thisNode = s;
do
{
result.Push(thisNode);
thisNode = map[thisNode];
} while (thisNode != fromState);
while (result.Count > 0)
yield return result.Pop();
yield break;
}
q.Enqueue(s);
}
}
}
// no path exists
}
private static readonly Dictionary<WorkItemType, List<Transition>> _allTransistions = new Dictionary<WorkItemType, List<Transition>>();
/// <summary>
/// Deprecated
/// Get the transitions for this <see cref="WorkItemType"/>
/// </summary>
/// <param name="workItemType"></param>
/// <returns></returns>
public static List<Transition> GetTransitions(this WorkItemType workItemType)
{
List<Transition> currentTransistions;
// See if this WorkItemType has already had it's transistions figured out.
_allTransistions.TryGetValue(workItemType, out currentTransistions);
if (currentTransistions != null)
return currentTransistions;
// Get this worktype type as xml
XmlDocument workItemTypeXml = workItemType.Export(false);
// Create a dictionary to allow us to look up the "to" state using a "from" state.
var newTransistions = new List<Transition>();
// get the transistions node.
XmlNodeList transitionsList = workItemTypeXml.GetElementsByTagName("TRANSITIONS");
// As there is only one transistions item we can just get the first
XmlNode transitions = transitionsList[0];
// Iterate all the transitions
foreach (XmlNode transitionXML in transitions)
{
// See if we have this from state already.
string fromState = transitionXML.Attributes["from"].Value;
Transition transition = newTransistions.Find(trans => trans.From == fromState);
if (transition != null)
{
transition.To.Add(transitionXML.Attributes["to"].Value);
}
// If we could not find this state already then add it.
else
{
// save off the transistion (from first so we can look up state progression.
newTransistions.Add(new Transition
{
From = transitionXML.Attributes["from"].Value,
To = new List<string> { transitionXML.Attributes["to"].Value }
});
}
}
// Save off this transition so we don't do it again if it is needed.
_allTransistions.Add(workItemType, newTransistions);
return newTransistions;
}