I have the following code:
Process scriptProc = new Process();
scriptProc.StartInfo.FileName = #"cscript";
scriptProc.StartInfo.WorkingDirectory = #"C:\MyPath\";
scriptProc.StartInfo.Arguments = "filename.vbs //X";
scriptProc.StartInfo.WindowStyle = ProcessWindowStyle.Hidden;
scriptProc.Start();
scriptProc.WaitForExit();
scriptProc.Close();
My VBS opens in an editor(Visual Studio) which is specified by the //X attribute, but this only opens if the script has no syntax errors, it is not opening in the editor if I have script errors, which basically makes the use of the debugger as redundant.
Is there any way with which I can debug a VBScript using C# only?
A debugger is a tool for dealing with run-time errors. So it can't be used to check for compile-time errors.
Unfortunately, the c|wscript.exe script hosts don't have an option like Perl's -c (syntax check). Running cscript maybebad.vbs to catch syntax errors may be not convenient if that executes a flawless shutdown/format my harddisk/... script accidentally/unwittingly. You could write a script that Execute(Global) the code of maybebad.vbs with a WScript.Quit 1
prepended.
There is the MS ScriptControl that could be used to avoid the shelling out; I'm not sure, whether that will streamline your 'debugging experience'.
The code below uses #Ekkehard.Horner approaches. Compile it, then drag and drop .vbs files onto executable to test whether the file has syntax errors or not:
using System;
using System.IO;
using System.Reflection;
using System.Windows.Forms;
using System.Runtime.InteropServices;
// Add reference to COM Microsoft Script Control 1.0
// Code works for .Net 2.0 and above
class Program
{
static void Main(string[] args)
{
// Check whether a file was dragged onto executable
if (args.Length != 1)
{
MessageBox.Show("Drag'n'drop .vbs file onto this executable to check syntax");
return;
}
MessageBox.Show("Syntax will be checked for\r\n" + args[0]);
String vbscode = "";
// Read the content of the file
try
{
StreamReader sr = new StreamReader(args[0]);
vbscode = sr.ReadToEnd();
}
catch (Exception e)
{
MessageBox.Show("File reading error " + e.Message);
return;
}
// Add statement raising runtime error -2147483648 in the first line to ScriptControl
int hr = 0;
try
{
vbscode = "Err.Raise &H80000000\r\n" + vbscode;
MSScriptControl.ScriptControl sc = new MSScriptControl.ScriptControl();
sc.Language = "VBScript";
sc.AddCode(vbscode);
}
catch (Exception e)
{
hr = Marshal.GetHRForException(e);
// First line of code executed if no syntax errors only
if (hr == -2147483648)
{
// Run time error -2147483648 shows that execution started without syntax errors
MessageBox.Show("Syntax OK");
}
else
{
// Otherwise there are syntax errors
MessageBox.Show("Syntax error");
}
}
}
}
In answer to your question, no, I'm afraid you cannot debug the VBScript from within a debugging context of C#. Try debugging your script directly with something like http://www.vbsedit.com. By launching the script in C# first, you're complicating matters.
Related
I have a Windows Forms application that uses Selenium. I have multiple production clients that need to run this application and I’ve noticed that in every new client (and also when I need to update the webdriver) I need to copy and paste the .exe to the [PATH] location (%USERPROFILE%\AppData\Local\Microsoft\WindowsApps) and I want to automate that with the setup file that gets generated by Visual Studio every time I publish the application.
I’ve found that you can install an extension called “Microsoft Visual Studio Installer Project”, include the .exe file on it and either make a new Form that’ll check if the webdriver is in place and if it’s not to copy it, or I can change the [PATH] of my IWebDriver object in order to reflect the new path of this file. As a bonus you can also add the the desktop icon.
But first I want to know if there’s a way to publish this webdriver.exe file to it’s proper address through the “Publish wizard” parameters before I start looking for workarounds.
This worked for my use case, for context, I'm using a windows forms project targeting .NET (framework) 4.7.1. these are snippets from my events "load" and "show" formated as a different function. I only included the logic behind the file check, download and unzip with overwite. Since the System.IO.Compression.ZipFile class for this version of .NET doesn't natively support overwrite files, I used Ionic's DotNetZip package downloaded from NuGet.
using Ionic.Zip;
using System;
using System.Diagnostics;
using System.IO;
using System.Net;
private void DriverCheck(){
string edge, edgeVersion, edgeDriverPath, edgeDriver, edgeDriverVersion;
edge = #"C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe";
edgeVersion = FileVersionInfo.GetVersionInfo(edge).FileVersion;
edgeDriverPath = Environment.GetEnvironmentVariable("LocalAppData") + "\\Microsoft\\WindowsApps\\";
edgeDriver = edgeDriverPath + "msedgedriver.exe";
try
{
edgeDriverVersion = FileVersionInfo.GetVersionInfo(edgeDriver).FileVersion;
}
catch
{
edgeDriverVersion = null;
}
if (!File.Exists(edgeDriver) || edgeVersion != edgeDriverVersion)
{
try
{
using (var client = new WebClient())
{
string winver;
if (Environment.Is64BitProcess)
{
winver = "64";
}
else
{
winver = "32";
}
string zipPath = edgeDriverPath + "edgedriver_win64.zip";
client.DownloadFile("https://msedgedriver.azureedge.net/" + edgeVersion + "/edgedriver_win" + winver + ".zip", zipPath);
using (ZipFile zip = ZipFile.Read(zipPath))
{
foreach (ZipEntry temp in zip)
{
temp.Extract(edgeDriverPath, ExtractExistingFileAction.OverwriteSilently);
}
}
}
}
catch (Exception ex)
{
MessageBox.Show("Error downloading webdriver:\n" + ex.Message);
Application.Exit();
}
}
}
I have to modify an existing SSIS package, to launch it from a web site. Currently, this package is launched on demand by double clicking it, shows a form to ask for an excel file location and some database credentials that are stored in package variables, and then loads data from the excel file into a DB. Since there are many errors that can occur in the process, there is a package variable that holds an internal state, to inform the user which part of the process failed.
Given that I have to launch the package from a web site, as a first approach I have split the package in two, a master package that gets the information from user, executes the slave package by passing the user parameters through package variables, gets the child package internal state and then it finishes by informing the user the final state of this process. The communication between packages is being done by using variables with the same name and package configuration (main package variables). This is true for all variables except for the internal state one, that exists just in the parent, but is used in the child. Since both share the same context, it works ok.
Now that the child package is isolated, I'm trying to replace the master one with a C# asp site. Currently I'm able to get the user parameters through a webform and execute the package, but I can't figure how to read the child's internal state variable from the web app.
This internal value is an integer from 0 to 12, where 0 means ok and any other means that something went wrong with loading a table, executing a SP or something else.
There is a way to get this package variable value from the web app, when the package finishes? Otherwise, I just realized that this could be wrote in a log file that could be read by the web app, but I was wondering if there is a more wise solution.
Just to let you know, this is how I'm passing variables from the web app to the package. The package is configured to set its variables from primary/main package variables.
using System;
using System.IO;
using System.Collections;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using System.Web.UI;
using System.Web.UI.WebControls;
using Microsoft.SqlServer.Dts.Runtime;
public partial class _Default : System.Web.UI.Page
{
protected void Page_Load(object sender, EventArgs e)
{
}
protected void Button1_Click(object sender, EventArgs e)
{
Application app = new Application();
Package package = null;
String PackagePath = "";
try
{
string fileName = FileUpload1.PostedFile.FileName.ToString();
fileName = Server.MapPath("App_Data//" + System.IO.Path.GetFileName(fileName));
FileUpload1.PostedFile.SaveAs(fileName);
//Load DTSX
PackagePath = #"C:\Program Files\Microsoft SQL Server\100\DTS\Packages\Null Project\Package.dtsx";
package = app.LoadPackage(PackagePath, null);
//Global package variables (same name)
Hashtable param = new Hashtable();
param["ServidorOrigen"] = "SQL_SERVER";
param["UserOrigen"] = "user";
param["PassOrigen"] = "pass";
param["BaseDatosOrigen"] = "test_database";
param["EstadoConexion"] = 0;
param["EstadoPaquete"] = 0;
param["ExcelRuta"] = fileName.ToString();
Variables vars = null;
foreach (DictionaryEntry entry in param)
{
package.VariableDispenser.LockOneForWrite(entry.Key.ToString(), ref vars);
try
{
vars[entry.Key.ToString()].Value = entry.Value.ToString();
}
catch
{
throw new Exception("variable " + entry.Key.ToString() + " not found in package");
}
finally
{
vars.Unlock();
}
}
//Execute DTSX
Microsoft.SqlServer.Dts.Runtime.DTSExecResult results = package.Execute();
//Collects debugging info
using (StreamWriter _testData = new StreamWriter(Server.MapPath("~/App_Data/log.txt"), true))
{
if (!package.Errors.Count.Equals(0)){
_testData.WriteLine(package.Errors.Count.ToString()); // Write the file.
ErrorEnumerator myEnumerator = package.Errors.GetEnumerator();
int i = 0;
while ((myEnumerator.MoveNext()) && (myEnumerator.Current != null))
_testData.WriteLine("[{0}] {1}", i++, myEnumerator.Current.Description);
}
}
}
catch (Exception ex)
{
throw ex;
}
}
}
You can store the state in a database table the same way you would with your log file and then just have your web app read that at a given interval.
I'm not sure how you are passing your variables from the web app to the ssis, but you could look into the ssis configuration stuff storing in sql databases.
I have a similar thing I do.
Config stuff saved to database from web app.
Web app calls a sql job.
Job starts ssis package.
Web app queries every minute to see if the job has finished and returns succeeded or failed to user.
I've been browsing for a good hour and have yet to find something that would help with this. I'm working on opening AutoCAD from the .NET API in VS2013 using C#, but for some reason, I can never get AutoCAD to actually launch. I'm using the following code:
using System;
using System.Runtime.InteropServices;
using Autodesk.AutoCAD.Interop;
using Autodesk.AutoCAD.Runtime;
using Autodesk.AutoCAD.ApplicationServices;
namespace IOAutoCADHandler
{
public static class ACADDocumentManagement
{
[CommandMethod("ConnectToAcad")]
public static void ConnectToAcad()
{
AcadApplication acAppComObj = null;
// no version number so it will run with any version
const string strProgId = "AutoCAD.Application";
// Get a running instance of AutoCAD
try
{
acAppComObj = (AcadApplication)Marshal.GetActiveObject(strProgId);
}
catch // An error occurs if no instance is running
{
try
{
// Create a new instance of AutoCAD
acAppComObj = (AcadApplication)Activator.CreateInstance(Type.GetTypeFromProgID(strProgId), true);
}
catch //// STOPS HERE
{
// If an instance of AutoCAD is not created then message and exit
// NOTE: always shows this box and never opens AutoCAD
System.Windows.Forms.MessageBox.Show("Instance of 'AutoCAD.Application'" +
" could not be created.");
return;
}
}
// Display the application and return the name and version
acAppComObj.Visible = true;
System.Windows.Forms.MessageBox.Show("Now running " + acAppComObj.Name +
" version " + acAppComObj.Version);
// Get the active document
AcadDocument acDocComObj;
acDocComObj = acAppComObj.ActiveDocument;
// Optionally, load your assembly and start your command or if your assembly
// is demandloaded, simply start the command of your in-process assembly.
acDocComObj.SendCommand("(command " + (char)34 + "NETLOAD" + (char)34 + " " +
(char)34 + #"C:\Users\Administrator\Documents\All Code\main-libraries\IOAutoCADHandler\bin\Debug\IOAutoCADHandler.dll" + (char)34 + ") ");
acDocComObj.SendCommand("DRAWCOMPONENT");
}
}
Unfortunately, it always stops at the nested catch statement and always displays the popup box without opening AutoCAD. Any suggestions on how to at least make AutoCAD open for me?
EDIT: Error message
The issue is you're coding (correctly) to the AutoCAD interop interface. I recommend against that (due to potential version changes).
The other issue is that the documentation for AutoCAD plugins using the newer .net api is for plugins when AutoCAD is already running.
Final issue could be that the program Id of AutCAD is a mystery. I have resorted to making that a configurable setting, but default to "AutoCAD.Application", which will take the currently registered AutoCAD.Application on the production machine. If there are multiple versions installed on the machine and you want to be specific, then you could append the version number (which you'll need to research) to the ProgID like: "AutoCAD.Application.19", or "AutoCAD.Application.20" for 2015.
For the first issue, one technique is to use dynamics for the autoCad objects, particularly for creating instances. I have used the ObjectARX api for creating my application in a dummy project, and then switching to dynamics when I'm happy with the properties and method names.
In a standalone .Net application that starts AutoCAD you could use something like:
// I comment these out in production
//using Autodesk.AutoCAD.Interop;
//using Autodesk.AutoCAD.Interop.Common;
//...
//private static AcadApplication _application;
private static dynamic _application;
static string _autocadClassId = "AutoCAD.Application";
private static void GetAutoCAD()
{
_application = Marshal.GetActiveObject(_autocadClassId);
}
private static void StartAutoCad()
{
var t = Type.GetTypeFromProgID(_autocadClassId, true);
// Create a new instance Autocad.
var obj = Activator.CreateInstance(t, true);
// No need for casting with dynamics
_application = obj;
}
public static void EnsureAutoCadIsRunning(string classId)
{
if (!string.IsNullOrEmpty(classId) && classId != _autocadClassId)
_autocadClassId = classId;
Log.Activity("Loading Autocad: {0}", _autocadClassId);
if (_application == null)
{
try
{
GetAutoCAD();
}
catch (COMException ex)
{
try
{
StartAutoCad();
}
catch (Exception e2x)
{
Log.Error(e2x);
ThrowComException(ex);
}
}
catch (Exception ex)
{
ThrowComException(ex);
}
}
}
When there are several versions of AutoCAD installed on a computer, creating an instance with the ProgID "AutoCAD.Application" will run the latest version started on this computer by the current user. If the version of the Interop assemblies used does not match the version that is starting, you'll get a System.InvalidCastException with an HRESULT 0x80004002 (E_NOINTERFACE).
In your specific case, the {070AA05D-DFC1-4E64-8379-432269B48B07} IID in your error message is the GUID for the AcadApplicationinterface in R19 64-bit (AutoCAD 2013 & 2014). So there is an AutoCAD 2013 or 2014 that is starting, and you cannot cast this COM object to a 2015 type because 2015 is R20 (not binary compatible).
To avoid that, you can add a specific version to your ProgID (like "AutoCAD.Application.20" for AutoCAD 2015 (R20.0) to 2016 (R20.1)) to start the version matching your Interop assemblies or you can use late binding (eg. remove your references to Autodesk.AutoCAD.Interop* and use the dynamic keyword instead of the AutoCAD types).
In the last case, you will lost autocompletion, but your program will work with all the versions of AutoCAD.
Check also 32-bit vs 64-bit because TypeLib/Interop assemblies are not the same.
I open the application in a much straight-forward way. First, be sure to reference the correct type library. The one I am using is AutoCAD 2014 Type Library, located at:
c:\program files\common files\autodesk shared\acax19enu.tlb
To initialize the application:
using AutoCAD;
namespace test
{
class Program
{
static void Main(string[] args)
{
AutoCAD.AcadApplication app;
app = new AcadApplication();
app.Visible = true;
Console.Read();
}
}
}
Try this:
"sourcefile" is the original file
"newfile" is the new file
[CommandMethod("ModifyAndSaveas", CommandFlags.Redraw | CommandFlags.Session)]
public void ModifyAndSaveAs()
{
Document acDoc = Application.DocumentManager.Open(sourcefile);
Database acDB = acDoc.Database;
Transaction AcTran = acDoc.Database.TransactionManager.StartTransaction();
using (DocumentLock acLckDoc = acDoc.LockDocument())
{
using (AcTran)
{
BlockTable acBLT = (BlockTable)AcTran.GetObject(acDB.BlockTableId, OpenMode.ForRead);
BlockTableRecord acBLTR = (BlockTableRecord)AcTran.GetObject(acBLT[BlockTableRecord.ModelSpace], OpenMode.ForRead);
var editor = acDoc.Editor;
var SelectionSet = editor.SelectAll().Value;
foreach (ObjectId id in SelectionSet.GetObjectIds())
{
Entity ent = AcTran.GetObject(id, OpenMode.ForRead) as Entity;
//modify entities
}
AcTran.Commit();
}
}
acDB.SaveAs(newfile, DwgVersion.AC1021);
}
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Diagnostics;
using System.Runtime.InteropServices;
namespace Tekkit
{
class Program
{
static void Main(string[] args)
{
//make sure to add last 2 using statements
ProcessStartInfo start = new ProcessStartInfo("calc.exe");
Process.Start(start);//starts the process
}
}
}
I have not found yet a file-rename-function in .NET for C#, so I'm a bit confused how I would rename a file. I use the command prompt with Process.Start, but this isn't really professional and a black DOS window is popping up each time. Yes, I know there is something in the Visual Basic namespace, but this is not my intention to add the "visual-basic.dll" to my project.
I found some examples which "move" the file to rename it. It is a quite painful method and a shoddy workaround for things. Such footwork I can program myself.
Every language has renaming commands, so I am stunned that C# hasn't or I haven't found out yet. What is the right command?
For large files and to rename on CD, this code works, but your project will be partly converted into Visual Basic (as I understand it, maybe it is not so):
//Add the Microsoft.VisualBasic.MyServices reference and namespace in a project;
//For directories:
private static bool RenameDirectory(string DirPath, string NewName)
{
try
{
FileSystemProxy FileSystem = new Microsoft.VisualBasic.Devices.Computer().FileSystem;
FileSystem.RenameDirectory(DirPath, NewName);
FileSystem = null;
return true;
}
catch {
return false;
} //Just shut up the error generator of Visual Studio
}
//For files:
private static bool RenameFile(string FilePath, string NewName)
{
try
{
FileSystemProxy FileSystem = new Microsoft.VisualBasic.Devices.Computer().FileSystem;
FileSystem.RenameFile(FilePath, NewName);
FileSystem = null;
return true;
}
catch {
return false;
} //Just shut up the error generator of Visual Studio
}
A rename is just a move and vice versa, see the MSDN : File.Move
In the OS the operations are the same for all intents an purposes. That's why in explorer a move on the same partition is near instantaneous - just adjusts the file name and logical location. To Rename a file in the same directory you Move it to a new File Name in the same directory.
using System;
using System.IO;
class Test
{
public static void Main()
{
string path = #"c:\temp\MyTest.txt";
string path2 = #"c:\temp2\MyTest.txt";
try
{
if (!File.Exists(path))
{
// This statement ensures that the file is created,
// but the handle is not kept.
using (FileStream fs = File.Create(path)) {}
}
// Ensure that the target does not exist.
if (File.Exists(path2))
File.Delete(path2);
// Move the file.
File.Move(path, path2);
Console.WriteLine("{0} was moved/renamed to {1}.", path, path2);
// See if the original exists now.
if (File.Exists(path))
{
Console.WriteLine("The original file still exists, which is unexpected.");
}
else
{
Console.WriteLine("The original file no longer exists, which is expected.");
}
}
catch (Exception e)
{
Console.WriteLine("The process failed: {0}", e.ToString());
}
}
}
Using MSDN I got the class to write a wrapper for my command line tool.
I now am facing a problem, if I execute the exe through the command line with arguments, it works perfect without any errors.
But when I try to pass the arguments from the Wrapper it crashes the program.
Wanted to know if I am passing the arguments properly and if I am wrong, could somebody point out please.
This is the LaunchEXE class from MSDN
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Diagnostics;
using System.IO;
namespace SPDB
{
/// <summary>
/// Class to run any external command line tool with arguments
/// </summary>
public class LaunchEXE
{
internal static string Run(string exeName, string argsLine, int timeoutSeconds)
{
StreamReader outputStream = StreamReader.Null;
string output = "";
bool success = false;
try
{
Process newProcess = new Process();
newProcess.StartInfo.FileName = exeName;
newProcess.StartInfo.Arguments = argsLine;
newProcess.StartInfo.UseShellExecute = false;
newProcess.StartInfo.CreateNoWindow = true; //The command line is supressed to keep the process in the background
newProcess.StartInfo.RedirectStandardOutput = true;
newProcess.Start();
if (0 == timeoutSeconds)
{
outputStream = newProcess.StandardOutput;
output = outputStream.ReadToEnd();
newProcess.WaitForExit();
}
else
{
success = newProcess.WaitForExit(timeoutSeconds * 1000);
if (success)
{
outputStream = newProcess.StandardOutput;
output = outputStream.ReadToEnd();
}
else
{
output = "Timed out at " + timeoutSeconds + " seconds waiting for " + exeName + " to exit.";
}
}
}
catch (Exception e)
{
throw (new Exception("An error occurred running " + exeName + ".", e));
}
finally
{
outputStream.Close();
}
return "\t" + output;
}
}
}
This is the way I am passing arguments from my main program (Form1.cs)
private void button1_Click(object sender, EventArgs e)
{
string output;
output = LaunchEXE.Run(#"C:\Program Files (x86)\MyFolder\MyConsole.exe", "/BACKUP C:\\MyBackupProfile.txt", 100);
System.Windows.Forms.MessageBox.Show(output);
}
The command line tool accepts the following command and works perfectly:
C:\Program Files (x86)\MyFolder>MyConsole.exe /BACKUP C:\MyBackupProfile.txt
I have two options for you -
1) Please try running your Visual Studio on "administrator mode".
or 2) Try to implement this instead. https://github.com/commandlineparser/commandline.
"The Command Line Parser Library offers to CLR applications a clean and concise API for manipulating command line arguments and related tasks. It allows you to display an help screen with an high degree of customization and a simple way to report syntax errors to the user. Everything that is boring and repetitive to be programmed stands up on library shoulders, letting you concentrate yourself on core logic. This library provides hassle free command line parsing with a constantly updated API since 2005."
Worked great for me.
I have a feeling it doesn't like the spaces in your path. See this post: C# How to use Directory White Spaces into process.arguements?