I'm trying to read some configuration in my global.aspx Application_Start method. When I read ConfigurationManager.GetSection("system.web/httpHandlers") everything is fine:
ConfigurationManager.GetSection("system.web/httpHandlers")
{System.Web.Configuration.HttpHandlersSection}
base {System.Configuration.ConfigurationSection}: {System.Web.Configuration.HttpHandlersSection}
Handlers: Count = 48
But when I read ConfigurationManager.GetSection("system.webServer/handlers") (which contains my custom handlers, it returns null. What am I doing wrong?
The section looks like this:
<system.webServer>
<handlers>
<add verb="*" path="*.loc" name="LocalizedResourceStreamer"
type="CFW.WebUI.HttpHandlers.LocalizedResourceStreamer,WebUI" />
</handlers>
</system.webServer>
Notes:
Web.configs are nested, ConfigurationManager.GetSection takes nesting into account by default.
The overall problem is trying to see if *.loc files are being served.
So far:
Looks like the system.webServer is ignored.
Depending on your OS/setup, the system.webServer element may be configured to be ignored - and so the config system will not be constructing any inner configuration items from it. E.g. on my machine (WinXP, IIS 5.1), it's set to ignored by default.
Check the machine.config on the machine where this code is running, and see how the system.webServer element is configured. I don't have machines available with suitable later OSes at the moment, but it may be that this element is always set to be ignored - after all, that part of the config is for IIS' use, rather than our own.
try :
**p.s. my web.config contains : <httpHandlers> and not handlers as yours. change as necessarily :) - also the webserver vs system.web **
Configuration webConfig = WebConfigurationManager.OpenWebConfiguration(Request.ApplicationPath);
ConfigurationSection webConfigSections = webConfig.GetSection("system.web/httpHandlers");
if (webConfigSections != null)
{
// PropertyInformationCollection t = webConfigSections.ElementInformation.Properties;
XDocument xmlFile = XDocument.Load(new StringReader(webConfigSections.SectionInformation.GetRawXml()));
IEnumerable<XElement> query = from c in xmlFile.Descendants("add") select c;
foreach (XElement band in query)
{
}
}
p.s. thats the problem with this section - he doesnt have a uniquee element name that can be taken. thats why you take it whole("add" element) and parse it.
Related
I want ta change another Webconfig but I'm getting erors
What I tried is
string path = "E:\\username\\myprojects\\myproject\\Web.config";
Configuration webConfig = WebConfigurationManager.OpenWebConfiguration(path);
webConfig.AppSettings.Settings.Add("DBNAME", "DEV_TEMP");
webConfig.Save();
This said " Web.config denied relative virtual paths"
and another I tried on webconfig
<configuration>
<appSettings>
<add key="DBNAME" value="DEV_DEVELOPMENT"/>
</appSettings>
DBNAME value change to "DEV_TEMP"
XDocument doc=XDocument.Load("E:\\username\\myprojects\\myproject\\Web.config");
doc.Element("appSettings").Element("DBNAME").Value = "DEV_TEMP";
doc.Save("Web.config");
and it says Null reference
Assuming you have a good reason to do this, you could use something like:
var addElement = doc.Element("configuration").Element("appSettings").Elements().First(x => x.Attribute("key").Value == "DBNAME");
var valueAttribute = addElement.Attribute("value");
valueAttribute.Value = "DEV_TEMP";
Note that you also need to get the configuration element, and you can get the add element by selecting the element where the key attribute matches DBNAME as there could be multiple add elements. Also value is the attribute of add, not an element itself. See the XML Document Object Model (DOM)
for further information.
I have an ASP.NET MVC application which creates a huge dictionary. In order for it to work, I enabled the gcAllowVeryLargeObjects in my web.config under the <configuration> section:
<runtime>
<gcAllowVeryLargeObjects enabled="true" />
</runtime>
This works in my local environment, but when deploying it in another environment which is not under my control, they get System.OutOfMemoryException: Array dimension exceeds supported range, which is the same error I had before adding that configuration.
What I need
I'd like to check on runtime if gcAllowVeryLargeObjects was correctly read from the web.config.
Why I need it
I would like to make the application log if the gcAllowVeryLargeObjects mode is on, in order to verify that it is working in the other environment.
I think that maybe some configuration on their IIS server might be overriding or forcing the ignore of the setting in the app's web.config.
After 3 years still unanswered, and I had the same question. Took me 2 days, and finally I got it working:
Console.WriteLine("Is64BitProcess :" + Environment.Is64BitProcess); // this works on 64 bit processes only
// make sure that the flag is set, for using objects > 2GB:
Configuration configFile = ConfigurationManager.OpenExeConfiguration(ConfigurationUserLevel.None);
ConfigurationSection oRuntimeSection = configFile.GetSection("runtime");
string str = oRuntimeSection.SectionInformation.GetRawXml();
if (str == null)
{
oRuntimeSection.SectionInformation.SetRawXml("<runtime><gcAllowVeryLargeObjects enabled = \"true\" /></runtime>");
configFile.Save();
}
When I try to access configuration section that is encrypted and cannot be decrypted properly (for example, someone just grabbed config file from another machine blindly) - Configuration class is throwing an exception. I want to catch that situation and rewrite the section completely in such case.
I've tried to remove and add back the section in question, but it seems that removal is ignored - second statement in 'catch' throws another exception about such section already exists:
try
{
// this getter might throw an exception - e.g. if encrypted on another machine
var connStrings = config.ConnectionStrings;
}
catch (ConfigurationException)
{
config.Sections.Remove("connectionStrings");
config.Sections.Add("connectionStrings", new ConnectionStringsSection());
}
It might be related to the fact I have connectionStrings section residing in separate file, i.e. my config file has something like <connectionStrings configSource="connections.config"/>, while actual encrypted content is in the connections.config file.
Is it possible to do what I need without falling back to direct XML manipulations, by using .NET Configuration classes only?
I'm pretty sure this will do what you want. In my case, I just included a bogus connectionString setting:
<connectionStrings>
<add connectionString="foo"/>
</connectionStrings>
I didn't include a "name" property, so trying to read ConnectionStrings will blow up, just like in your case:
try {
var x = ConfigurationManager.ConnectionStrings; //blows up
}
catch {
var config = ConfigurationManager.OpenExeConfiguration(ConfigurationUserLevel.None);
config.Sections.Remove("connectionStrings");
config.Save(); //Now save the changes
ConfigurationManager.RefreshSection("connectionStrings"); //and reload the section
var x = ConfigurationManager.ConnectionStrings; //can now read!
}
You still won't be able to do this:
config.Sections.Add("connectionStrings", new ConnectionStringsSection());
Because (at least on my machine), it's picking up connectionStrings from the Machine.config.
I suspect though that in your case that's fine. You are now at a point where if you want to, you can add your own connection strings and you don't really need to completely blow away the connection string section.
I want to change the location where my application looks for the app.config file.
I know that I can use ConfigurationManager.OpenExeConfiguration() to access an arbitrary config file - however, when the .Net Framework reads the config file (for ConnectionStrings or EventSources, for instance), it will look at the default location. I want to actually change the location, globally for the entire .Net Framework (for my application, of course).
I also know that I can use AppDomainSetup to change the location of the app.config for a new AppDomain. However, that doesn't apply to the primary AppDomain of the application.
I also know that I can override function Main() and create a new AppDomain as above and run my application in that new AppDomain. However, that has other side-effects - for instance, Assembly.GetEntryAssembly() will return a null reference.
Given how everything else works in .Net, I would expect there to be some way to configure the startup environment of my application - via a Application Manifest, or some such - but I have been unable to find even a glimmer of hope in that direction.
Any pointer would be helpful.
David Mullin
I used the approach with starting another AppDomain from Main(), specifying the "new" location of the configuration file.
No issues with GetEntryAssembly(); it only returns null, when being called from unmanaged code - or at least it doesn't for me, as I use ExecuteAssembly() to create/run the second AppDomain, much like this:
int Main(string[] args)
{
string currentExecutable = Assembly.GetExecutingAssembly().Location;
bool inChild = false;
List<string> xargs = new List<string>();
foreach (string arg in xargs)
{
if (arg.Equals("-child"))
{
inChild = true;
}
/* Parse other command line arguments */
else
{
xargs.Add(arg);
}
}
if (!inChild)
{
AppDomainSetup info = new AppDomainSetup();
info.ConfigurationFile = /* Path to desired App.Config File */;
Evidence evidence = AppDomain.CurrentDomain.Evidence;
AppDomain domain = AppDomain.CreateDomain(friendlyName, evidence, info);
xargs.Add("-child"); // Prevent recursion
return domain.ExecuteAssembly(currentExecutable, evidence, xargs.ToArray());
}
// Execute actual Main-Code, we are in the child domain with the custom app.config
return 0;
}
Note that we are effectively rerunning the EXE, just as a AppDomain and with a different config. Also note that you need to have some "magic" option that prevents this from going on endlessly.
I crafted this out from a bigger (real) chunk of code, so it might not work as is, but should illustrate the concept.
I am not sure why you want to change the location of your config file - perhaps there can be different approach for solving your actual problem. I had a requirement where I wanted to share configuration file across related applications - I had chosen to use own xml file as it had given me extra benefit of having complete control over the schema.
In your case, it's possible to externalize sections of your config file to a separate file using configSource property. See here under "Using External Configuration Files" to check how it has been done for connection strings section. Perhaps, this may help you.
var configPath = YOUR_PATH;
if (!Directory.Exists(ProductFolder))
{
Directory.CreateDirectory(ProductFolder);
}
if (!File.Exists(configPath))
{
File.WriteAllText(configPath, Resources.App);
}
var map = new ExeConfigurationFileMap
{
ExeConfigFilename = configPath,
LocalUserConfigFilename = configPath,
RoamingUserConfigFilename = configPath
};
Configuration config = ConfigurationManager.OpenMappedExeConfiguration(map, ConfigurationUserLevel.None);
Then use config member as you want.
Another approach is to leave the config file with the executable file and move the relevant changeable sections to external xml files which can be in whatever location you choose.
If you are using your config file in a readonly capacity, then you can add the relevant chunks to an XML file in a different location using XML Inlcude. This won't work if you are trying to write values back directly to app.config using the Configuration.Save method.
app.config:
<?xml version="1.0"?>
<configuration xmlns:xi="http://www.w3.org/2001/XInclude">
<appSettings>
<xi:include href="AppSettings.xml"/>
</appSettings>
<connectionStrings>
<xi:include href="ConnectionStrings.xml"/>
</connectionStrings>
<startup><supportedRuntime version="v4.0" sku=".NETFramework,Version=v4.7"/></startup>
</configuration>
ConnectionStrings.xml:
<?xml version="1.0"?>
<add name="Example1ConnectionString"
connectionString="Data Source=(local)\SQLExpress;Initial Catalog=Example1DB;Persist Security Info=True;User ID=sa;Password=password"
providerName="System.Data.SqlClient" />
<add name="Example2ConnectionString"
connectionString="Data Source=(local)\SQLExpress;Initial Catalog=Example2DB;Persist Security Info=True;User ID=sa;Password=password"
providerName="System.Data.SqlClient" />
AppSettings.xml:
<?xml version="1.0"?>
<add key="Setting1" value="Value1"/>
<add key="Setting2" value="Value2"/>
A file URI looks like this:
file:///C:/whatever.txt
You can even define failover files in case the one you are trying to reference is missing. This pattern is from https://www.xml.com/pub/a/2002/07/31/xinclude.html:
<xi:include href="http://www.whitehouse.gov/malapropisms.xml">
<xi:fallback>
<para>
This administration is doing everything we can to end the stalemate in
an efficient way. We're making the right decisions to bring the solution
to an end.
</para>
</xi:fallback>
I've got the function for changing the values in web.config
but my problem is it is not getting the path of web.config correctly and throwing
"Could not find file 'C:\Users\maxnet25\Web.config'"
It was giving error on xmlDoc.Load() function.
My code:
public void UpdateConfigKey(string strKey, string newValue)
{
XmlDocument xmlDoc = new XmlDocument();
xmlDoc.Load(AppDomain.CurrentDomain.BaseDirectory + "..\\..\\Web.config");
if (!ConfigKeyExists(strKey))
{
throw new ArgumentNullException("Key", "<" + strKey + "> not find in the configuration.");
}
XmlNode appSettingsNode = xmlDoc.SelectSingleNode("configuration/appSettings");
foreach (XmlNode childNode in appSettingsNode)
{
if (childNode.Attributes["key"].Value == strKey)
childNode.Attributes["value"].Value = newValue;
}
xmlDoc.Save(AppDomain.CurrentDomain.BaseDirectory + "..\\..\\Web.config");
xmlDoc.Save(AppDomain.CurrentDomain.SetupInformation.ConfigurationFile);
Label1 .Text ="Key Upated Successfullly";
}
What error messsage is being given?
Either way, you're not really going about modifying web.config in the right way. You should probably take a look at the System.Configuration.ConfigurationManager class as this provides programmatic access to the web.config file in a structured manner. Note that to access this class you need to add a reference to System.Configuration.dll to your project to bring the ConfigurationManager into scope.
If you look at the example code for the GetSection method, it shows how to create/add settings in the appSettings section of a .net config file, so that example should be enough to get you where you want to go.
If you definately want to use this approach to manipulate your web.config file, I suspect that:
AppDomain.CurrentDomain.BaseDirectory + "..\\..\\Web.config")
is incorrect, based on the path that you've shown in the error message. Try removing the ..\..\ and seeing if that works. AppDomain.CurrentDomain.BaseDirectory should be pointing at the location of your web.config file without modification.
Assuming this is indeed an ASP.NET website, instead of this:
AppDomain.CurrentDomain.BaseDirectory + "..\\..\\Web.config"
Use this:
HttpContext.Current.Server.MapPath("~/Web.config")
On a side note, please be aware that anytime you make a change to web.config, your web application restarts. You might not need to worry about that depending on what your web app does though.
Try using Server.MapPath() to resolve the location of your web.config. If you're in a page, Server is one of the page properties. If not, you can find it in HttpContext.Current.
As an example...
HttpContext.Current.Server.MapPath("~/web.config")
...should return the physical path to the web.config at the top of your web application.
Now, you're probably much better off using the WebConfigurationManager, as shown in this post. The approach is much cleaner, but requires a reference to System.Configuration.
Have you added a web.config to your web site?
You should use either:
System.Configuration.ConfigurationManager
for app.config files, or:
System.Web.Configuration.WebConfigurationManager
for web.config files.
You can actually use System.Configuration.ConfigurationManager with web.config files as well, and to be honest, I'm not actually sure if there's any benefit for using one over the other.
But either way, you should not be using the Xml namespaces and writing/modifying the raw XML.