|
|
So I have it working where I can get the wwwroot from the controller, but I would like to be able to get it from a class static function that the controller calls.
I’m at the point where I have to decide whether to refactor my static class functions to pass the Environment, or search for a way to access it in the static functions, like for sending email and picking up a HTML template to populate.
This is sort of a dumb question, but is it possible to do?
Or should just rewrite my class functions and pass the Enviroment?
In addition, I need to get my settings in appsettings.json as well, in which I can access them in the view and controller, it not in a class, so the same question applies for that as well.
I’ve searched all over the Internet and It’s just examples of controller and view. On a side question, what’s with the binding options in configureServices in Startup.cs? I don’t have my computer in front of me to give better details about it.
Future Questions:
I just have a few more issues to go, and I should be able to complete my port over to core.
Cookies, core seems to use the single cookie value and not the cookie collection ; in which I posted a question on this earlier.
System.drawing Color, missing in that Nuget Package.
User-Agent
IpAddress
Server.MapPath - what is your work around?
A couple more
Warming;
Any typos or misspelling I will blame on my phone and not me.
I’m actually impressed with how well this works on my phone.
Job well done!
If it ain't broke don't fix it, I broke the rule on this.
Discover my world at jkirkerx.com
modified 29-Jun-18 2:30am.
|
|
|
|
|
Refactor the static methods and pass the IHostingEnvironment to them. Otherwise, they'll be impossible to test.
Similarly with the configuration: pass the IConfiguration (or the strongly-typed configuration object) to the methods.
Not sure what you mean by bindingOptions?
Multi-value cookies were never standardized, so Core doesn't support them. There are various options for handling multiple values suggested in this StackOverflow thread[^] - the Cookie Manager project[^] looks promising.
Several suggestions for Core image processing libraries on the .NET blog: .NET Core Image Processing | .NET Blog[^]
User agent should be available with: Request.Headers[HeaderNames.UserAgent]
How to get user Browser name ( user-agent ) in Asp.net Core? - Stack Overflow[^]
IP address is available from HttpContext.Connection.RemoteIpAddress - inject the IHttpContextAccessor service to get the current context.
Get Client IP Address in ASP.NET Core 2.0 - Edi Wang[^]
Server.MapPath doesn't really apply any more, since Core doesn't support virtual directories. You just use IHostingEnvironment[^] to get the physical root path, and then Path.Combine[^] to build the path to a folder or file within that path.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Things re much clearer now, thanks!
You gave me the info I needed to push ahead now.
This is the binding thing in ConfigureServices. I was trying to get the Settings and Configuration direct from the static class function and added the 2nd call. The 3rd call was from another online source. I suspect the 2nd and 3rd calls are not needed, but I'll have to comment them out and see what happens to determine that.
services.Configure(Configuration.GetSection("Settings"));
services.Configure(options => Configuration.GetSection("Settings").Bind(options));
services.AddSingleton(Configuration);
As for System.Drawing, I'm using CoreCompat.System.Drawing which works so far, it just have Color, but has Bitmap and stuff like that.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
I have 2 issues here on replacing MapPath. After reading many articles on the subject, guess I am the sloppy programmer that used MapPath inside functions to handle physical file locations for image and html file reading and writing.
On the first issue, I seeded my database with data that includes blobs of image bytes and then wrote the bytes to the folder as an image. I'm seeding my database in Program.cs which works fine, it creates the database but I can't grasp the MapPath or Environment.
I noticed in CreateWebHostBulder, if I add .ConfigureAppConfiguration((hostingContext, config) => that I can access the enviroment, so I can pass it to SeedData.Intialize(services, environment);, but then I can't access the DBcontext. But in the code below, I can access the DBContext, but can't figure out how to access the environment.
To the best of my understanding, I'm injecting the environment in startup.cs, which runs after program.cs. I'm lead to believe that I can't use or Initialize the enviroment in the actual SeedData either, so I thought I could pass it along.
public static void Main(string[] args)
{
var host = CreateWebHostBuilder(args).Build();
using (var scope = host.Services.CreateScope())
{
var services = scope.ServiceProvider;
try
{<br />
var context = services.GetRequiredService();
context.Database.Migrate();
SeedData.Initialize(services);
}
catch (Exception ex)
{
var logger = services.GetRequiredService>();
logger.LogError(ex, "An error occurred seeding the DB.");
}
}
host.Run();
}
On issue 2, I just can't seem to understand how create a class for MapPath. I tried this, the code below, but I'm testing on the Database Seed function so I know it's not injected yet because it's injected in startup.cs, and not program.cs.
So I'm flying around in circles, and not really sure how to approach this.
public class ServerPath
{
private IHostingEnvironment _hostingEnvironment;
public ServerPath(IHostingEnvironment environment)
{
_hostingEnvironment = environment;
}
public string MapPath(string path)
{
string filePath = string.Empty;
try
{
filePath = Path.Combine(_hostingEnvironment.WebRootPath, path);
}
catch(Exception e)
{
Console.WriteLine(e.Message.ToString());
}
return filePath;
}
public string CombinePath(string path, string fileName)
{
string filePath = string.Empty;
try
{
filePath = Path.Combine(_hostingEnvironment.WebRootPath, path + "\" + fileName);
}
catch (Exception e)
{
Console.WriteLine(e.Message.ToString());
}
return filePath;
}
}
}
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
Maybe just getting the current Directory and adding wwwroot to it is the way to go for seeding.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
The code in Startup runs as part of the BuildWebHost method in the Program class. That gets invoked both by the application, and by the design-time EF commands (add-migration, update-database, etc.), so you can't seed the data in the Startup class.
In 2.0 projects using EF Core 2.0, Program.BuildWebHost is invoked to obtain the application services. Unlike 1.x, this has the additional side effect of invoking Startup.Configure. If your 1.x app invoked database initialization code in its Configure method, unexpected problems can occur.
So the seeding code needs to be in Main, on in the context's OnModelCreating method[^].
In theory, you should be able to inject any registered service into your DbContext constructor, which would let you access your ServerPath service from the OnModelCreating method.
Alternatively, you can retrieve the services from the Main method using services.GetRequiredService<ServerPath>(); within the try block.
(Personally, I tend to keep the migrations and seed data in a separate console app, because I don't want the website connecting to SQL using an account that has permissions to create and modify databases.)
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Hmm....
I'll leave the seed in Main for now, and make sure everything works before reconsidering a change.
But I changed the file paths from Url types to physical types with Path.Combine and I successfully seeded my database with the blobs and Url paths and wrote the images to the folders. Getting past System.Drawing yesterday was whole nother issue.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
Last question of the day I hope.
I've been job hunting over the last 3 months. In my area there are all these Senior MVC Full Stack Developer job postings in which they ask for the following qualifications. I suspect it's all these different recruiting companies looking for one person in general, or someone considering leaving Silicon Valley and heading south.
MVC 5,
.Net Core
c#, VB
CSS, CSS3
HTML5
Javascript, JQuery
SQL Server, Stored Transactions
ORM, CRUD, Azure, Git, NodeJS
AngularJS 2-5
What I can't figure out is why the AngularJS with MVC. AngularJS seems to be a client side version of MVC. From what I have read, a crawler can't read an AngularJS page created 100% by AngularJS, so you have to use something that will download a static page first like MVC, then you can inject some AngularJS to say continue showing more cars.
I would think that ReactJS would be more suitable over AngularJS, being able to use the same server side controller and reducing code writing. Then say if the price of the car changes, the ReactJS polling will update that without a page refresh.
Last but not least, how the heck do these guys have the time to master all of these skills. Do these programmers really exist?
I didn't pickup Angular because I was busy learning Bootstrap 3 and 4, and writing responsive CSS.
I'm just trying to understand the posting, and would love to visit the job requester to see how they are using this and learn.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
You're querying the combination of angular and MVC, but not c# and vb.net? 
You get all sorts of rubbish in jobs ads, they're just throwing lots of tech in to cast their net wide and get as many applicants as possible, hoping to maybe get someone who knows every technology out there in the process. You won't really need all of those things though. It's up to you to decide if you want to spend time dealing with people who post adverts like that, or if you want to deal with people who are a bit more focused in what they want.
|
|
|
|
|
Good Reply. Guess non programmers are composing these job postings and just throwing out rubbish.
Perhaps I'm suppose to call them on it, or play the game.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
jkirkerx wrote: AngularJS 2-5
To clarify, the team announced that separate terms should be used for each framework with "AngularJS" referring to the 1.X versions and "Angular" without the "JS" referring to versions 2 and up.
I'm amazed they're not asking for "8+ years experience with ASP.NET Core", given it was released back in 2016. 
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Didn't know that.
I had one recruiter call me and ask me questions about Angular. In the end I asked here why and she wasn't able to answer the question. Then I said your not a programmer are you and she replied no. They must copy and paste these job requirements and claim them as their own.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
Well I have two functions called Startup and I just need one I think. One was stock called Startup(IConfiguration configuration) that gets called from Program.cs I think, and the other I added from examples in how to read the appsettings.json file called Startup(IHostingEnviroment env) and how to initialize the database.
So basically being so new here, I must of confused versions such as core 1 vs core 2
I would assume the first is correct.
public IConfiguration Configuration { get; }
public Startup(IConfiguration configuration)
{
Configuration = configuration;
}
And the 2nd one is incorrect for core 2.1
public Startup(IHostingEnvironment env)
{
var builder = new ConfigurationBuilder()
.SetBasePath(env.ContentRootPath)
.AddJsonFile("appsettings.json", optional: true, reloadOnChange: true)
.AddJsonFile($"appsettings.{env.EnvironmentName}.json", optional: true)
.AddEnvironmentVariables();
using (var client = new IndigoDBContext())
{
client.Database.EnsureCreated();
client.Database.Migrate();
}
Configuration = builder.Build();
}
But I don't see another place that has the environment except for Configure. Maybe the 2nd code goes there.
I'm trying to get my database created, and set migration.
Question is: I suspect I'm on the wrong track here, and I need a little help getting squared up on this. I probably have duplicates of or version overlaps here.
This is my ConfigureServices Function, I'm pretty sure I have this right
public void ConfigureServices(IServiceCollection services)
{
services.Configure<CookiePolicyOptions>(options =>
{
options.CheckConsentNeeded = context => true;
options.MinimumSameSitePolicy = SameSiteMode.None;
});
<pre>
services.AddMvc().SetCompatibilityVersion(CompatibilityVersion.Version_2_1)
.AddApplicationPart(typeof(ErrorController).GetTypeInfo().Assembly);
services.AddHttpContextAccessor();
services.Configure<AppConfiguration>(Configuration.GetSection("AppConfiguration"));
services.Configure<SmtpConfig>(Configuration.GetSection("SmtpConnection"));
services.TryAddSingleton<IHttpContextAccessor, HttpContextAccessor>();
services.TryAddSingleton<IActionContextAccessor, ActionContextAccessor>();
services.AddEntityFrameworkSqlite().AddDbContext<indigodbcontext>();
services.AddAuthentication(CookieAuthenticationDefaults.AuthenticationScheme).AddCookie();
services.AddHttpContextAccessor();
}
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
You're right - the IHostingEnvironment overload is not needed in core 2. The web host builder in Program.cs will initialize the configuration system for you:
Loads app configuration from:
- appsettings.json
- appsettings.{Environment}.json
- User secrets when the app runs in the Development environment using the entry assembly.
- Environment variables.
- Command-line arguments.
My DbContext configuration is slightly different, but I'm using full SQL Server:
services.AddDbContext<CmsDbContext>(options => options.UseSqlServer(Configuration.GetConnectionString("DefaultConnection"))); Configuring a DbContext - EF Core | Microsoft Docs[^]
You don't want to call EnsureCreated if you're using migrations - it doesn't use the migrations. Database.Migrate() will create the database if it doesn't exist.
asp.net - How and where to call Database.EnsureCreated and Database.Migrate? - Stack Overflow[^]
I notice you've got two services.AddHttpContextAccessor(); lines. You only need one of them.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
I got rid of the first one and kept the 2nd Startup(IHostingEnviroment env)
And I just deleted the 2nd services.AddHttpContextAccessor()
Then I noticed that my database was created, but in the root of the project solution and not in the wwwroot in which I created an App_Data folder. Not sure where to really put the Sqlite file.
I tried a similar DBContext to your, but it generated an error.
This works, well is created the database, not sure if it really works
services.AddEntityFrameworkSqlite().AddDbContext<IndigoDBContext>();
but when I use yours, I get a blank page
services.AddDbContext<IndigoDBContext>(options => options.UseSqlite(Configuration.GetConnectionString("SqLiteConnection")));
Maybe it has something to do with my DBContext OnConfiguring(DbContextOptionsBuilder optionsBuilder)
protected override void OnConfiguring(DbContextOptionsBuilder optionsBuilder)
{
var connectionStringBuilder = new SqliteConnectionStringBuilder { DataSource = "jkirkerx.sqlite;" };
var connectionString = connectionStringBuilder.ToString();
var connection = new SqliteConnection(connectionString);
optionsBuilder.UseSqlite(connection);
}
Well that's a different subject.
I'm satisfied with my Startup.cs at the moment now
Thanks Richard!
This is hard!, it's a totally different way of thinking. But I think I see the purpose of it. I predict that EG. I won't have to create a new DBContext in Linq everytime, and will just be able to use the one created in startup.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
When I wrote my first Angular app in MVC, I added a RenderSection in the Body element so I could add the AngularJS component.
This is in my _homeLayout View.
<body @RenderSection("AngularJS", required: false)>
But in Core 2.1, it complains about C# being there. I read some articles about tag helpers and looked at some work arounds but I can't think of a way or method to replace this with.
I knew this would come back and haunt me in the future. Just looking for a better idea.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
A tag helper would be a fairly simple way to do this:
[HtmlTargetElement(Attributes = ViewBagKeyAttributeName)]
public class HtmlElementAttributeTagHelper : TagHelper
{
private const string ViewBagKeyAttributeName = "asp-attributes-key";
[ViewContext]
[HtmlAttributeNotBound]
public ViewContext ViewContext { get; set; }
[HtmlAttributeName(ViewBagKeyAttributeName)]
public string AttributesKey { get; set; }
public override void Process(TagHelperContext context, TagHelperOutput output)
{
if (!string.IsNullOrWhiteSpace(AttributesKey))
{
var attributes = ViewContext.ViewData[AttributesKey];
switch (attributes)
{
case IDictionary<string, object> dict:
{
foreach (var attr in dict)
{
output.Attributes.Add(attr.Key, attr.Value);
}
break;
}
case object obj:
{
foreach (var attr in HtmlHelper.AnonymousObjectToHtmlAttributes(obj))
{
output.Attributes.Add(attr.Key, attr.Value);
}
break;
}
}
}
}
} In Views\_ViewImports.cshtml, add a line to register the tag helper:
...
@addTagHelper *, YourProjectAssemblyName In Views\Shared\Layout.cshtml, add the tag helper to the body:
<body asp-attributes-key="BodyAttributes"> In your controller or view, store the body attributes in the ViewData (or ViewBag, which is basically the same thing):
ViewData["BodyAttributes"] = new Dictionary<string, object>
{
["ng-app"] = "",
...
};
ViewBag.BodyAttributes = new
{
ng_app = "",
...
}; Any attributes you add to the ViewData collection will be combined with any existing attributes on the body element.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
I read up on Tag Helpers, and it looked so complicated compared to the old way.
So what folder would I place that in to stay organized?
Like ViewComponents?
FYI:
I'm slowly getting my stuff from the 2nd project wired up in the first project.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
I tend to put mine in a folder called "Helpers", but there's no specific requirement. Just so long as you're consistent, you can put them wherever you want.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
I decided to take up .Net Core 2.1 and port over a project to it, which now seems like a bad idea.
I've noticed that some of the support in system.web has been deleted such as
HttpContext to get querystrings, Request.Url, Session Value, Request.Headers for User Agent
HttpCookies
And reading the Web.Config file that doesn't exist anymore.
My original project design had a separate project that contained extra controllers, controller Attributes, security Attributes
and used the above to operate. it also contained the Entity Framework context, models, seeds, etc and needs a connection; and contains all the email stuff as well which requires access to appSettings.json.
I haven't been able to run the new app yet, for I'm refactoring the external .net core 2.1 project.
My question is, will I end up learning that I can't load dependency injections from the primary project to the secondary project?
Won't be able to load connection strings, and I'll have to move parts back to the primary project?
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
Maybe take advice from your own signature
|
|
|
|
|
No Joke because it's not broken.
This is for job training. Here in my area, Employers are looking for .Net MVC people that know MVC 5, Core 2 and AngularJS 2 through 5., NodeJS and more. I'm just trying to keep up with the bar since it's been raised again.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
jkirkerx wrote: My question is, will I end up learning that I can't load dependency injections from the primary project to the secondary project?
Simple answer: no.
If anything, DI is easier in ASP.NET Core than in previous versions, because it's baked-in. Just declare the dependencies in your constructor, and register your services in the Startup.ConfigureServices method, and the framework will wire everything together for you.
(If you don't want to explicitly register every single implementation of a service, you can use something like Scrutor[^] to find and register all implementations in an assembly.)
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Thanks Richard.
I didn't want to find out at the end that it could not be done and have to start again.
If it ain't broke don't fix it
Discover my world at jkirkerx.com
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
