Hi

Please work on docker version 👍

currently, you can not create a new post without the required input fields being filled. Right now, if you do not provide a slug in the input field, it will let you continue causing a null reference exception. Here is the image of the error.

when can solve this by validating the slug input field before a new post is created.

Hi,
Thx for your work ! it's very cool

How can i create complexe url?
Like "/cat1/mytitle" ?
thx !

Seem to be able to reliably re-create this.
Believe behaviour only shows for Un-Published Posts

1. Login
2. Create new test Post (or change title of existing)
3. Un-tick Is Published
4. Save
5. Signout

Browser is redirected to http://miniblogcore20171027024748.azurewebsites.net/blog/Test/

Status code page, status code: 404

As its not visible to logged out users, as its un-published.

Looks like a Return-Url issue maybe?

(ps: Nice work!)

Hi
Do you think you're going to add a feature so that the articles have a main image?
Thank you

sometimes the file names or suffixes have invalid path names on blog hosted by operation system; also the relative pathname is used to create the absolute with hardcoded "/" - works only on windows

Sorry. I am a bit of a newbie here.

I have an MVC .net site running on IIS. I would like to have this run under a directory of that site but I cannot figure out how to install it?

I have set up the blog to run under it's own domain and IIS site (IIS 10) without issue. I would like to however run the site from sub-directory from the other domain.

To that end I have added an application to my other IIS site, named it "blog" and pointed it to the root of the working site. This keeps resulting in a 500 error, but I cannot seem to find what the error says despite enabling detailed errors in the web.config.

Any ides how I may get the blog to run in a sub-directory?

Thanks in advance.

I will continue to try and find the error message.

Looking through and can't see any reference to Syntax Highlighting support.

Is this provided / available or a "todo"?

I'd like to move an old Wordpress Blog to MiniBlog.Core. Can I use this tool? Miniblog Formatter

I need to look into this more, so just posting as a to-do.
But today, while editing a blog post and adding images on about 5 occasions (did resize most, but not all) after hitting save I got a 404 and seems site crashed in Azure and automatically restarted...

How can I change default demo password?
According to "appsettings.json" in comment, // Generate a new password hash with salt here https://onlinehasher.azurewebsites.net.
but, url seem does not work anymore.
Can anyone help?

Hello,

Maybe I'm doing something wrong but I'm getting this error when I'm trying to run solution.

I see that you registered it here, but I cannot find the file:

    // Register service worker
    if ('serviceWorker' in navigator) {
        navigator.serviceWorker.register("/serviceworker.js");
    }

Could you guide me where I have to look?

Might be a little easier for some people to use if there was a basic admin page that could modify these settings.

edit, looking a bit closer... it's using file storage for the entire blog. Not that it is a bad thing, just not what I was expecting. Probably not worth doing this unless the posts and comments changed too.

Looking through and can't see any reference to Markdown support.

Is this provided / available or a "todo"?

It would be much appreciated if the latest update to core 2.1 could be made available through the nuget package manager :-)

Even in the smallest of Blogs people tend to want to be able to schedule posts. The field ispublished could be replaced with a published field mapping to a DateTime?.

Question, is

public abstract class InMemoryBlogServiceBase : IBlogService
located in IBlogService.cs

is only a not used artefact, or is it used by the caching system?

I've been working on a PR (versus the stuff I whipped up real quick as a proof of concept) to integrate a couple other blog services that are not file based.

The first one being Azure table storage. There's a method on the service interface that is coupled to a file provider.

Task<string> SaveFile(byte[] bytes, string fileName, string suffix = null)
https://github.com/madskristensen/Miniblog.Core/blob/master/src/Services/IBlogService.cs#L107

While I could or would normally just ignore it and have it throw an exception, the issue here is how coupled this method is to the whole app not just the service itself. It's called once from each the blog controller and the metaweblog service.

Any thoughts on this specific coupling and how it could be reduced to just the FileBlog Service?

Here's the commit for the service I am proposing. But I could see this being a similar issue with a database provider.

tbasallo@fc8ccff

If there's no interest in supporting other services, then no big deal. But since an interface is used I imagine that the intent was to allow other providers.

A refactor of MetaWeblogService appears to have broken the WLS login:

        private void ValidateUser(string username, string password)
        {
            if (_userServices.ValidateUser(username, password))
            {
                throw new MetaWeblogException("Unauthorized");
            }

            var identity = new ClaimsIdentity(CookieAuthenticationDefaults.AuthenticationScheme);
            identity.AddClaim(new Claim(ClaimTypes.Name, username));

            _context.HttpContext.User = new ClaimsPrincipal(identity);
        }

As you can probably see, a successful user validation results in an exception, whereas a failure provides access...

When choosing New post when editing a post the form just loops around. The button should either be disabled in edit mode or open an empty form for a new post.

e.g.

Trying to add a miniblog.core blog to OLW returns the following error:

open livewriter connect is not able to connect, because we have no blog:name value in aspsettings anymore

1. When trying to add a Comment to a Post, clicking the Post Comment button results in a 404 error.

2. This behavior is also present on the Azure demo site. Screenshot attached.

3. You are a rock star of the coding community. Thank you for your contributions.

I've been trying to work out what is special about the miniblog.core routing/setup that stops it working as a IIS application (works fine when published to a root IIS site. Miniblog is unable to find Views and css files for example, but I'm unable to understand why ¯_(ツ)_/¯ Can anyone help/give me some pointers? Thanks.

Add a feature to show comment when that is approved.

Very Thanks

• RSS feed
• MetaWeblog API
• Windows/Open Live Writer support
• Commenting system
• Gravatar support
• OpenGraph support
• Post editing
• Post creation
• Post deletion
• Categories
• Robots.txt and sitemap.xml support
• Social sharing
• File/image uploads

I can see security headers added on live site, but can't find middleware reponsible for it?

I am looking for a blog engine component to serve as a back end to the Aurelia front end. It is being used as a model for several articles on distributed processes and identity management, that are planned for the Carmel project, dedicated to Aurelia Community. The article Aurelia blog describes its current functionality and plans for the near future.

I know how busy Mads is and I also know how "wide spread" he likes to be, meaning to say that he has wide spectrum of interests. So, it is possible that he may be interested in the idea to use miniblog.core as the engine for multiple front ends.

@madskristensen sorry this is not to say that something wrong with this project, but is this project setup with some CI or github hook to azure app?

I have been having this issue for quite a while with one of my project as well that one day I managed to deploy site once on azure web app and subsequent deployment ended up in same error and its very hard to debug. I have lost the confidence on auto deployment recently and been doing manual web deploy.

I would say web app deployment isn't as stable as it should be as this issue or similar has already been escalated on other places and most of the time the solutions presented don't work consistently.

here, here , here, here, here, here, here, here, here

Do you have any knowledge on it being down?

The rendering of partial views does not work on linux. The source of the problem is rather trivial: ASP.Net Core seems to be case sensitive on linux, and the code is not fully aware of this yet.

According to gravatar.com: Creating the Hash we had to

1. Trim leading and trailing whitespace from an email address
2. Force all characters to lower-case
3. md5 hash the final string

The given php example is

$email = trim( "[email protected] " ); // "[email protected]"
$email = strtolower( $email ); // "[email protected]"
echo md5( $email );
// "0bc83cb571cd1c50ba6f3e8a78ef1346"

The default encoding in php is UTF8

All in all, we had to change the code to

public string GetGravatar()
{
    using (var md5 = System.Security.Cryptography.MD5.Create())
    {
        var email = Email;
        // 1. Trim leading and trailing whitespace from an email address
        email = email.Trim();
        // 2. Force all characters to lower-case
        email = email.ToLowerInvariant();
        // 3. md5 hash the final string
        byte[] inputBytes = Encoding.UTF8.GetBytes(email);
        byte[] hashBytes = md5.ComputeHash(inputBytes);

        // Convert the byte array to hexadecimal string
        var sb = new StringBuilder();
        for (int i = 0; i < hashBytes.Length; i++)
        {
            sb.Append(hashBytes[i].ToString("X2"));
        }

        var gravatarId = sb.ToString().ToLowerInvariant();

        return $"https://www.gravatar.com/avatar/{gravatarId}?s=60&d=blank";
    }
}

As the SaveFilesToDisk method is called after the post has been persisted the path rewrite in this method can have no effect. As such the base64 image is still loaded in the blog code source.

This appears to simply require swapping the order of the lines unless I am missing something.

The category page selector on bottom of post, is not correct. It uses the page number from post page selector.

for example go to: https://madskristensen.net/28/
and select there the category link, build to: https://madskristensen.net/blog/category/random%20thoughts/28

and we get an empty category page, because there are at the moment of writing only 26 ones.
I will digging and build a PR

Hi there,
Sorry I'm an almost complete nub when it comes to developing but I'm trying to use this excellent project to learn some useful stuff
One thing I'm trying to figure out is what are the files I need to backup to restore my blog
Is it just the "Posts" folder ?
Thanks !

JS Output from serviceworker is not minified, imo because the call is without .js, so it will not detected by weboptimizer.

Are you interested in implementing other providers (in the core project) and taking pull requests, or would you prefer for users to implement as they need on their own and leave outside of this project?

I ask because with the interface created, it would be trivial for others to add other providers (SQL [using things like NPoco, dapper, ADO.NET, Azure Blob/Table Storage, etc.).