|
zezba9000 wrote: supports newer C# versions
Disingenuous claim - it *forces* you to use newer C# versions
zezba9000 wrote: newer framework features,
Disingenuous - it's still just .Net
zezba9000 wrote: runs on more platforms
Mostly false - it only runs on Windows if you add a UI. On linux, you have to run it through a glorified parser(the dotnet command). When you get right down to it, .Net isn't "native" on ANY platform.
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
I work with dotnet 5 every day and I cannot say that any of the claims here are actually factual. I run core 3.1 ( which is NOT exactly the same as dotnet 5 ) on a Raspberry PI with full UI. (UWP).
I Now run dotnet 5 as a portable web site ( runs on everything I've tried so far, with no real complications ).
As for size of the executable folder. Try this..... Just create a demo app using full framework ( web forms ) then create the demo Blazor app, now do a comparison using windows explorer-Properties.
Be aware if you used the ".ALL" nuget package in your conversion ( Huge legacy, Full Framework library ) and then select the option that automatically loads the whole Kestral web server into the runtime folder when you deploy you will get a lot of things you really don't need.
Actually the whole paradigm has changed.
|
|
|
|
|
Member 7921483 wrote: web site
Web apps, by their very nature, areplatform agnostic. The UI is handled by the browser, NOT the "app" itself. Desktop apps written with Core that have a UI are NOT cross platform. My claim stands.
Member 7921483 wrote: on a Raspberry PI with full UI. (UWP)
Is that a web app, or a desktop app?
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
Your "feelings" don't effect the facts. Sorry.
I think .NET has been a mess for years with its runtime fragmentation and lack of AOT in any meaningful sense... but the reality is if you want to use new lang or framework features fundamentally based on CoreLib changes you need to update.
|
|
|
|
|
Wait till you implement a small microservice, and 6 months later you need another microservice so you fire up another ASP.NET Core 3.1 project and use NuGet to get the various packages you need, and suddenly, running on the same server, all these bizarre dependent DLL's can't be found because whatever you did to get them on your server for the first project, well, the second project, which now uses newer versions, those don't exist on the server.
And given this is a production server, I'm reluctant to update all those DLL's, as it might break something in the other existing projects.
So my solution? I copy the entire SLN of the microservice that is running, replace various C# files with the functionality I need in the new service, and it works.
F***ing mess, is what NuGet and Core 3.1 is.
.NET was supposed to "solve" DLL hell. It's come back with a vengeance.
Yes, I suppose I could edit some obscure config file somewhere that says "use this version of this package", or manually select the package version I used previously when installing the NuGet package, but really? Why? Why do I need to go through all this extra work?
Oh, and I also blame the open source community. So many updates, constantly, it doesn't take six months for something to change. In some cases, it takes mere hours.
|
|
|
|
|
It's a setting.. you can deploy your app as Framework Dependent (need to have it installed on the target machine, deploy size is same as before) or Framework Independent (need not install the framework on the target machine, the deploy size is much bigger)
While you would think "just install the damn framework", if you ever had to handle with your IT support that refuse to install new version for whatever reason, the larger in size, yet framework independent version is a god send! 
|
|
|
|
|
On what systems are your deploying where you worry about 10Mb?
That's maybe half the size of a single RAW photo taken on a new iPhone.
cheers
Chris Maunder
|
|
|
|
|
I'm not "worried". I'm old, and I started working in the industry back in the late 70's, when we only had single sided 5-1/4 inch floppies and 64k of RAM. Conservation of resources was a really big deal back then, and while I realize that resources are generally more abundant today, I still code as if they weren't.
I realize that the .Net framework requires almost 5gb of disk space to install, but I view that as part of the OS, and not part of my application when I deploy it. When my app grows to over 10mm for a simple "hello world" application, I get a bit annoyed.
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
I get it.
I was taking a trip down memory lane with some old 3 1/2 inch disks and was thinking 'how did we ever survive??' 640Kb RAM, 20Mb HDD, 4.77MHz. Ah, those were the days...
cheers
Chris Maunder
|
|
|
|
|
Trying to tighten up my CSS skills so I'm reading, CSS: The Definitive Guide[^] (4th ed) O'Reilly pub., and I learned something about CSS selectors I've never seen before.
If you have only a bit of CSS knowledge (like me) then you probably know you can create :
1) id selectors (indicated by # symbol)
#mainThang{color:blue;border: 5px solid red;}
<div id="mainThang>This my example.</div>2) class selectors (indicated by . symbol)
.articleTitle {font-size:xx-large;color:red}
<div class="articleTitle">This Explains Everything</div>
All Elements With An Attribute
But, did you know you could select all elements that have a particular attribute?
For example you can select every img element on your page that has an alt tag on it.
Obviously, if the img doesn't have that attribute it isn't chosen.
You use [ ] brackets to indicate the attribute of the element:
img[alt] {border: 3px solid red;}
<img alt="elephant pic" src="fakeUrl.png">
<img alt="bird pic" src="fakeUrl2.png">
<img src="fake_unknown.png">
The first two images would get the style applied, but the last one wouldn't.
I like learning. Learning's my favorite. 
|
|
|
|
|
Interesting. I avoid messing with CSS as much as possible, jQuery is my thing, but I suppose I should spend a little more time on learning about it.... maybe next year.
- I would love to change the world, but they won’t give me the source code.
|
|
|
|
|
� Forogar � wrote: I avoid messing with CSS as much as possible,
Me too. Boostrap [^] will definitely spoil you if you try it out.
It basically style controls for you in HTML the same we were able to depend upon Microsoft to do all that for us on WinForms in the past.
|
|
|
|
|
Yeah, we use Bootstrap as well. I like it because "it just works".
- I would love to change the world, but they won’t give me the source code.
|
|
|
|
|
Forogar wrote: I like it because "it just works".
Absolutely! It's actually quite easy to learn from the bootstrap examples too.
|
|
|
|
|
|
Richard Deeming wrote: You can also select elements based on full or partial matches of the attribute value
Yeah, there's a lot more granularity there than I had ever known. Good stuff really.
|
|
|
|
|
|
|
W3Schools is a good reference. I use it all the time. Thanks
|
|
|
|
|
Yes, but the only reason is I have and have read that book.
I'm not sure how many cookies it makes to be happy, but so far it's not 27.
JaxCoder.com
|
|
|
|
|
|
raddevus wrote: I like learning. Learning's my favorite.
Followed by ranting about what I learned, at least for me. 
|
|
|
|
|
wait until you find out about the pseudo-selectors like nth-child... 
Phil
|
|
|
|
|
switch (es) {
case CS_DISENABLED: return ::CreateSolidBrush(brRed); break;
case CS_ENABLED: return ::GetSysColorBrush(COLOR_WINDOW); break;
case CS_UNDEF: return ::CreateSolidBrush(brRed); break;
default: return ::CreateSolidBrush(brRed); break;
}
What's wrong here?
I'll wait.
Software Zen: delete this;
|
|
|
|
|
1) none of the break statements are ever hit
2) default does not require nor need break statement
3) DISENABLED is a very odd word.
those are all I got right now. 
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
