Jump to page: 1 2
Thread overview
Web Programming in D
Feb 08
aberba
Feb 08
Sergey
Feb 08
aberba
Feb 08
Martyn
Feb 08
Sergey
Feb 08
Martyn
Feb 08
Sergey
Feb 08
aberba
5 days ago
karitoy
5 days ago
Sergey
February 08

https://forum.dlang.org/post/cffdqyvthqvurbcpxvbo@forum.dlang.org

On Thursday, 22 June 2023 at 06:15:47 UTC, Web Dev wrote:

>

The D language is among the languages ‚Äč‚Äčthat I have been interested in for a long time, and it is pleasing to see that there is an active community by following the forum for a while.

As part of my job, I usually develop web software. There have been projects where I have used Python, PHP and recently Go for the backend.

However, I recently thought of using the D language in web projects as well. When I look at the frameworks developed on the web, I see Hunt and Vibe.D options.

I think Hunt is no longer in development but Vibe.D continues to be developed.

Has anyone done a web project with Vibe.D before? What are the positive and negative aspects? Does it make sense to use the D language on the web for now?

Personally I see nothing specifically wrong with vibe.d. I've talked to people using vibe.d professionally in the community too. The only major issue will be the few third-party packages. Everything else is doable.

Vibe.d uses diet templates which can take a while to compile but that's about it.

February 08

On Thursday, 8 February 2024 at 00:47:52 UTC, aberba wrote:

>

Personally I see nothing specifically wrong with vibe.d. I've talked to people using vibe.d professionally in the community too. The only major issue will be the few third-party packages. Everything else is doable.

Vibe.d uses diet templates which can take a while to compile but that's about it.

Interesting. I thought the large number of packages is the main advantage of the vibe :D
While disadvantages:

  • bad maintenance
  • a lot of code
  • hard to understand code => hard to fix bugs
  • slow compilation time
  • with diet it becomes even slower
February 08

On Thursday, 22 June 2023 at 06:15:47 UTC, Web Dev wrote:

>

As part of my job, I usually develop web software. There have been projects where I have used Python, PHP and recently Go for the backend.

It seems where you work you have the flexibility to use different tools for different projects. I am not sure why you would do this - but I am sure there are many.

I think the decision is likely a business one, rather than the language itself. D is certainly capable being used for Websites or REST APIs etc.

If you work on your own then I guess you are free to use whatever language you want without any pushback. For me, even as a lead, if I hinted at the next project being in a different language (not just D) I would have a number of puzzled (and unhappy) faces in my team.

  • "Why?" my C# team would scream

If I entertain the idea we all decided to use another language for legit reasons in our next project, especially for web development, the contenders are likely to be the popular choices like:-

  • PHP
  • Python
  • Ruby
  • Go

Blimey, even Rust could be a serious contender for the web space now.

Of course, what possible reason would we leave C# in favour another? For our circumstances.. very little.

Also, picking "popular" languages will help hiring new developers, or replace existing ones.

I have toyed with D using Vibe.d and I do like it. However, it is a lot of bloat and diet can slow things down.

While it could still actively maintained, it is unlikely to be as full featured to others in php, python, go, etc.

Personally, I would not give up other technologies for it in the workplace. The risks are too high, especially with other team members and the "pointing fingers" blame culture. Although we I did push for D and something went terribly wrong, pointing fingers at me is valid.

Your circumstances might be different, though.

February 08

On Thursday, 8 February 2024 at 12:25:49 UTC, Martyn wrote:

>

as full featured to others in php, python, go, etc.

Thank you for your reply.

I had an idea to create a kind of list of features of web-frameworks (to see what D ecosystem is lacking).
Current source of data is this page https://en.wikipedia.org/wiki/Comparison_of_server-side_web_frameworks - it has "classic" list of features.

Also I've checked documentation of some frameworks with features, that mentioned on their pages.
Some features that usually mentioned:

  • Authorization (including oauth)
  • Middleware
  • DB interaction (ORM)
  • Routing, handlers
  • Caching
  • Sessions

Would you be so kind to name some list of features (based on your real life experience) that are vital from your opinion and is not on the surface?

February 08

On Thursday, 8 February 2024 at 12:46:05 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 12:25:49 UTC, Martyn wrote:

>

as full featured to others in php, python, go, etc.

.. snip

Please dont take my comment out of context.... you missed vital bits before it:-

it is unlikely to be as full featured to others in php, python, go, etc.

Despite this... Yes, I will admit it is disingenuous of me to make this claim. D (or vibe.d particularly) are not popular choices in the web world. As it is not popular, it is not as actively used so while "full featured" is inaccurate it is still prone to bugs that would likely be addressed and fixed with other popular alternatives.

Everyones position is different. If you (or the original poster) are working solo or have a team with the same mindset to try/use D then thats a different story. You're all in it together! For me, I would be arguing to use D with my present (and past) team.

I would have to sell D, first
Even if I win them over with D's power and flexibility - The question would remain.. why should we use it over PHP, Python, Ruby, Go, Rust, Java, etc? Could we hire developers easily with D than Python? Its an obvious question.

I would have to sell DUB.
It is already difficult selling DUB from a bunch of C# developers using Nuget. Again.. why D+DUB when you have package managers available for other popular alternatives.

I would have to sell vibe.d
In all honesty, I think I could sell vibe.d well for simple demonstrations. It might take a bit of time to get them accustomed to using Diet.

Despite my efforts, I will likely lose the battle to use D. We will use a language that is popular, highly active, with web frameworks heavily tried and tested! When you think of a least popular choice, you naturally think "what could be missing? Does it needs to play catchup?"

My comment is not in any way pooping on D or libraries like vibe.d. I do appreciate the effort that has gone into the project. However, it is difficult to push for D+vibe.d because there is a risk compared to using other, more popular frameworks out there in python, php, go, etc?

Last note -- if i was with a team who were all positive to using the D language (and even things like Vibe.d)... I would be all in! So please dont think I am being negative. This is just my answer to questions like "Should I use D for my next web project" - it depends.

February 08

On Thursday, 8 February 2024 at 14:36:48 UTC, Martyn wrote:

>

On Thursday, 8 February 2024 at 12:46:05 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 12:25:49 UTC, Martyn wrote:

>

as full featured to others in php, python, go, etc.

.. snip

Please dont take my comment out of context.... you missed vital bits before it:-

it is unlikely to be as full featured to others in php, python, go, etc.

Don't get me wrong, I didn't have any context of getting vibe in production.
And your point from first message also was very clear and I definitely understood it.
There are many points why it is almost impossible to bring D into production, at least with current level of the development (which is not about compiler itself or even ecosystem, but about the risks and userbase).

But my question was more about features for web-frameworks not related to D.
Just features themself. And their usability in other frameworks.

February 08

On Thursday, 8 February 2024 at 10:21:29 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 00:47:52 UTC, aberba wrote:

>

Personally I see nothing specifically wrong with vibe.d. I've talked to people using vibe.d professionally in the community too. The only major issue will be the few third-party packages. Everything else is doable.

Vibe.d uses diet templates which can take a while to compile but that's about it.

Interesting. I thought the large number of packages is the main advantage of the vibe :D

Well vibe.d has the most packages in the D ecosystem for sure and what I would go for when using D. But it the context of all other web development languages/frameworks, vibe has very limited options. You'd need to write a lot more code yourself.

February 08

On Thursday, 8 February 2024 at 15:50:41 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 14:36:48 UTC, Martyn wrote:

>

On Thursday, 8 February 2024 at 12:46:05 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 12:25:49 UTC, Martyn wrote:

>

as full featured to others in php, python, go, etc.

.. snip

Please dont take my comment out of context.... you missed vital bits before it:-

it is unlikely to be as full featured to others in php, python, go, etc.

Don't get me wrong, I didn't have any context of getting vibe in production.
And your point from first message also was very clear and I definitely understood it.
There are many points why it is almost impossible to bring D into production, at least with current level of the development (which is not about compiler itself or even ecosystem, but about the risks and userbase).

But my question was more about features for web-frameworks not related to D.
Just features themself. And their usability in other frameworks.

Vibe.d is actually very packed with features. It however has got it's own way of doing session and auth using high level APIs which I'm not a fun of their design. With that said, I see nothing that can't be done in vibe.d.

For example, Node.js has express framework which is very popular for web. Arguably more popular than those traditional (python, C#, etc) solutions. It's a simple framework and very similar to vibe.d.

Vibe.d has auth (can do oauth), session, has orms available, crypto/hashing, websocket, etc.

The D community is mostly centered around system programming and many folks here who might not be developing full stack web professionally are too critical of vibe.d in ways that IMO are unrealistic. By that criteria, node.js (JavaScript) wouldn't even be qualify to be used for anything. Node.js (JavaScript) is the top language for web dev both serverside and clientside.

5 days ago

On Thursday, 8 February 2024 at 15:50:41 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 14:36:48 UTC, Martyn wrote:

>

On Thursday, 8 February 2024 at 12:46:05 UTC, Sergey wrote:

>

On Thursday, 8 February 2024 at 12:25:49 UTC, Martyn wrote:

>

as full featured to others in php, python, go, etc.

.. snip

Please dont take my comment out of context.... you missed vital bits before it:-

it is unlikely to be as full featured to others in php, python, go, etc.

Don't get me wrong, I didn't have any context of getting vibe in production.
And your point from first message also was very clear and I definitely understood it.
There are many points why it is almost impossible to bring D into production, at least with current level of the development (which is not about compiler itself or even ecosystem, but about the risks and userbase).

But my question was more about features for web-frameworks not related to D.
Just features themself. And their usability in other frameworks.
Hi Sergey, if you are just about to learn Vibe.d, I have written a tutorial on it here:
https://reyvaleza.gitbook.io/vibe.d-tutorial/
You can take a cursory look at that tutorial and see if Vibe.d has the features you are looking for.

5 days ago

On Thursday, 8 February 2024 at 00:47:52 UTC, aberba wrote:

>

https://forum.dlang.org/post/cffdqyvthqvurbcpxvbo@forum.dlang.org

On Thursday, 22 June 2023 at 06:15:47 UTC, Web Dev wrote:

>

[...]

Personally I see nothing specifically wrong with vibe.d. I've talked to people using vibe.d professionally in the community too. The only major issue will be the few third-party packages. Everything else is doable.

Vibe.d uses diet templates which can take a while to compile but that's about it.

emmankoko5199@gmail.com. can you ping me?

« First   ‹ Prev
1 2