New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Request for stable release (=version 1.0) #709

Open
304NotModified opened this Issue Sep 17, 2018 · 4 comments

Comments

Projects
None yet
2 participants
@304NotModified

304NotModified commented Sep 17, 2018

Hi @FlorianRappl

First of all, thanks for this great library! I think this library is the best in it's category :)

Unfortunately, it's a bit pity there is no stable (1.0) version released of this library. For me it's a bit frustrating as I expected a stable a few years ago. While there is some development now after years, I'm afraid there won't be a stable release ever.

I understand if you're busy with other things in live and sometimes things won't go as planned. But thinking out-of-the-box, I think the open issues for 1.0 is to huge.

Also the breaking changes are a bit disturbing, as libraries build on this library has a hard time to get stable. (for example the excellent library HtmlSanitizer - see issues)

So I have a proposal:

  • Skip all issues from v0.10 that aren't bugs. Those aren't critical. Sure, better performance is better, build on Linux is cool, but it's isn't really necessary for a 1.0
  • Skip v0.11. I think that would be nice for 1.1. Or maybe 2.0
  • Release 1.0 alphas/betas soon. An alpha could be released already with the 4 open bugs in v0.10
  • Last but not least (not sure if it was planned already), please adopt semver (= no breaking changes in 1.x)

Cheers,
Julian

PS: I apologize if I sound blunt. Direct language is very common in my home country 👼

@FlorianRappl

This comment has been minimized.

Show comment
Hide comment
@FlorianRappl

FlorianRappl Sep 17, 2018

Contributor

The reason for 0.10 is to get feedback for the API and have a stable 1.0. And yes, of course semver is used for 1.0 onwards (I think this is even described in one of the docs).

However, most bullet points still assume that some development is done. The truth is as simple as bitter: The .NET ecosystem and its community did not keep my motivation for development in this space. I still want to conclude the work on this one (i.e., at least leave a 1.0 version behind where only some hotfixes / improvements should be merged in from time-to-time), but time is an issue.

Nevertheless, even though I cannot promise anything I still have AngleSharp on the radar and I'll try to deliver hopefully more soon-ish than late / never-ish. Contributions in any space are welcome. One of the greatest killers was Microsoft's ever growing change. Once we had the nice multi-platform build using dotnet with project.json established the system was outdated. Then NuGet went on to abandon multi-platform support and it became hard to make the system build again on AppVeyor. Hence I believe that a 0.10 will drop most of the multi-platform to only go with .NET Standard (that is of course implied multi-platform, however, not containing all earlier supported platforms).

Regarding 0.11 - I inserted this one for some modifications prior to 1.0 that still should validated before going semver and having to release 2.0 instantly.

TL;DR: The proposal is of course nice and goes in line with my thoughts, however, zero progress is still zero progress no matter how much the scope is reduced.

HTH and thanks for your support! 🍻

Contributor

FlorianRappl commented Sep 17, 2018

The reason for 0.10 is to get feedback for the API and have a stable 1.0. And yes, of course semver is used for 1.0 onwards (I think this is even described in one of the docs).

However, most bullet points still assume that some development is done. The truth is as simple as bitter: The .NET ecosystem and its community did not keep my motivation for development in this space. I still want to conclude the work on this one (i.e., at least leave a 1.0 version behind where only some hotfixes / improvements should be merged in from time-to-time), but time is an issue.

Nevertheless, even though I cannot promise anything I still have AngleSharp on the radar and I'll try to deliver hopefully more soon-ish than late / never-ish. Contributions in any space are welcome. One of the greatest killers was Microsoft's ever growing change. Once we had the nice multi-platform build using dotnet with project.json established the system was outdated. Then NuGet went on to abandon multi-platform support and it became hard to make the system build again on AppVeyor. Hence I believe that a 0.10 will drop most of the multi-platform to only go with .NET Standard (that is of course implied multi-platform, however, not containing all earlier supported platforms).

Regarding 0.11 - I inserted this one for some modifications prior to 1.0 that still should validated before going semver and having to release 2.0 instantly.

TL;DR: The proposal is of course nice and goes in line with my thoughts, however, zero progress is still zero progress no matter how much the scope is reduced.

HTH and thanks for your support! 🍻

@304NotModified

This comment has been minimized.

Show comment
Hide comment
@304NotModified

304NotModified Sep 17, 2018

Thanks for the fast response! 👍

I fully understand your situation and therefore I try to get a stable, with the minimal effort needed.

One of the greatest killers was Microsoft's ever growing change.

Luckily I think it's stable now :) Maybe not fully mature, but stable.

Regarding 0.11 - I inserted this one for some modifications prior to 1.0 that still should validated before going semver and having to release 2.0 instantly.

Do you think "HTML 5.2" (#618) would introduce breaking changes? (PS #535 could be fixed easily without breaking changes)

I don't wanna repeat myself, but there are barely issues in the milestones 0.10, 0.11 and 1.0 that should block a stable release.

And yes, of course semver is used for 1.0 onward

👍 great news! As a developer, I really like semver

304NotModified commented Sep 17, 2018

Thanks for the fast response! 👍

I fully understand your situation and therefore I try to get a stable, with the minimal effort needed.

One of the greatest killers was Microsoft's ever growing change.

Luckily I think it's stable now :) Maybe not fully mature, but stable.

Regarding 0.11 - I inserted this one for some modifications prior to 1.0 that still should validated before going semver and having to release 2.0 instantly.

Do you think "HTML 5.2" (#618) would introduce breaking changes? (PS #535 could be fixed easily without breaking changes)

I don't wanna repeat myself, but there are barely issues in the milestones 0.10, 0.11 and 1.0 that should block a stable release.

And yes, of course semver is used for 1.0 onward

👍 great news! As a developer, I really like semver

@FlorianRappl

This comment has been minimized.

Show comment
Hide comment
@FlorianRappl

FlorianRappl Sep 17, 2018

Contributor

Do you think "HTML 5.2" (#618) would introduce breaking changes?

No the 0.11 is more a "guess" than a fixed set. I just placed also some other things (e.g., #618) in this milestone, but nothing cut in stone and such issues could easily be moved if there is no other reason for a 0.11.

Contributor

FlorianRappl commented Sep 17, 2018

Do you think "HTML 5.2" (#618) would introduce breaking changes?

No the 0.11 is more a "guess" than a fixed set. I just placed also some other things (e.g., #618) in this milestone, but nothing cut in stone and such issues could easily be moved if there is no other reason for a 0.11.

@304NotModified

This comment has been minimized.

Show comment
Hide comment
@304NotModified

304NotModified Sep 17, 2018

OK, but why before 1.0? 1.1 sounds reasonable to me.

304NotModified commented Sep 17, 2018

OK, but why before 1.0? 1.1 sounds reasonable to me.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment