FPC 3.2.4-rc1 available

[Thaddy]

Marco has experience as release manager, just you forgot...

[marcov]

If want opinion, then I make also: If some one is think I make spend more time motivate why my work is need for eligible after > 4 year then better have think again. That is why exist commit message -> so take time and read what is motivate.

How will you track what is already merged to fixes or not for the last 1.5 years that still need to be done for 3.2.4? Anyway, it seems that at least Florian is merging via the -x mechanism. There is no history for main->fixes in the gitlab environment. The proposal (no matter what you think of it) is to build such thing based on MRs for the new major branch, but that is not even definitive, and doing it for fixes is burning bridges. (including 3.2.6 if the major branch delays more than expected)

I understand that if you are on trunk, you probably don't care, but many people will use 3.2.4.

[Martin_fr]

How will you track what is already merged to fixes

There are indeed some commits without the "cherry picked from". Seems so that in the latest commits, all parties have that set.

That said, I don't know how that works with older merges, that may be done with tooling in mind that uses other means, and if it allows to easily find a complete set.

---
**IF** you don't have conflicts, then you can usually use "git cherry" => it will find commits that don't have a matching commit (by changeset).
But when you have conflicts, and had to edit the merges...

Its not the same as svn, that had a "database" like list of actual merged revisions (which allowed adding numbers if automation failed).

[Thausand]

I not forget Thaddy 

It never a problem. Now I read ML and issues and all talk other people make.

Because I know marco is RM for past then I not understand why is question his experience.

How will you track what is already merged to fixes or not for the last 1.5 years that still need to be done for 3.2.4?

Because they on you mergelist

Anyway, it seems that at least Florian is merging via the -x mechanism. There is no history for main->fixes in the gitlab environment. The proposal (no matter what you think of it) is to build such thing based on MRs for the new major branch, but that is not even definitive, and doing it for fixes is burning bridges. (including 3.2.6 if the major branch delays more than expected)

I make understand that not all change is worse. But lack understanding and motivation is imho no good.

I not have any say for Florian but I read proposal from other and they have make no sense in positive way for me. Only motivate is look like 'ship it" any way and push  burden for person who is make commit. This ok when make change intrusive and have take responsibility but not for not do any thing 4 year and now tell person who commit for have motivate why commit is might need eligbile for new release. It > 4 year and I not know what I have eat when breakfast 4 year ago, you have ? 

I understand that if you are on trunk, you probably don't care, but many people will use 3.2.4.

I have personal no care because yes I make use trunk. But trunk is contain fixes that is need for make target for new compiler. End user is not know how apply fixes then rebuild compiler. So every time when fix not merged is mean wait next release and have cross finger it is merge. It is make 2 cycle release for have final fix for end user. More easy is have release for self and have forget all drama.

[wpostma]

This project is widely considered "dead" by most people on the internet because of no actual release for five and a half years.

The nightly builds are so well hidden (there's NO link to them from the download page), and everything on the wiki is 30 years old.

But yeah, the problem is a simple matter.    It's not simple. 

[marcov]

This project is widely considered "dead" by most people on the internet

Well, then we are Delphi compatible in that aspect :-)

[Thausand]

The nightly builds are so well hidden (there's NO link to them from the download page), and everything on the wiki is 30 years old.

https://www.freepascal.org/develop.html

[Thaddy]

This project is widely considered "dead" by most people on the internet because of no actual release for five and a half years.

It is still within four years until almost four  weeks: 20'th of May 2021.
And you forget about activity.
Debian contains code that is over 20 tears old for some processor families.. Why would they do that? It is only considered "dead" for people that do not know. You may consider yourself what kind of animal I should address you with.

Working code is never dead. It works...

To help you a little:

Code: Pascal  [Select][+][-]

1. // full program
2. {$mode objfpc}{$apptype console}
3. uses dateutils, sysutils;
4.   function Age(Releasedate:string):integer;inline;
5.   begin
6.     Result := YearsBetween(Now,StrToDateTime(Releasedate));
7.   end;
8. begin
9.   // the Age routine takes the defaultformatsettings.
10.   writeln(Age('20-05-2021'));
11.   readln;  
12. end.

..... You are way off, this says 4 at the moment, at least 7 months.
By then your toddler walks..

[Fred vS]

Read all the mails of Michael and Graeme c.s., and see what alternate procedures they describe except for the "flag for merge at the moment" and the "release manager that dutifully and automatically presses the button when prompted". They don't even commit to maintaining an eligible revision list, making falling back to the current process in the case of problems, or simply to mop up revisions that were wrong.

Nevertheless, they are proposing something concrete.
https://gitlab.com/freepascal.org/fpc/source/-/work_items/41735
And that gets things moving, unlike the empty talk and excuses we're used to hearing.

[Thaddy]

Empty? Scrap that, it is impatience by others. Outside pressure seems to help in this case, but is frankly a gotspe.
I like it the old ways: it is done when it is done, instead of our commercial cousin that calls a bugfix or two a major release...On all their three platforms they support..)

Perspective Fred, this is all about psychology, nothing else.
A release adds nothing to a user experience if it is not done properly.

I am ashamed about the community pushing for something that is in effect just a name, nothing more, nothing less. I would simply call RC1 a release to shut them up.

And then the lot starts compaining it is rubbish and again the major platforms are advantaged at the cost of others... Yeah, right. What is the prescription for your glasses?  (Certainly if you push for a niche platform, plz help out)

And I am now playing "I hate to say I told you so" by the Hives again.
There is a gross limitation on respect by many of you.
Or utter and complete lack of brains.           

Go ahead and use the unleashed branch: good luck with the minor platforms...                                                                                           

[Fred vS]

Yeah, right. What is the prescription for your glasses? 

My reading glasses?
No prescription, I tried them on right there in a Chinese shop; They're 2.5x magnification and cost me 3 euros.
And I can perfectly read all your unnecessary ad hominem attacks.

There is a gross limitation on respect by many of you.

Where is the lack of respect?
Respect must be mutual, towards users as well.

[Thaddy]

Respect must be mutual, towards users as well.

Respect to us users has over the years been shown by many of the core team, as well as patience in handling silly requests for an early release. Just in case you missed that. Not very high on my Einstein ranking. (On which I score also average otherwise)

[valdir.marcos]

Respect must be mutual, towards users as well.

Respect to us users has over the years been shown by many of the core team, as well as patience in handling silly requests for an early release. Just in case you missed that. Not very high on my Einstein ranking. (On which I score also average otherwise)

Status of FPC 3.4.0 or FPC 4.0.0 [major release]
https://forum.lazarus.freepascal.org/index.php/topic,73355.0.html

FPC 3.2.0 [major release] was released on June 20, 2020. That's 5 years and 10 months ago.

FPC 3.2.2 [minor release] was released on May 21, 2021. That's 4 years and 11 months ago.


Are you serious, Thaddy? "Silly requests for an early release"? 

Have you been living under a tree?

You are barking up the wrong tree, again!

[MarkMLl]

There is a gross limitation on respect by many of you.

Where is the lack of respect?
Respect must be mutual, towards users as well.

While I disagree with those who argue that because there has not been a new release for several years it will impact the credibility of Pascal, it does have to be said that it is fairly common for somebody to say something like "Use generics for that... oops, you'll have to use trunk". It's also fairly common for somebody to say "Yes, there's problems with the documentation and in particular the indexing but they're fixed in trunk".

And the person who is most likely to do that is Thaddy.

MarkMLl

[Thaddy]

Well, if I see something I could do I would certainly help. Even making coffee or pancakes. I have time on my hands.
My point is I find people moaning where that is not due. Alas that is also quite common. But it is an opinion.