Sign in Join

Live chat for Free Pascal, Lazarus, Delphi and every Pascal programmer.

Mailing lists / fpc-devel / 2016-04

FPC 3.0.2 release target

39 messages · 13 people · –

Reply by email Subscribe to this list Your subscription

Read-only archive. Use j and k to move between messages.

Ondrej Pokorny

What are the plans on releasing 3.0.2?
We'd like to release Lazarus 1.6.2 - it would be good if it was released with FPC 3.0.2 because of the Currency bug.

Ondrej

Michael Van Canneyt
to Ondrej Pokorny
Ondrej Pokorny wrote
What are the plans on releasing 3.0.2? We'd like to release Lazarus 1.6.2 - it would be good if it was released with FPC 3.0.2 because of the Currency bug.

I also think so, we've already been merging the necessary fixes to fixes_3_x

Now we need someone to volunteer to be buildmaster. I will ask the previous buildmaster if he is game... :-)

Michael.

Marco van de Voort
to Michael Van Canneyt

In our previous episode, Michael Van Canneyt said:

I also think so, we've already been merging the necessary fixes to fixes_3_x

For people that missed this hint: PLEASE retest your issues with the fixes branch (3.0.1), specially if your problems were mostly library in nature.

Luiz Americo Pereira Camara
to Marco van de Voort

2016-04-12 5:03 GMT-03:00 Marco van de Voort :

Show quoted text 7 lines
In our previous episode, Michael Van Canneyt said: > > I also think so, we've already been merging the necessary fixes to > fixes_3_x For people that missed this hint: PLEASE retest your issues with the fixes

Fixes branch is missing packages/googleapi

When asked previously there was no technical reason to not ship it

Luiz

Kazantsev Alexey
to Dejan Boras
Dejan Boras wrote
Would it be possible to include bug #29321 fix in 3.0.2? as this solves my problems with array helpers.

And #27349, please.

Kazantsev Alexey
Juha Manninen
to Ondrej Pokorny

Is there any estimate for FPC 3.0.2 release?
FPC 3.0.0 was released already at November 26, over half a year ago. Fixes branches of both FPC and Lazarus have lots of important commits. Lazarus 1.6.2 + FPC 3.0.2 would make a nice and solid combination.

Juha

Michael Van Canneyt
to Juha Manninen
Juha Manninen wrote
Is there any estimate for FPC 3.0.2 release? FPC 3.0.0 was released already at November 26, over half a year ago. Fixes branches of both FPC and Lazarus have lots of important commits. Lazarus 1.6.2 + FPC 3.0.2 would make a nice and solid combination.

We're set to create the branch at the end of May, which is about now. The last merges have been performed. I assume Marco will tag somewhere this week, and then building starts.

Michael.

Michael Van Canneyt
to Ondrej Pokorny
Ondrej Pokorny wrote 10 lines
On 29.05.2016 10:33, Michael Van Canneyt wrote: > We're set to create the branch at the end of May, which is about now. > The last merges have been performed. I assume Marco will tag somewhere this > week, and then building starts. Great! Do you have a merge list on the net somewhere (like Lazarus has: http://wiki.freepascal.org/Lazarus_1.6_fixes_branch ) or do I have to check the SVN merge info manually to see what has been merged?

Marco maintains a merge list, but I don't know the URL offhand.

All in all, the best (and authoritative) source is svn mergeinfo :-)

Michael.

Marco van de Voort
to Michael Van Canneyt: Marco maintains a merge list, but I don't know the URL offhand. All in all, the best (and…

In our previous episode, Michael Van Canneyt said:

Show quoted text 5 lines
> Do you have a merge list on the net somewhere (like Lazarus has: > http://wiki.freepascal.org/Lazarus_1.6_fixes_branch ) or do I have to check > the SVN merge info manually to see what has been merged? Marco maintains a merge list, but I don't know the URL offhand.

I maintain a list that splits up the eligible revisions into categories, not a merged list.

All in all, the best (and authoritative) source is svn mergeinfo :-) The last merges have been performed. I assume Marco will tag somewhere this week, and then building starts.

fpcbuild still needs to be checked for merges, the list is short though, at http://www.stack.nl/~marcov/mergelogsbuild30/resultset.html

Sven Barth
to Maciej Izak
"Maciej Izak" wrote 9 lines
2016-04-11 15:27 GMT+02:00 Ondrej Pokorny : > > What are the plans on releasing 3.0.2? Is possible to add r31911 to 3.0.2? http://bugs.freepascal.org/view.php?id=29045

It can't be merged as is, but I've provided Marco with a custom patch.

Please note that I had no problems compiling your examples with 3.0.0 (the problem would only manifest with generic records in that version).

Regards,
Sven

Maciej Izak
to Marco van de Voort

2016-05-30 11:29 GMT+02:00 Marco van de Voort :

Show quoted text 8 lines
In our previous episode, Sven Barth said: > > Is possible to add r31911 to 3.0.2? > > > > http://bugs.freepascal.org/view.php?id=29045 > > It can't be merged as is, but I've provided Marco with a custom patch. Done.

Sven & Marco - Thanks!

What about Arnaud Bouchez proposition? That is really important in production for most of us... Modifications in mentioned branch are simple (I think so) and *critical* for proper work.

Best regards, Maciej Izak
Dejan Boras
to Alfred

Is there a chance to merge the fix for #29321 as well in 3.0.2? Last time I tested 3.0.1 (few days ago), this wasn't included. Not really in an environment now where I can test it again.

Unknown author
to Dejan Boras

It says fixed in 3.1.1. It doesn't say it would be back-ported, but indeed you can request it.
And indeed in this case I support your request. (Since it was an oversight in the first place?)
I have no clue if this is easy to do, but I think it is warranted.

Regards,

Thaddy

Dejan Boras wrote 9 lines
Is there a chance to merge the fix for #29321 as well in 3.0.2? Last time I tested 3.0.1 (few days ago), this wasn't included. Not really in an environment now where I can test it again. _______________________________________________ fpc-devel maillist - http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
Sven Barth
to Maciej Izak: 2016-05-30 11:29 GMT+02:00 Marco van de Voort : Sven & Marco - Thanks! What about Arnaud…
"Maciej Izak" wrote 16 lines
2016-05-30 11:29 GMT+02:00 Marco van de Voort : > > In our previous episode, Sven Barth said: > > > Is possible to add r31911 to 3.0.2? > > > > > > http://bugs.freepascal.org/view.php?id=29045 > > > > It can't be merged as is, but I've provided Marco with a custom patch. > > Done. Sven & Marco - Thanks! What about Arnaud Bouchez proposition? That is really important in

production for most of us... Modifications in mentioned branch are simple (I think so) and *critical* for proper work.

Let me first finish the packages stuff as that will lead to a breaking change in the RTTI binary format (PPTypeInfo instead of PTypeInfo, necessary due to Windows' PE format) and I don't want to add any new RTTI stuff before this is dealt with.

Regards,
Sven

Maciej Izak
to Sven Barth: production for most of us... Modifications in mentioned branch are simple (I think so)…

2016-05-30 14:41 GMT+02:00 Sven Barth :

Show quoted text 5 lines
Let me first finish the packages stuff as that will lead to a breaking change in the RTTI binary format (PPTypeInfo instead of PTypeInfo, necessary due to Windows' PE format) and I don't want to add any new RTTI stuff before this is dealt with.

I do not want to grumble, but... Release date for packages? It is really painful. Maybe you need some support? My work with smart pointers is almost done and after smart pointers release I can put effort into packages. The work of other developers outside FPC core team (also my work) is blocked for more than half a year... or even more.

At this moment I am blocked with many parts of my work (for example few of kinds of smart pointers can't work without fixing #28911 ... even I can't fix that myself because at the moment you don't want patch for this like for #29245 etc. -,-).

IMO merging changes into trunk for mentioned branch shouldn't have affect on packages progress.

Best regards, Maciej Izak
Michael Van Canneyt
to Maciej Izak
Maciej Izak wrote 12 lines
2016-05-30 14:41 GMT+02:00 Sven Barth : > Let me first finish the packages stuff as that will lead to a breaking > change in the RTTI binary format (PPTypeInfo instead of PTypeInfo, > necessary due to Windows' PE format) and I don't want to add any new RTTI > stuff before this is dealt with. > I do not want to grumble, but... Release date for packages? It is really painful. Maybe you need some support? My work with smart pointers is almost done and after smart pointers release I can put effort into packages. The work of other developers outside FPC core team (also my work) is blocked for more than half a year... or even more.

This is not an argument.

I will repeat: FPC is and remains a hobby project for the core team.

Expecting some kind of timeline is simply deceiving yourself.

Michael.

Maciej Izak
to Michael Van Canneyt

2016-05-30 15:40 GMT+02:00 Michael Van Canneyt :

This is not an argument.

Yes this is not an argument but reality. It blocks a lot of code.

I will repeat: FPC is and remains a hobby project for the core team.

Nothing new: most of 3rd contributions for FPC/Lazarus stuff is produced as hobby and in free time.

Expecting some kind of timeline is simply deceiving yourself.

No. That is kind of motivation and discipline.

no offence :)

Best regards, Maciej Izak
Michael Van Canneyt
to Maciej Izak
Maciej Izak wrote 6 lines
2016-05-30 15:40 GMT+02:00 Michael Van Canneyt : > This is not an argument. > Yes this is not an argument but reality. It blocks a lot of code.

Certainly.
I suffer from it myself from time to time, but such is life...

Show quoted text 6 lines
> Expecting some kind of timeline is simply deceiving yourself. No. That is kind of motivation and discipline. no offence :)

None taken, simply because I don't understand what you mean with your last sentence... :)

Michael.

Ondrej Pokorny
to Ondrej Pokorny: Could you please include http://mantis.freepascal.org/view.php?id=28578 (r31434) into…
Ondrej Pokorny wrote 8 lines
On 12.04.2016 9:22, Michael Van Canneyt wrote: > I also think so, we've already been merging the necessary fixes to > fixes_3_x Could you please include http://mantis.freepascal.org/view.php?id=28578 (r31434) into 3.0.2 as well?

Could you please include http://mantis.freepascal.org/view.php?id=28578 (r31434) into 3.0.2 as well? (Or confirm that it won't be merged.)

It's not a biggy, it just makes debugging on Windows more comfortable.

Ondrej

Dejan Boras
to Unknown author: It says fixed in 3.1.1. It doesn't say it would be back-ported, but indeed you can…

It's not a big change for #29321, from what I see in the svn diff for r32845. If not, I can always patch fpc myself.

Show quoted text 28 lines
It says fixed in 3.1.1. It doesn't say it would be back-ported, but indeed you can request it. And indeed in this case I support your request. (Since it was an oversight in the first place?) I have no clue if this is easy to do, but I think it is warranted. Regards, Thaddy On 30-May-16 12:42 PM, Dejan Boras wrote: Is there a chance to merge the fix for #29321 as well in 3.0.2? Last time I tested 3.0.1 (few days ago), this wasn't included. Not really in an environment now where I can test it again. _______________________________________________ fpc-devel maillist - _______________________________________________ fpc-devel maillist - http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
Dimitrios Chr. Ioannidis
to Sven Barth: get It's less the binary/RTTI format (which is what the branch provides), but more the…

Στις 2016-05-30 15:43, Sven Barth έγραψε:

Show quoted text 16 lines
Am 30.05.2016 14:07 schrieb "Michael Van Canneyt" : > > > > On Mon, 30 May 2016, wrote: > >> New feature. Belongs in 3.1.1. > > > And the implementation is contested, which is the reason why it doesn't get > merged to trunk. It's less the binary/RTTI format (which is what the branch provides), but more the Invoke() implementation that would follow...

I know that I'm probably late, but ...

Sven,

currently fpc is ahead of Delphi in Linux platform but this will probably change ( if this time, they keep their Road Map that is) .

Plz, think again regarding the proposition at
http://bugs.freepascal.org/view.php?id=26774#c91291 . Make it a switch or a directive during fpc build or what ever, just don't let time passes by.

regards,

Dimitrios Chr. Ioannidis
Luiz Americo Pereira Camara
to Dimitrios Chr. Ioannidis

2016-06-08 15:40 GMT-03:00 Dimitrios Chr. Ioannidis <

:
Show quoted text 12 lines
I know that I'm probably late, but ... Sven, currently fpc is ahead of Delphi in Linux platform but this will probably change ( if this time, they keep their Road Map that is) . Plz, think again regarding the proposition at http://bugs.freepascal.org/view.php?id=26774#c91291 . Make it a switch or a directive during fpc build or what ever, just don't let time passes by.

Just to clarify, the bug report and previous message of Alfred request to be included in trunk and not in 3.0.2

Luiz

Sven Barth
to Dimitrios Chr. Ioannidis: Στις 2016-05-30 15:43, Sven Barth έγραψε: I know that I'm probably late, but ...

Am 08.06.2016 20:41 schrieb "Dimitrios Chr. Ioannidis" <

Show quoted text 27 lines
: Στις 2016-05-30 15:43, Sven Barth έγραψε: > > Am 30.05.2016 14:07 schrieb "Michael Van Canneyt" > : > > > > > > > > On Mon, 30 May 2016, wrote: > > > >> New feature. Belongs in 3.1.1. > > > > > > And the implementation is contested, which is the reason why it > doesn't get > > merged to trunk. > > It's less the binary/RTTI format (which is what the branch provides), > but more the Invoke() implementation that would follow... I know that I'm probably late, but ... Sven, currently fpc is ahead of Delphi in Linux platform but this will

probably change ( if this time, they keep their Road Map that is) .

Plz, think again regarding the proposition at

http://bugs.freepascal.org/view.php?id=26774#c91291 . Make it a switch or a directive during fpc build or what ever, just don't let time passes by.

Since I've now merged the critical RTTI related changes from my packages branch I'll take a look at it, play with it a bit and maybe also implement a new target (m68k or Power for example to test big endian support). If there are no apparent problems I'll merge it if no other core dev objects.

@Steve: would you please change your usages of PTypeInfo to PPTypeInfo in typinfo (after you've catched up with trunk again)?

Regards,
Sven