Preparing FPC 3.2.4, point out road blocks now

[TRon]

Those are bad too I suppose:

Dynamic vs static. Personally I think it is bad but generally it is understandable.

[zeljko]

Aleluyah https://gitlab.com/freepascal.org/fpc/source/-/commits/release_3_2_4-branch?ref_type=heads

[BrunoK]

Please, if not done, include:

Code: Pascal  [Select][+][-]

1.  packages/fcl-db/src/base/sqlscript.pp | 1 +
2.  1 file changed, 1 insertion(+)
3.  
4. diff --git a/packages/fcl-db/src/base/sqlscript.pp b/packages/fcl-db/src/base/sqlscript.pp
5. index afc1944..79495e4 100644
6. --- a/packages/fcl-db/src/base/sqlscript.pp
7. +++ b/packages/fcl-db/src/base/sqlscript.pp
8. @@ -509,6 +509,7 @@ begin
9.    FSkipStackIndex:=0;
10.    Faborted:=False;
11.    FLine:=1;
12. +  FCol:=1;
13.    DefaultDirectives;
14.    Repeat
15.      NextStatement();

I've lost quite a few hours because of that.

[zeljko]

@BrunoK better would be to open issue at fpc bugtracker.

[BrunoK]

@BrunoK better would be to open issue at fpc bugtracker.

As far as I know, the bugtracker is for trunk. This bug is corrected in trunk.
Here we are considering 3.2.4

[zeljko]

Can you write revision id ?

[BrunoK]

Can you write revision id ?

Nothing I could find in the history. It has been corrected, I dont know when or eventualy my local Git copy of 3.2.2 is damaged.

[dsiders]

Can you write revision id ?

https://gitlab.com/freepascal.org/fpc/source/-/commit/672f90bf40cc6fe8d3dbb2be0402e41d7113fc00
Issue #39356.

Cherry-picked in https://gitlab.com/freepascal.org/fpc/source/-/commit/0376cb7ae8d10bcf3ad754e0f2539c462b5cfdd4 7 months ago.

[zeljko]

So it is there already, nothing to be done.

[BrunoK]

Thanks you for the references, couldn't find them. Couldn't figure out the search words.

[dsiders]

Thanks you for the references, couldn't find them. Couldn't figure out the search words.

I used git blame to find the commits for that line of code instead of searching the issue list.

[dbannon]

If anyone is interested in this, perhaps it belongs in a new thread ?

Apropos libfontconfig, I booted a *nix system for the first time after a busy month, and looked into this. I found that there already is an override in the code, at least in trunk, try the following:

1. put unit libfontconfig in your uses.
2. put

Code: Pascal  [Select][+][-]

1.  LoadFontConfigLib('libfontconfig.so.666',false);

as first line in your .lpr to try to load libfontconfig.so.666 before the rest is tried. Untested,  no guarantees, but worth a try.

Thanks Marcov.  I have been travelling but finally got around to building a FPC without my patch to test your idea. Not sure where 'libfontconfig.so.666' comes from, but applying your logic to 'libfontconfig.so.1' almost works !  But not quite.

While it is possible to preempt libfontconfig.pp's attempt to load "libfontconfig.so"  by calling LoadFontConfigLib('libfontconfig.so.1', false) earlier, its not useful because (eg) fptty.pp later calls LoadFontConfigLib('') and that triggers an EInOutException when it finds the library is already loaded. fptty.pp sees the Exception and refuses to proceed.

That Exception appears to be inappropriate in my view.

I have tested proceeding without an Exception if the library is already loaded and it worked, on my system, as it should.

That is, however, a far more obtrusive change than the earlier suggested one.

So, setting the DefaultLibName, for Linux, to 'libfontconfig.so.1' is still my recommendation.       

Davo

PS : I will add my notes to the bug report if no one disputes them, save someone else from starting from scratch.

[ALLIGATOR]

I see in the 2020 announcement that this feature is not planned to be included in the 3.2 branch, but maybe something has changed in 4 years and there is a chance to include it in 3.2.4  ? Via modeswitch

[marcov]

No major language features in 3.2.x. That is for the next major version. (3.4 or 4.0)

[BSaidus]

Hello  .
So, what about the release of the 3.2.4 version of FPC.
We are waiting.
I do not want to ask but, I'll ask at my own risks.