An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171206/03619cda/attachment.html>

Mattias Gaertner wrote

The Lazarus team is glad to announce the release of Lazarus 1.8.

Congrats to the Lazarus team for this expected released.

I just downloaded the sources from sourceforge and after compile it, the splash window and the about dialog show version 1.9.0, I tried the next downloads.

https://sourceforge.net/projects/lazarus/files/Lazarus Zip _ GZip/Lazarus 1.8.0/lazarus-1.8.0.zip
<https://sourceforge.net/projects/lazarus/files/Lazarus%20Zip%20_%20GZip/Lazarus%201.8.0/lazarus-1.8.0.zip>

https://sourceforge.net/projects/lazarus/files/Lazarus Zip _ GZip/Lazarus 1.8.0/lazarus-1.8.0.tar.gz
<https://sourceforge.net/projects/lazarus/files/Lazarus%20Zip%20_%20GZip/Lazarus%201.8.0/lazarus-1.8.0.zip>

If I download the sources from svn tag, the version is displayed correctly.

Best Regards.

Victor Campillo via Lazarus hat am 7. Dezember 2017 um 10:53 geschrieben: [...] I just downloaded the sources from sourceforge and after compile it, the splash window and the about dialog show version 1.9.0, I tried the next downloads.

Oops. My fault. I will upload the right versions.

Mattias

El 06/12/17 a les 20:15, Mattias Gaertner via Lazarus ha escrit:

The Lazarus team is glad to announce the release of Lazarus 1.8.

I see that, while under linux there's no problem, under windows it cannot be built with fpc-2.6.4.
Now I managed to build it by transplanting the missing unicode units from fpc-3.0, but I'm not sure it will work correctly. I guess I'll have to switch to fpc 3.0.4 but I still fear the unicode changes in 3.0 will bite me.

Bye

Luca Olivetti wrote

I guess I'll have to switch to fpc 3.0.4 but I still fear the unicode changes in 3.0 will bite me.

At some point you'll have to anyway.
Better bite the bullet.

I hardly had to make any had some callback procedures I had to change the signature of.

Then again, I don't do any databse stuff, which may have more problems IIRC.

Bart

Luca Olivetti wrote

I see that, while under linux there's no problem, under windows it cannot be built with fpc-2.6.4.

How does it fail? Lazarus 1.8 was meant to still compile with FPC 2.6.4. However with Lazarus trunk I would be happy to go ahead and forget 2.6.4. If the fix for Lazarus 1.8 is easy, please upload a patch.

Juha

Juha Manninen wrote

How does it fail? Lazarus 1.8 was meant to still compile with FPC 2.6.4. However with Lazarus trunk I would be happy to go ahead and forget 2.6.4. If the fix for Lazarus 1.8 is easy, please upload a patch.

Tested on Win32 where it fails in interfaces\win32\win32int because unit "character" is not found. It is in rtl\objpas of fpc3.04, but not of fpc 2.6.4, it came in with fpc 3.0.0

El 09/12/17 a les 00:41, Werner Pamler via Lazarus ha escrit:

Yes, my dirty hack was to copy character.pas from fpc-3.0.0 (and all other needed units[*]) to interfaces\win32.
I also add to edit character.pas to add the missing resource strings and the definition of ENoConstructException

--- fpc-3.0.0/rtl/objpas/character.pas 2013-08-20 06:55:05.000000000 +0200 +++ laz_1_8/lcl/interfaces/win32/character.pas 2017-12-09 11:51:22.606416879 +0100
@@ -195,6 +195,19 @@
SysUtils,

    RtlConsts;

+ResourceString
+ SStringIndexOutOfRange = 'String index %d out of range [1 - %d]';
+ SHighSurrogateOutOfRange = 'High surrogate $%x out of range [$D800 - $DBFF]';
+ SLowSurrogateOutOfRange = 'Low surrogate $%x out of range [$DC00 - $DFFF]';
+ SInvalidUTF32Char = 'Invalid UTF32 character $%x. Valid UTF32 character must be in range [$0 - $10FFFF] except surrogate range [$D800-$DFFF]';
+ SInvalidHighSurrogate = 'Invalid high surrogate at index %d. High surrogate must be followed by a low surrogate pair';
+ SInvalidUnicodeCodePointSequence = 'Invalid unicode code point sequence';
+ SClassCantBeConstructed = 'Class %s can not be constructed'; +
+type
+
+ ENoConstructException = class(Exception);
+

  const
    LETTER_CATEGORIES = [
      TUnicodeCategory.ucUppercaseLetter,

TUnicodeCategory.ucLowercaseLetter,

[*] the complete list of missing units is

? character.pas
? unicodedata.inc
? unicodedata.pas
? unicodedata_be.inc
? unicodedata_le.inc
? unicodenumtable.pas
? weight_derivation.inc

I know I'll have to upgrade to fpc 3 someday, but what worries me about it is my (and other 3rd party units) use of ansistrings as binary buffers. I did some quick tests some time ago an nothing seemed broken, but I'm somewhat reluctant since my software has to run unattended 24/7 (or more like 24/5).
OTOH I can always keep a copy of fpc-2.6.4 and lazarus 1.6.4 recompile if something goes wrong.

Bye

Werner Pamler wrote

Tested on Win32 where it fails in interfaces\win32\win32int because unit "character" is not found. It is in rtl\objpas of fpc3.04, but not of fpc 2.6.4, it came in with fpc 3.0.0

Damn, it is added by me.
It is used only for function TCharacter.IsSurrogate(). Do you know a replacement function without copying the whole unit from FPC 3.0? LazUTF16 only has:
function IsUTF16CharValid(AChar, ANextChar: WideChar): Boolean; A fix could be committed to the 1.8 fixes branch as a special case If no easy fix is found then I could add:
if {$IF FPC_FULLVERSION<30000}False{$ELSE}TCharacter.IsSurrogate(WideChar(OrgCharCode)){$ENDIF} then
and the code works with FPC 2.6.4 as it did before my changes.

The call to TCharacter.IsSurrogate() fixes this issue only partly : https://bugs.freepascal.org/view.php?id=32101
UTF8KeyPress is still not called. Somebody with better knowledge of LCL-Win32 should fix it.

Luca Olivetti wrote

I know I'll have to upgrade to fpc 3 someday, but what worries me about it is my (and other 3rd party units) use of ansistrings as binary buffers.

The automatic conversions by FPC 3 could indeed mess things up. However you should change such binary buffers to use TBytes or similar in any case. There have always been warnings against using String type for binary data. In future your code may use $mode DelphiUnicode and it breaks for sure.
Maybe switching to FPC 3 is the right moment to fix the buffers, too.

Juha

El 09/12/17 a les 13:23, Juha Manninen via Lazarus ha escrit:

I think I had this discussion before: I'm not going to use TBytes until it offers the same convenience as an ansistring (i.e. easy concatenation, insertion and deletion, not to mention search and replace, etc.).
Besides, some of it is in 3rd party libraries (but I think they explicitly use ansistring, not the generic string type).

There have always been warnings against using String type for binary data. In future your code may use $mode DelphiUnicode and it breaks for sure.

There weren't when I started using them in delphi 20 years ago (with no dynamic arrays in sight).

Bye

I think I had this discussion before: I'm not going to use TBytes until

it offers the same convenience as an ansistring (i.e. easy concatenation, insertion and deletion, not to mention search and replace, etc.).

Isn't RawByteString a more suited replacement then? In the few places i relied on the bytes in a string when i was converting my code from FPC 2.x to FPC 3.x i used a RawByteString and things seem to work fine. -------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171210/a3a624d5/attachment.html>

El 10/12/17 a les 15:02, Kostas Michalopoulos via Lazarus ha escrit:

https://www.freepascal.org/docs-html/rtl/system/rawbytestring.html

it's an ansistring.

Bye

El 09/12/17 a les 11:57, Luca Olivetti via Lazarus ha escrit:

I've done some tests with 3.0.4 and it turns out I'm bitten by the extensive changes in tthread: freeing a thread in its OnTerminate[*] randomly (i.e. not every time) produces a sigsev (under win32). With 2.6.4 it didn't happen (though freeing a thread in OnTerminate had another problem undes Linux, see
https://bugs.freepascal.org/view.php?id=32759)

[*] I don't know if freeing a thread in OnTerminate is allowed. FPC documentation here
https://www.freepascal.org/docs-html/rtl/classes/tthread.onterminate.html says nothing about it, but
the example here (mentioned in the bug report)
http://docwiki.embarcadero.com/CodeExamples/Tokyo/en/TThreadYield_(Delphi) seems to imply that it should.

For the time being I'm using FreeOnTerminate:=true and avoid freeing the thread in OnTerminate (though there are other changes required to my code and "fixing" working code is always an issue). I checked the user changes pages for 3.0.0, 3.0.2 and 3.0.4 and there's no mention of changes in threads (the only one is this one http://wiki.freepascal.org/User_Changes_3.0#cthreads:_critical_sections_and_recursive_mutex_support and doesn't apply here).

Bye

It is definitely not allowed in FPC, and I will document it as such.

Michael.

El 11/12/17 a les 13:43, Michael Van Canneyt ha escrit:

Ouch, that's not delphi compatible then (unless the example is flawed). In any case it's better to be not compatible and clearly documented than vaguely documented and maybe (just maybe) compatible (i.e. I don't really care about delphi compatibility since I don't use it, but sometimes you have to rely on delphi examples/documentation).

Bye

The following is clearly nonsense:

  FThread.FreeOnTerminate := False;  // Free FThread in the OnTerminate proc. This will set Terminated to true

Terminated is already True in OnTerminate.

So I think the example is flawed.

From the TThread implementation:

  if (FThreadID = GetCurrentThreadID) then
     begin
       if not(FFreeOnTerminate) and not FFinished then
         raise EThreadDestroyCalled.Create('A thread cannot destroy itself except by setting FreeOnTerminate and leaving!');

I think we cannot be more clear than that :)

Michael.

El 11/12/17 a les 16:14, Michael Van Canneyt ha escrit:

Not really: OnTerminate is called in the context of the main thread, so FThreadId is different than GetCurrentThreadId.

Bye

As usual after istalled fpc packages (fpc-src_3.0.4-1_amd64.deb <https://sourceforge.net/projects/lazarus/files/Lazarus%20Linux%20amd64%20DEB/Lazarus%201.8.0/fpc-src_3.0.4-1_amd64.deb/download> and fpc_3.0.4-1_amd64.deb
<https://sourceforge.net/projects/lazarus/files/Lazarus%20Linux%20amd64%20DEB/Lazarus%201.8.0/fpc_3.0.4-1_amd64.deb/download>) the "make all" for source 1.8.0 fails.
It is ok using fpc-src_3.0.4-rc1_amd64.deb
<https://sourceforge.net/projects/lazarus/files/Lazarus%20Linux%20amd64%20DEB/Lazarus%201.8.0RC5/fpc-src_3.0.4-rc1_amd64.deb/download> and fpc_3.0.4-rc1_amd64.deb
<https://sourceforge.net/projects/lazarus/files/Lazarus%20Linux%20amd64%20DEB/Lazarus%201.8.0RC5/fpc_3.0.4-rc1_amd64.deb/download>

/usr/bin/ld: attenzione: ./link.res contiene sezioni di output; forse è stata dimenticata -T
/usr/bin/ld: /usr/lib/fpc/3.0.4/units/x86_64-linux/rtl/cprt0.o: unrecognized relocation (0x2a) in section `.text'
/usr/bin/ld: link finale non riuscito: Bad value
Error: Error while linking
Fatal: There were 1 errors compiling module, stopping Fatal: Compilation aborted
Makefile:2737: set di istruzioni per l'obiettivo "lazres" non riuscito make[2]: *** [lazres] Errore 1
make[2]: uscita dalla directory "/home/pierluigi/programs/lazarus/tools" Makefile:3155: set di istruzioni per l'obiettivo "all" non riuscito make[1]: *** [all] Errore 2
make[1]: uscita dalla directory "/home/pierluigi/programs/lazarus/tools" Makefile:3267: set di istruzioni per l'obiettivo "lazbuild" non riuscito

Regards

2017-12-06 20:15 GMT+01:00 Mattias Gaertner via Lazarus <

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171213/93093ee4/attachment.html>

There is a similar problem with the fpc rpms.
GNU binutils 2.26 added a new relocation type. I need to rebuild the debs on an older Linux distribution.

Mattias

I uploaded new amd64 debs built on a system with binutils 2.24.

Mattias

I am trying to update to lazarus 1.8 on my OSX (I was using 1.6.4). I downloaded and installed, fpc, fpc src and lazarus. First run is ok.
Then if I recompile lazarus (clean build all) everything seems to be ok but if I try to start lazarus I got an error message. See message1. I close the message try to restart again lazarus and I got the message2.

What is going wrong?

Il 06/12/2017 20:15, Mattias Gaertner via Lazarus ha scritto:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: message1.jpg
Type: image/jpeg
Size: 13559 bytes
Desc: not available
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171213/834f0b8f/attachment-0002.jpg> -------------- next part --------------
A non-text attachment was scrubbed...
Name: message2.jpg
Type: image/jpeg
Size: 14699 bytes
Desc: not available
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171213/834f0b8f/attachment-0003.jpg>

Now it's ok, thanks
Salvatore

2017-12-13 12:25 GMT+01:00 Mattias Gaertner via Lazarus <

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171213/a8cdd62d/attachment.html>

Anyone has suggestions?

Il 13/12/2017 12:44, Andrea Mauri ha scritto:

An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171215/d690b29c/attachment.html>

Did you run Lazarus as the same user as it was installed? Is the user an admin user?
How did you open the IDE?

Can you send me the text of IDE menu
View / IDE Internals / About IDE / General?

Mattias

On Sun, Dec 10, 2017 at 4:17 PM, Luca Olivetti via Lazarus <

wrote: https://www.freepascal.org/docs-html/rtl/system/rawbytestring.html it's an ansistring.

This is IMO misleading since nowadays AnsiString has codepage information and a plain AnsiString as shown in the page you linked can be any codepage whereas RawByteString is always CP_NONE. RawByteString and AnsiString are not equivalent.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171215/4be6ea4f/attachment.html>

On Fri, 15 Dec 2017 13:32:24 +0200

At compile time it is CP_NONE, at runtime it can be any codepage.

RawByteString and AnsiString are not equivalent.

True.

Mattias

On Fri, Dec 15, 2017 at 1:41 PM, Mattias Gaertner via Lazarus <

wrote: At compile time it is CP_NONE, at runtime it can be any codepage.

My understanding (which, if it is wrong, sorry but not my fault since the docs are a bit vague and all the names involved are totally unintuitive :-P) is that RawByteString has basically no codebase so it is treated as a byte string pretty much the same way AnsiString was treated pre-FPC3. So i'd expect the codebase to always be CP_NONE and to change it, i'd have to assign the RawByteString to some string type with an associated codepage (basically, making the RBS an "escape hatch" to the whole conversion machinery).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171215/47053a12/attachment.html>

same user. only one user on my Mac, I suppose I am admin (I am a OS X newbie).
I opened using launchpad and directly from finder.

Here the text of IDE menu:

Lazarus version: 1.8.0

Lazarus svn revision: exported

Lazarus build date: 2017/12/04

Lazarus was compiled for i386-darwin

Lazarus was compiled with fpc 3.0.4

Global IDE options:

Primary config directory=/Users/giacca/.lazarus

Secondary config directory=/etc/lazarus

LazarusDirectory=/Developer/lazarus/

Real LazarusDirectory=/Developer/lazarus/

Default CompilerFilename=/usr/local/bin/ppc386

Real Default CompilerFilename=/usr/local/bin/ppc386

Project CompilerFilename=$(CompPath)

Real Project CompilerFilename=/usr/local/bin/ppc386

CompilerMessagesFilename=

Real
CompilerMessagesFilename=/usr/local/share/fpcsrc/compiler/msg/errore.msg

FPC source directory=/usr/local/share/fpcsrc

Real FPC source directory=/usr/local/share/fpcsrc/

Test directory=~/tmp/

Real Test directory=/Users/giacca/tmp/

Working directory=/

Application.ExeName=/Developer/lazarus/lazarus.app/Contents/MacOS/lazarus

Parameters:

/Developer/lazarus/lazarus.app/Contents/MacOS/lazarus

Environment variables:

TMPDIR=/var/folders/z2/958dg9vn2s93cm1rky6j_d9w0000gn/T/

TMPDIR=/var/folders/z2/958dg9vn2s93cm1rky6j_d9w0000gn/T/

__CF_USER_TEXT_ENCODING=0x1F5:0x0:0x0

SHELL=/bin/bash

HOME=/Users/giacca

Apple_PubSub_Socket_Render=/private/tmp/com.apple.launchd.M23sK4oq3B/Render

SSH_AUTH_SOCK=/private/tmp/com.apple.launchd.VYyuGjeRvT/Listeners

PATH=/usr/bin:/bin:/usr/sbin:/sbin

LOGNAME=giacca

XPC_SERVICE_NAME=lazarus.freepascal.ide.8732

USER=giacca

2017-12-15 12:11 GMT+01:00 Mattias Gaertner via Lazarus <

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20171218/354ba09d/attachment.html>