Re: [fpc-devel] Attn Florian: r47852 broke building of i386

2021-01-03 Thread J. Gareth Moreton via fpc-devel
Sorry, it was!  I got a clean copy of the RTL and it compiled successfully.  Sorry about that.  Glad I checked here first before reporting a bug! Gareth aka. Kit On 03/01/2021 17:51, Florian Klämpfl via fpc-devel wrote: Am 02.01.2021 um 18:47 schrieb J. Gareth Moreton via fpc-devel: There sti

Re: [fpc-devel] Attn Florian: r47852 broke building of i386

2021-01-03 Thread J. Gareth Moreton via fpc-devel
Oh boy, if it's an accidental modification, I'll be kicking myself!  Thanks. Gareth aka. Kit On 03/01/2021 17:51, Florian Klämpfl via fpc-devel wrote: Am 02.01.2021 um 18:47 schrieb J. Gareth Moreton via fpc-devel: There still seems to be a problem with i386-win32 - this is from using FPC 3.2.

Re: [fpc-devel] Attn Florian: r47852 broke building of i386

2021-01-03 Thread Florian Klämpfl via fpc-devel
Am 02.01.2021 um 18:47 schrieb J. Gareth Moreton via fpc-devel: There still seems to be a problem with i386-win32 - this is from using FPC 3.2.0 as a base compiler and calling: make distclean all install CPU_TARGET=i386 OS_TARGET=win32 ... C:/FPC/3.2.0/bin/i386-Win32/ppc386.exe -Ur -Xs -O2 -n

[fpc-devel] TRegExpr serious regression in 3.2.0

2021-01-03 Thread Martin Frb via fpc-devel
In 3.0.4 the below gives true for all 4 lines In 3.2.0 the last 2 matches give false. 0*  does not match 00 when it appears in the middle of an expression program Project1; {$mode objfpc}{$H+} uses   RegExpr; var   FRegX: TRegExpr; begin   FRegX := TRegExpr.Create;   FRegX.ModifierI := True;