Menu
▾
▴
tcl-bugs — DO NOT EMAIL DIRECTLY - USE BUG TRACKER INSTEAD
You can subscribe to this list here.
| 2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(548) |
Dec
(254) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2001 |
Jan
(207) |
Feb
(364) |
Mar
(1050) |
Apr
(1250) |
May
(304) |
Jun
(235) |
Jul
(347) |
Aug
(896) |
Sep
(1110) |
Oct
(666) |
Nov
(657) |
Dec
(500) |
| 2002 |
Jan
(706) |
Feb
(677) |
Mar
(785) |
Apr
(606) |
May
(449) |
Jun
(957) |
Jul
(1125) |
Aug
(628) |
Sep
(413) |
Oct
(686) |
Nov
(368) |
Dec
(428) |
| 2003 |
Jan
(458) |
Feb
(956) |
Mar
(690) |
Apr
(525) |
May
(643) |
Jun
(449) |
Jul
(788) |
Aug
(408) |
Sep
(569) |
Oct
(815) |
Nov
(985) |
Dec
(604) |
| 2004 |
Jan
(558) |
Feb
(716) |
Mar
(897) |
Apr
(626) |
May
(1010) |
Jun
(693) |
Jul
(890) |
Aug
(641) |
Sep
(1050) |
Oct
(1014) |
Nov
(1107) |
Dec
(738) |
| 2005 |
Jan
(587) |
Feb
(472) |
Mar
(661) |
Apr
(761) |
May
(839) |
Jun
(740) |
Jul
(657) |
Aug
(610) |
Sep
(566) |
Oct
(760) |
Nov
(768) |
Dec
(544) |
| 2006 |
Jan
(359) |
Feb
(261) |
Mar
(1211) |
Apr
(546) |
May
(356) |
Jun
(340) |
Jul
(292) |
Aug
(535) |
Sep
(611) |
Oct
(855) |
Nov
(713) |
Dec
(579) |
| 2007 |
Jan
(426) |
Feb
(561) |
Mar
(438) |
Apr
(595) |
May
(621) |
Jun
(668) |
Jul
(354) |
Aug
(370) |
Sep
(832) |
Oct
(1139) |
Nov
(1228) |
Dec
(839) |
| 2008 |
Jan
(682) |
Feb
(404) |
Mar
(801) |
Apr
(676) |
May
(552) |
Jun
(827) |
Jul
(758) |
Aug
(902) |
Sep
(335) |
Oct
(815) |
Nov
(870) |
Dec
(1504) |
| 2009 |
Jan
(913) |
Feb
(889) |
Mar
(521) |
Apr
(616) |
May
(415) |
Jun
(346) |
Jul
(597) |
Aug
(423) |
Sep
(321) |
Oct
(680) |
Nov
(779) |
Dec
(711) |
| 2010 |
Jan
(725) |
Feb
(632) |
Mar
(889) |
Apr
(575) |
May
(432) |
Jun
(354) |
Jul
(236) |
Aug
(602) |
Sep
(683) |
Oct
(520) |
Nov
(538) |
Dec
(372) |
| 2011 |
Jan
(380) |
Feb
(76) |
Mar
(315) |
Apr
(307) |
May
(149) |
Jun
(220) |
Jul
(288) |
Aug
(607) |
Sep
(272) |
Oct
(239) |
Nov
(170) |
Dec
(124) |
| 2012 |
Jan
(387) |
Feb
(293) |
Mar
(292) |
Apr
(435) |
May
(381) |
Jun
(405) |
Jul
(587) |
Aug
(334) |
Sep
(278) |
Oct
(133) |
Nov
(353) |
Dec
(314) |
| 2013 |
Jan
(448) |
Feb
(335) |
Mar
(327) |
Apr
(177) |
May
(184) |
Jun
(142) |
Jul
(65) |
Aug
(52) |
Sep
(119) |
Oct
(156) |
Nov
(192) |
Dec
(68) |
| 2014 |
Jan
(123) |
Feb
(266) |
Mar
(192) |
Apr
(161) |
May
(558) |
Jun
(212) |
Jul
(327) |
Aug
(401) |
Sep
(552) |
Oct
(529) |
Nov
(337) |
Dec
(304) |
| 2015 |
Jan
(189) |
Feb
(334) |
Mar
(346) |
Apr
(282) |
May
(379) |
Jun
(286) |
Jul
(306) |
Aug
(216) |
Sep
(279) |
Oct
(465) |
Nov
(291) |
Dec
(312) |
| 2016 |
Jan
(505) |
Feb
(386) |
Mar
(545) |
Apr
(340) |
May
(306) |
Jun
(386) |
Jul
(558) |
Aug
(427) |
Sep
(427) |
Oct
(361) |
Nov
(410) |
Dec
(311) |
| 2017 |
Jan
(503) |
Feb
(309) |
Mar
(426) |
Apr
(638) |
May
(778) |
Jun
(688) |
Jul
(299) |
Aug
(548) |
Sep
(524) |
Oct
(651) |
Nov
(1050) |
Dec
(844) |
| 2018 |
Jan
(619) |
Feb
(349) |
Mar
(456) |
Apr
(441) |
May
(493) |
Jun
(413) |
Jul
(434) |
Aug
(395) |
Sep
(500) |
Oct
(827) |
Nov
(792) |
Dec
(512) |
| 2019 |
Jan
(457) |
Feb
(390) |
Mar
(621) |
Apr
(642) |
May
(642) |
Jun
(538) |
Jul
(477) |
Aug
(632) |
Sep
(715) |
Oct
(949) |
Nov
(886) |
Dec
(582) |
| 2020 |
Jan
(594) |
Feb
(442) |
Mar
(701) |
Apr
(1385) |
May
(856) |
Jun
(550) |
Jul
(810) |
Aug
(886) |
Sep
(1186) |
Oct
(1095) |
Nov
(842) |
Dec
(675) |
| 2021 |
Jan
(701) |
Feb
(579) |
Mar
(654) |
Apr
(922) |
May
(697) |
Jun
(558) |
Jul
(335) |
Aug
(243) |
Sep
(283) |
Oct
(276) |
Nov
(469) |
Dec
(626) |
| 2022 |
Jan
(520) |
Feb
(687) |
Mar
(675) |
Apr
(389) |
May
(512) |
Jun
(492) |
Jul
(446) |
Aug
(489) |
Sep
(1115) |
Oct
(801) |
Nov
(665) |
Dec
(418) |
| 2023 |
Jan
(440) |
Feb
(579) |
Mar
(893) |
Apr
(917) |
May
(782) |
Jun
(398) |
Jul
(850) |
Aug
(596) |
Sep
(928) |
Oct
(602) |
Nov
(1089) |
Dec
(696) |
| 2024 |
Jan
(691) |
Feb
(458) |
Mar
(650) |
Apr
(670) |
May
(829) |
Jun
(831) |
Jul
(597) |
Aug
(1113) |
Sep
(436) |
Oct
(1424) |
Nov
(728) |
Dec
(710) |
| 2025 |
Jan
(707) |
Feb
(505) |
Mar
(875) |
Apr
(620) |
May
(843) |
Jun
(790) |
Jul
(612) |
Aug
(855) |
Sep
(927) |
Oct
(682) |
Nov
|
Dec
|
|
From: <no...@tc...> - 2025-10-27 11:33:39
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [dcbdfeaf81ede3d0f900f63a905a256cd76f6f96c0c8cd4b465d55e3ea1bcf1f] [Tabs on macOS have noticable visual overlap/cutoff issues] By anonymous For Tk On 2025-10-27T11:26:05.554 Details https://core.tcl-lang.org/tk/tinfo?name=dcbdfeaf81ede3d0f900f63a905a256cd76f6f96c0c8cd4b465d55e3ea1bcf1f Ticket https://core.tcl-lang.org/tk/tktview/1b33bbae49cc36e4ff595d5f52c601baac2610ac Changed Fields assignee: nobody closer: nobody comment: Tabs on macOS visibly overlap with each other, and the right-most tab is cut off. Here's an image of the issue: <https://i.sstatic.net/M6fvFM7p.png> Example code to reproduce the issue: ```python from tkinter import * from tkinter import ttk root = Tk() root.geometry("400x600") notebook = ttk.Notebook(root) first_tab = ttk.Frame(notebook) notebook.add(first_tab, text="one") second_tab = ttk.Frame(notebook) notebook.add(second_tab, text="two") third_tab = ttk.Frame(notebook) notebook.add(third_tab, text="three") notebook.pack(fill="both", expand=1) root.mainloop() ``` (Image and example code are from <https://stackoverflow.com/q/79761277/4901968>, but the issue also reproduces on my machine with the example code, IDLE's settings GUI, and my own application.) foundin: 9.0.2 is_private: 0 login: anonymous priority: 5 Medium resolution: None severity: Minor status: Open submitter: anonymous subsystem: 02. Appearance title: Tabs on macOS have noticable visual overlap/cutoff issues type: Bug ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 11:33:36
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [d49dc5ab2c3f6bf8738c9dc5fae219598c0eb1ba69676032b3607dc1a40628bd] [Tabs on macOS have noticable visual overlap/cutoff issues] By anonymous For Tk On 2025-10-27T11:27:53.760 Details https://core.tcl-lang.org/tk/tinfo?name=d49dc5ab2c3f6bf8738c9dc5fae219598c0eb1ba69676032b3607dc1a40628bd Ticket https://core.tcl-lang.org/tk/tktview/1b33bbae49cc36e4ff595d5f52c601baac2610ac Changed Fields comment: Tabs on macOS visibly overlap with each other, and the right-most tab is cut off. Here's an image of the issue: <https://i.sstatic.net/M6fvFM7p.png> Example code to reproduce the issue: ```python from tkinter import * from tkinter import ttk root = Tk() root.geometry("400x600") notebook = ttk.Notebook(root) first_tab = ttk.Frame(notebook) notebook.add(first_tab, text="one") second_tab = ttk.Frame(notebook) notebook.add(second_tab, text="two") third_tab = ttk.Frame(notebook) notebook.add(third_tab, text="three") notebook.pack(fill="both", expand=1) root.mainloop() ``` (Image and example code are from <https://stackoverflow.com/q/79761277/4901968>, but the issue also reproduces on my machine running macOS Sonoma 14.8 with the example code, IDLE's settings GUI, and my own application.) login: anonymous ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 11:18:45
|
Automated mail by fx, on behalf of no...@tc... Wiki Change [ed7c15346cfb960d839bbad61bbcf85c458d1b9da2e210933ba615750b7bc319] Page branch/c11-full-hash-foreach By dkf For Tcl On 2025-10-27T11:16:30.286 Details https://core.tcl-lang.org/tcl/wiki?name=branch%2Fc11-full-hash-foreach ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 10:49:20
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [aa1f1e6cce0f522aecff70aa8a765d3a7fe7f0737e36ab825134f33eab50fbd8] ["update idletasks" command is about 8 times slower] By oehhar For Tk On 2025-10-27T10:40:32.700 Details https://core.tcl-lang.org/tk/tinfo?name=aa1f1e6cce0f522aecff70aa8a765d3a7fe7f0737e36ab825134f33eab50fbd8 Ticket https://core.tcl-lang.org/tk/tktview/7168473e104fe406d4302c69605ab268ea380d4f Changed Fields icomment: Sorry, this is correctly the Tk repo. You are apparently on the Mac platform. The MAC code got completly changed and now has a different drawing strategy. It now draws on a shadow page, instead directly on the screen. The older interface got deprecated by Apple and there were always drawing issues, that regions were not updated. To my knowledge, the new drawing is just a big step foreward. I suppose, the different timings you are seeing are due to that reform. login: oehhar username: oehhar ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 10:48:49
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [143ede2b49e2e2211eb351b420ce5cffe46b49347a511db7eccb29ec44fad898] [Mingw-64 has SEH] By oscarfv For Tk On 2025-10-27T10:40:12.845 Details https://core.tcl-lang.org/tk/tinfo?name=143ede2b49e2e2211eb351b420ce5cffe46b49347a511db7eccb29ec44fad898 Ticket https://core.tcl-lang.org/tk/tktview/7d6590d964a6f5c29edcaaac9d66dffdc261e40c Changed Fields closedate: 2025-10-27 10:40:12 closer: oscarfv icomment: Yeah, the patch makes no sense. Digging on the git history, it looks like a half-baked early experiment. Please accept my apologies. login: oscarfv ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 10:34:20
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [3caf8f52d82308b133e809091b7146772f343e648ca0937facad322958ff1b9d] ["update idletasks" command is about 8 times slower] By achirous For Tk On 2025-10-27T10:31:46.207 Details https://core.tcl-lang.org/tk/tinfo?name=3caf8f52d82308b133e809091b7146772f343e648ca0937facad322958ff1b9d Ticket https://core.tcl-lang.org/tk/tktview/7168473e104fe406d4302c69605ab268ea380d4f Changed Fields icomment: Here is more build info: tcl_platform(byteOrder) = littleEndian tcl_platform(engine) = Tcl tcl_platform(machine) = arm64 tcl_platform(os) = Darwin tcl_platform(osVersion) = 25.0.0 tcl_platform(pathSeparator) = : tcl_platform(platform) = unix tcl_platform(pointerSize) = 8 tcl_platform(wordSize) = 8 I thought this was the TK repo. Could someone please point me in the right direction if this is the wrong place to post this? login: achirous ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 10:33:50
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [98591c32155b825473ff018be24c68ed04261c28c6132e898f01774e72d4c46b] [Don't use static libgcc when building as shared lib] By oscarfv For Tk On 2025-10-27T10:26:14.964 Details https://core.tcl-lang.org/tk/tinfo?name=98591c32155b825473ff018be24c68ed04261c28c6132e898f01774e72d4c46b Ticket https://core.tcl-lang.org/tk/tktview/71b6efd8f0ac51bf56a5e944f3128a1df9901191 Changed Fields closedate: 2025-10-27 10:26:14 closer: oscarfv icomment: One big problem with linking statically to libgcc in plugin-based applications (such as tcl/tk when using binary packages) is that destruction of global objects may break, exchange of C runtime objects may also break (due to differences on internal details), exception handling may break (there are multiple options in that field: SEH, Dwarf and setjmp/longjmp is still around.) So, it is quite risky to mix gcc dlls without dynamically linking libgcc as well *and* using the same libgcc flavor. Contrary to the the _MINGW_ANSI_STDIO issue, this is more of a nuisance, as there is no way of removing -static-libgcc without patching the tcl/tk build. Can we have a configure option to disable -static-libgcc? Would such a patch acceptable? login: oscarfv ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 10:18:55
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [93826b53189e923242d1a45a89fbe94c12b0bd9b4005f8fc071054dc01cd95c1] [Makefile forces -D__USE_MINGW_ANSI_STDIO=0] By oscarfv For Tk On 2025-10-27T10:08:48.607 Details https://core.tcl-lang.org/tk/tinfo?name=93826b53189e923242d1a45a89fbe94c12b0bd9b4005f8fc071054dc01cd95c1 Ticket https://core.tcl-lang.org/tk/tktview/b405da2aad74715fdf8c729fdcfcc0a75d56b64b Changed Fields closedate: 2025-10-27 10:08:48 closer: oscarfv icomment: Thank you for the detailed explanation. A few remarks: I64 is supported by mingw-w64 for many years now. On your explanation I think you are mentioning that the tcl/tk build sets __USE_MINGW_ANSI_STDIO=1 when it actually is =0. MSYS2/Mingw-packages wants to be a complete repository. We even have twapi now :-) Hence, compatibility with VS is a non-goal, although I understand why it is important for you as a policy. For these reasons we probably end overriding CFLAGS to remove __USE_MINGW_ANSI_STDIO=0 and be consistent with the rest of the MSYS2/Mingw-packages ecosystem. I mention this just in case it is relevant for future exchanges. It would be nice if we had an explicit knob (i.e. a configure option) instead of a blanket override, but for now CFLAGS= would do. Thank you again. login: oscarfv ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 09:33:51
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [4a491752f7be5f181b8ab7b39a84a14d53f53acd58112da3bad062040b357ac7] [Install pkg-config file] By jan.nijtmans For Tk On 2025-10-27T09:19:17.438 Details https://core.tcl-lang.org/tk/tinfo?name=4a491752f7be5f181b8ab7b39a84a14d53f53acd58112da3bad062040b357ac7 Ticket https://core.tcl-lang.org/tk/tktview/2c240bb0f59d4f124bdbdf5c3ce28d70c41c95fb Changed Fields icomment: Looks good to me. Thanks! login: jan.nijtmans ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 09:23:44
|
Automated mail by fx, on behalf of no...@tc... By jan.nijtmans For Tk On 2025-10-27T09:08:41.073 Changed https://core.tcl-lang.org/tk/info/ad32c99356f987d46727974c301fc45572ab2bb0325bff8ec6432abc01aa3664 Tag +closed Tag +comment Fix for [7d6590d964]: Mingw-64 has SEH Build fails: [https://github.com/tcltk/tk/actions/runs/18832518106/job/53726499263] Tag -sym-core-bug-7d6590d964 ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 09:23:30
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [cfaa37561b92c2a5c0d2651a02dac2392b1cd336ede2c89577ee4a5c39370ff7] [Mingw-64 has SEH] By jan.nijtmans For Tk On 2025-10-27T09:07:43.642 Details https://core.tcl-lang.org/tk/tinfo?name=cfaa37561b92c2a5c0d2651a02dac2392b1cd336ede2c89577ee4a5c39370ff7 Ticket https://core.tcl-lang.org/tk/tktview/7d6590d964a6f5c29edcaaac9d66dffdc261e40c Changed Fields closedate: 2025-10-27 09:07:43 closer: jan.nijtmans icomment: As I suspected, the build fails with your suggestion: See: [https://github.com/tcltk/tk/actions/runs/18832518106/job/53726499263] login: jan.nijtmans ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 09:23:29
|
Automated mail by fx, on behalf of no...@tc... Commit [f4fafa6feb2c48686011edee9c9d24f04e9594333f26fc24f02cc29c47fa7207] By dkf For Tcl (branch: c11-full-hash-foreach) On 2025-10-27T09:07:01.679 Details https://core.tcl-lang.org/tcl/info/f4fafa6feb2c48686011edee9c9d24f04e9594333f26fc24f02cc29c47fa7207 Description Extend the use of the FOREACH_HASH macro set Tcl-wide Changed Files 25 edited generic/tclAssembly.c generic/tclBasic.c generic/tclBinary.c generic/tclCmdIL.c generic/tclCompCmdsSZ.c (truncated) ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 09:03:59
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [5db318af49f5343b897bd26bd045ae2021f7c5003cd63f8597ada538f4dc7437] [Don't use static libgcc when building as shared lib] By jan.nijtmans For Tk On 2025-10-27T09:00:15.972 Details https://core.tcl-lang.org/tk/tinfo?name=5db318af49f5343b897bd26bd045ae2021f7c5003cd63f8597ada538f4dc7437 Ticket https://core.tcl-lang.org/tk/tktview/71b6efd8f0ac51bf56a5e944f3128a1df9901191 Changed Fields assignee: jan.nijtmans closedate: 2025-10-27 09:00:15 closer: jan.nijtmans foundin: 8.6 icomment: This one is also intentional. When building Tcl in any environment, it should be possible to copy the tclsh executable and the dll's (e.g. with an installer) to another machine, and it works. If you don't compile with -static-libgcc, an additional dependency with a Mingw-dependant dll is created (I don't remember which one). A Visual Studio compiled version of Tcl only needs tcl86.dll, zlib1.dll (and libtommath.dll, for 9.0+). So if we want to make gcc/clang's version of tcl86.dll a drop-in replacement for the Visual Studio build of tcl86.dll, we need -static-libgcc. It only makes the dll a tiny bit larger. So, sorry, I have to reject this ticket as well :-( I fully understand that in a MSYS2/Mingw environment, you want to make a different choice. login: jan.nijtmans resolution: Rejected status: Closed ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 09:03:49
|
Automated mail by fx, on behalf of no...@tc... Commit [356e364f5318bc6d8c1e94c8186bb657a469928c503c11fbc3450aead456be78] By oehhar For Tk (branch: core-9-0-branch) On 2025-10-27T08:55:33.993 Details https://core.tcl-lang.org/tk/info/356e364f5318bc6d8c1e94c8186bb657a469928c503c11fbc3450aead456be78 Description [6051a9fc] Also back-out changes message Changed Files 1 edited changes.md ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 08:48:46
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [42a193f1074216afdd2b8057d497a58826f0886f2273bb94bb6ba8c7b7958a4a] [Makefile forces -D__USE_MINGW_ANSI_STDIO=0] By jan.nijtmans For Tk On 2025-10-27T08:45:50.933 Details https://core.tcl-lang.org/tk/tinfo?name=42a193f1074216afdd2b8057d497a58826f0886f2273bb94bb6ba8c7b7958a4a Ticket https://core.tcl-lang.org/tk/tktview/b405da2aad74715fdf8c729fdcfcc0a75d56b64b Changed Fields assignee: jan.nijtmans closedate: 2025-10-27 08:45:50 closer: jan.nijtmans foundin: 8.6 icomment: Thanks for submitting all those reports! I'm sure that there are issues with MSYS2/Mingw in Tcl, I would like to fix those here. However, this one is done intentionally. I'll explain. The gcc/clang Tcl build is supposed to be a drop-in replacement for a Visual Studio build of Tcl. That means, whenever there is a incompatibility between gcc/clang and Visual Studio, Visual Studio always wins (no matter whether we like it or not, it might very well be that the gcc/clang implementation is better). Also, binary extensions (e.g. itcl) can be compiled with Visual Studio, this should work fine even though Tcl is compiled with gcc/clang. So, you are right that "If an application that uses __USE_MINGW_ANSI_STDIO=1 embeds tcl/tk, it *might* cause issues.". The other way around means that if an application is compiled with Visual Studio and uses a Tcl dll built with gcc/clang, this *might* cause issues as well. Since Visual Studio always wins, we have to choose for __USE_MINGW_ANSI_STDIO=1. I'm aware that in a MSYS2/Mingw environment this is not an issue: All Tcl extensions will be built with __USE_MINGW_ANSI_STDIO=1. So I fully agree that you will want to make a different choice. I'll show a real-life example. Suppose that an extension does something like: <pre> __int64 value = foo(); Tcl_Panic("Value: %" TCL_LL_MODIFIER "d", value); </pre> Older Visual Studio versions don't have the "ll" specifier, so in order to be compatible with those versions TCL_LL_MODIFIER resolves to "I64". But the Mingw stdio doesn't recognise "I64" (unless it is recently added ...). The last reason is: Using __USE_MINGW_ANSI_STDIO=1 together with --static-libgcc brings in a lot of stuff, increasing the Tcl dll size. (this is not the main argument, the main argument is compatibility with Visual Studio) I hope you understand that therefore I have to reject this ticket :-( login: jan.nijtmans resolution: Rejected status: Closed ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 08:18:49
|
Automated mail by fx, on behalf of no...@tc... Commit [d603862aaecf08472a4761ce1695bdcfe611d5c9348585009d7f5b9fcf603507] By apnadkarni For Tcl (branch: apn-init-refactor) On 2025-10-27T08:13:03.343 Details https://core.tcl-lang.org/tcl/info/d603862aaecf08472a4761ce1695bdcfe611d5c9348585009d7f5b9fcf603507 Description Merge trunk Changed Files 84 edited .github/tcltest.json .github/workflows/linux-build.yml .github/workflows/mac-build.yml .github/workflows/onefiledist.yml .github/workflows/win-build.yml (truncated) ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 05:33:53
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [2107ccd96df220704940106fbe6b52d6b9ef531bedf2d5a817aa3cf6aa2a767d] [wish crash on exit (tka11y branch)] By apnadkarni For Tk On 2025-10-27T05:31:27.140 Details https://core.tcl-lang.org/tk/tinfo?name=2107ccd96df220704940106fbe6b52d6b9ef531bedf2d5a817aa3cf6aa2a767d Ticket https://core.tcl-lang.org/tk/tktview/6ff107d83f1dfd1d5c65e32ed032177f0fc0f98f Changed Fields closedate: 2025-10-27 05:31:27 closer: apnadkarni icomment: Verified as no longer crashing. Closing. login: apnadkarni resolution: Fixed status: Closed ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 01:18:35
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [feece6a82c9336fe1a8ff0d5a779de295b12a799802bd316377faa38e7dc6ace] [Windows printing hangs on select printer event] By kevin_walzer For Tk On 2025-10-27T01:03:24.101 Details https://core.tcl-lang.org/tk/tinfo?name=feece6a82c9336fe1a8ff0d5a779de295b12a799802bd316377faa38e7dc6ace Ticket https://core.tcl-lang.org/tk/tktview/09925826c797fbd07ef9d331ce79bc577f1b715b Changed Fields assignee: nobody closer: nobody comment: Text and canvas printing on Windows hangs after a printer select event. Unclear when it stopped working. foundin: 9.1 is_private: 0 login: kevin_walzer priority: 5 Medium private_contact: fd9290ae5f1a8509202a94b7afe132a13188e0cd resolution: None severity: Critical status: Open submitter: kevin_walzer subsystem: 68. Win Window Operations title: Windows printing hangs on select printer event type: Bug ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-27 00:33:38
|
Automated mail by fx, on behalf of no...@tc... Commit [0ff7977e1a99303848f0a9555ef46cd698bb83ada635843fd788c6895c23c9fb] By kevin_walzer For TIP (branch: main) On 2025-10-27T00:23:04.658 Details https://core.tcl-lang.org/tips/info/0ff7977e1a99303848f0a9555ef46cd698bb83ada635843fd788c6895c23c9fb Description Minor correction to TIP 733 Changed Files 1 edited tip/733.md ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-26 23:18:41
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [efd0695bc9ee8074fd86adbeecd4aac4dbf825b9366bd014a53bf0aaabf72a48] [Makefile forces -D__USE_MINGW_ANSI_STDIO=0] By oscarfv For Tk On 2025-10-26T23:06:13.897 Details https://core.tcl-lang.org/tk/tinfo?name=efd0695bc9ee8074fd86adbeecd4aac4dbf825b9366bd014a53bf0aaabf72a48 Ticket https://core.tcl-lang.org/tk/tktview/b405da2aad74715fdf8c729fdcfcc0a75d56b64b Changed Fields assignee: nobody closer: nobody comment: win/Makefile.in contains: CFLAGS = @CFLAGS@ @CFLAGS_DEFAULT@ -D__USE_MINGW_ANSI_STDIO=0 Why the -D__USE_MINGW_ANSI_STDIO=0 ? In MSYS2/Mingw-packages we set -D__USE_MINGW_ANSI_STDIO=1 and in this case tk's makefile is overriding it. If an application that uses __USE_MINGW_ANSI_STDIO=1 embeds tcl/tk, it *might* cause issues. Ditto for tcl/win/Makefile.in. is_private: 0 login: oscarfv priority: 5 Medium private_contact: becb0e4021beda640acc1b17b371abd9e9618647 resolution: None severity: Minor status: Open submitter: oscarfv subsystem: 85. Win Build title: Makefile forces -D__USE_MINGW_ANSI_STDIO=0 type: Support ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-26 22:48:46
|
Automated mail by fx, on behalf of no...@tc... By jan.nijtmans For Tk On 2025-10-26T22:47:48.280 Changed https://core.tcl-lang.org/tk/info/c060ba0048fc43ef82845e58134802199eeef8cfeb290b1005576cb1c6315d7e Tag +sym-core-bug-816739b7d1 ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-26 22:48:31
|
Automated mail by fx, on behalf of no...@tc... Commit [50a8805cb46fc939b27e83ddf422f83d937795d6afbb5f39a953b5f32154244d] By jan.nijtmans For Tk (branch: main) On 2025-10-26T22:46:29.566 Details https://core.tcl-lang.org/tk/info/50a8805cb46fc939b27e83ddf422f83d937795d6afbb5f39a953b5f32154244d Description Use more bool in internal API's Changed Files 8 edited generic/tkConfig.c generic/tkInt.h generic/tkPanedWindow.c generic/tkText.c generic/tkText.h (truncated) ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-26 22:33:56
|
Automated mail by fx, on behalf of no...@tc... Ticket Change [6d7add87da96cd9eb7ada13df7a332e194dc01e9716708b2b517c42afba867b3] [Mingw-64 has SEH] By oscarfv For Tk On 2025-10-26T22:28:22.485 Details https://core.tcl-lang.org/tk/tinfo?name=6d7add87da96cd9eb7ada13df7a332e194dc01e9716708b2b517c42afba867b3 Ticket https://core.tcl-lang.org/tk/tktview/7d6590d964a6f5c29edcaaac9d66dffdc261e40c Changed Fields closedate: 2025-10-26 22:28:22 closer: oscarfv icomment: After your comment I looked at how this patch is used on our builds. It turns out that this patch was dropped shortly after the creation of the mingw-w64-tk package but none of us maintainers noticed the fact and hence it was never deleted. In msys2/mingw-packages there is the bad practice of not documenting why patches are added or dropped. Then, many years after, this happens :-( I'm very sorry for wasting your time. Hopefully the other tickets will prove useful. My intent is to have a patch-free msys2/mingw-w64 tcl/tk 9 build. login: oscarfv resolution: Invalid status: Closed ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-26 22:33:47
|
Automated mail by fx, on behalf of no...@tc... Commit [335229fa9cfc4842c959780d6f4468e4d5a1af53fb514a7c62970946761b8db3] By jan.nijtmans For Tcl (branch: main) On 2025-10-26T22:31:25.459 Details https://core.tcl-lang.org/tcl/info/335229fa9cfc4842c959780d6f4468e4d5a1af53fb514a7c62970946761b8db3 Description Eliminate compiler warning (only seen in DEBUG) Changed Files 2 edited generic/regc_nfa.c generic/regcomp.c ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
|
From: <no...@tc...> - 2025-10-26 22:18:39
|
Automated mail by fx, on behalf of no...@tc... Commit [dfd477e3a499f3c21722b3c29d1053b5216c1f43fe1f96fb779841779608b876] By jan.nijtmans For ITcl (branch: main) On 2025-10-26T22:08:10.598 Details https://core.tcl-lang.org/itcl/info/dfd477e3a499f3c21722b3c29d1053b5216c1f43fe1f96fb779841779608b876 Description Add IWYU pragma annotations Changed Files 2 edited generic/itcl.h generic/itclInt.h ------------------------------------------------------------ See Tcl/Tk development @ http://core.tcl-lang.org/ ------------------------------------------------------------ |
286 messages has been excluded from this view by a project administrator.
