The second bug appears to be an assertion due to an invalid vector access.
The stack trace is below.
Truncated backtrace:
Thread no. 1 (10 frames)
#2 std::__replacement_assert at
/usr/include/c++/11/x86_64-redhat-linux/bits/c++config.h:2660
#3 std::vector, std::allocator > >::operator[]
at
All 4 CVEs were fixed in the 6.0.2 release and the release announcement was
updated last night to say this (to coincide with the public disclosure that
happened today). There will be another email on the developer list later
today with more details.
-Ian
On Wed, Feb 16, 2022 at 2:18 PM Steven A.
Hi Steve,
These warnings have already been addressed in the master branch because we
switched to C++17 (so those deprecated functions were removed and no longer
available). We didn't cherry-pick the change to v6 though, because it was
more of a cleanup than a bug fix (and v6 is staying with
Hi Steve,
On Tue, Jan 11, 2022 at 3:47 PM Steven A. Falco
wrote:
> I've opened https://gitlab.com/kicad/code/kicad/-/issues/10375
>
> I'll be adding more info to the issue as I do more testing. Right now,
> the issue just reports on visual artifacts that make accelerated graphics
> unusable on
Did you include the newlines in the translated text?
-Ian
On Mon, Nov 29, 2021 at 9:22 AM Marco Ciampa wrote:
> Hi devs,
> the message here:
>
> #: common/dialogs/dialog_global_lib_table_config.cpp:42
> #, c-format
> msgid ""
> "KiCad has been run for the first time using the new %s library
Please update your build. That version string is showing that your current
build is over 4 months old (it is from early June), and a loy of bugs have
been fixed since then.
-Ian
On Mon, 15 Nov 2021, 5:49 pm Brian, wrote:
> Sorry, I always forget to include that.
>
> Application: Pcbnew
>
So, for some reason GMail thinks this email is empty when I open it but
shows text in the inbox view, so I am going off of what I can see in the
"show original message" option.
For running with ASAN, you simply need to pass the CMake flag
KICAD_SANITIZE_ADDRESS and then I would suggest setting
Is it on a recent clone of the wxWidgets master branch? I thought this was
fixed upstream in July in https://github.com/wxWidgets/wxWidgets/pull/2436.
-Ian
On Thu, Sep 9, 2021 at 12:35 AM wrote:
> Hi Wayne,
>
> just did a git clone on the wxWdigets master.
>
> There does not seem to be an
On Mon, Aug 2, 2021 at 11:32 AM Carsten Schoenert
wrote:
> Hi Holger,
>
> Am 31.07.21 um 15:52 schrieb Holger Vogt:
> > Hello Carsten,
> >
> > there are two thing now open:
> >
> > I have made an update to ngspice in branch pre-master:
> > Now all boolean symbols transferred over the ABI are of
Steve,
I saw that failure last night also, and I think it may be a wxPython
problem with Python 3.10. I don't hav ea Rawhide VM available at the
moment, but what we should do is try the following:
1) Install Python 3.10 and python-wxpython4 in a Rawhide install
2) Run python -c "import
Depending on how the OpenSUSE build system si setup, a prebuilt version
from their repository might not work with a library with a different
version. Many build systems for Linux packages will encode a dependency on
a specific version of the shared library, which is included in the
filename. When
At this point, KiCad isn't defining a version requirement on ngspice, so
any way you choose to do it would work for us.
As for the issue compiling earlier, did you try to use your new ngspice
build with a pre-built KiCad version or did you build that KiCad version
yourself?
-Ian
On Tue, Jul 27,
Congrats Roberto!
On Tue, Jul 6, 2021 at 1:04 PM Wayne Stambaugh wrote:
> I am happy to announce that Roberto Fernandez Bautista has accepted an
> invitation to the KiCad lead development team. Roberto has made some
> significant contributions to the KiCad project and I look forward to his
>
I would prefer we keep this change to 5.99/6.0 only and not force a default
change in 5.1 builds. There are enough other dependency changes going on
for future v6 that packagers/users have to handle I think it is more
reasonable to force it there.
-Ian
On Tue, May 18, 2021 at 5:35 PM Eeli
Hi Clifford,
Welcome to the list! That sounds like a good suggestion, but we keep track
of feature requests in the issue database on GitLab
https://gitlab.com/kicad/code/kicad. Are you able to make a new issue
describing the feature you want added (the direct link to make a new issue
is
I have added support to build with the thread sanitizer on the master
branch (including annotating our coroutine implementation to track the
context switching). The thread sanitizer can be enabled using the flag
CMake KICAD_SANITIZE_THREADS.
As part of this change, I have changed the address
A point of clarification: this is only master branch (5.99) builds. None of
these changes will be in stable releases until v6 (so they aren't in 5.1).
-Ian
On Thu, Apr 15, 2021 at 5:38 PM Ian McInerney
wrote:
> FYI to all the packagers who work with the nightly packages on Linux, I
> hav
FYI to all the packagers who work with the nightly packages on Linux, I
have just merged in a set of changes overhauling the metadata naming and
contents (https://gitlab.com/kicad/code/kicad/-/merge_requests/742). These
changes make our files follow the standard file-naming system now
That is unrelated to the 3D models issue and is instead the software
rendering not working on macOS - so it can be fixed by simply switching to
the accelerated (OpenGL) canvas. We have an open ticket to remove the
software rendering from Pcbnew in 5.1 here
Hmm, I wonder if this is another instance of a pathing issue. We had a
similar bug on Linux that was caused by the 3d viewer looking for the
plugin to load OCE files in the wrong location, so it just wouldn't load
any files. That was https://gitlab.com/kicad/code/kicad/-/issues/7750,
which I have
The main reason we don't have CI on 5.1 is that I was being lazy and didn't
setup any and just focused on getting the master branch CI working since
that is what sees the most development.
-Ian
On Wed, Feb 17, 2021 at 3:49 PM Jon Evans wrote:
> We don't have CI on the 5.1 branch right now.
On Wed, Feb 3, 2021 at 7:34 AM Holger Vogt wrote:
> To obtain version information of the currently available libngspice-0.so
> version, you might do the following:
>
> load libngspice-0 dynamically
> initialze libngspice
> send the command 'version -s'
> retrieve some text response similar to
On Thu, Jan 21, 2021 at 5:25 PM Ian McInerney
wrote:
> Yes, Steve made me aware of the lack of that file on older distros - and I
> am working on a solution. I am currently building out the CMake files so
> that they try to do the translation and then check the return code and
> fallback
r that ? Perhaps there is already an answer
> > about this point ? I think I can fix that by coping missing ITS files on
> > the appropriate directory but it's a dirty solution...
> >
> >
> >
> >
> > Le 18/01/2021 à 18:54, Ian McInerney a écrit :
>
translations need to be updated to use the KICAD_BUILD_I18N=ON flag and no
longer reference the i18n repository.
-Ian
On Sat, Jan 16, 2021 at 7:41 PM Ian McInerney
wrote:
> Since we now host the v6 translations inside the main code repository, I
> have consolidated the CMake scripts for bu
Since we now host the v6 translations inside the main code repository, I
have consolidated the CMake scripts for building the translation files into
the main build process inside this MR (
https://gitlab.com/kicad/code/kicad/-/merge_requests/628). That exposes a
new CMake option
In order for the version string to contain the git version information, the
build must be done inside a git repository.
-Ian
On Sun, Nov 15, 2020 at 6:36 PM wrote:
> Thanks to Mark Roszko’s help I can build now with MSVC and KiCad runs ok.
>
> But then, the version info looks strange (see
On Wed, Oct 28, 2020 at 4:07 PM Brian Piccioni
wrote:
> I found it in Common and reassigned it to ctl-F12 so I won't use it by
> accident. I can't see how to clear or disable a hotkey.
>
If you right-click on the action in the hotkey list there should be an
option to clear the assigned hotkey
If all you are doing is touching the translations, we will be skipping the
CI checks. Just make the merge request with the update files and it will be
merged. I will look into changing the CI so that it doesn't run if only the
translation files are updated.
On Thu, Oct 22, 2020 at 10:43 AM Marco
On Wed, Oct 14, 2020 at 3:52 PM Jon Evans wrote:
> "Footprint Text" off turns off all footprint text, including references
> and values. This isn't new behavior but I worry that the simplified
> objects panel has made it more confusing than it was in 5.1
>
>
This only turns it off when viewing
t; glcanvasegl
>
> I have done a 'git pull' at the top level.
>
> Am I missing something?
>
> Thanks
> Tom
>
> On Mon, 12 Oct 2020, Ian McInerney wrote:
>
> > The OpenGL failing is because wxWidgets has defaulted to an EGL backend
> instead of GLX and I
gt; I built KiCad with 'cmake -DCMAKE_BUILD_TYPE=Debug' so can investigate
> with gdb if needed.
>
> Could you give me any tips on what might be wrong and where to look?
>
> Thanks again
>
> Tom Crane
>
> On Thu, 8 Oct 2020, Ian McInerney wrote:
>
> > The build has failed
The build has failed because it appears that your version of
wxPython/Phoenix is using wxWidgets 3.1.5 and you are trying to use
wxWidgets 3.1.4 in the main KiCad build. Those two versions must be the
same in order for KiCad to build properly (otherwise there will be issues
with linking).
The
Out of curiosity, which issues are preventing the 10.12-10.13 builds? Is it
the disappearance of OCE from homebrew?
-Ian
On Wed, Sep 30, 2020 at 2:47 PM Adam Wolf
wrote:
> macOS is uploaded:
>
> https://kicad-downloads.s3.cern.ch/osx/stable/kicad-unified-5.1.7-0-10_14.dmg
>
> I was unable to
On Wed, Sep 30, 2020 at 2:53 PM Simon Richter
wrote:
> Hi,
>
> On Wed, Sep 30, 2020 at 08:46:46AM -0400, Wayne Stambaugh wrote:
>
> > I'm fine with adding glew to the third party directory. I'm assuming
> > that the plan would be to use cmake to determine if EGL support was
> > required and
tprint library, we have 7 footprints that specify
>>> non-unity scaling. (Banana_Jack_[1-3], NS-Tech_Grove_1x04, Fuse_Blade_ATO,
>>> Fuse_Blad_Mini, Oscillator_SMD_TXC0_G158).
>>>
>>> -Seth
>>>
>>>
>>>
>>>
>>> On Tue, Sep
, Fuse_Blade_ATO,
>> Fuse_Blad_Mini, Oscillator_SMD_TXC0_G158).
>>
>> -Seth
>>
>>
>>
>>
>> On Tue, Sep 29, 2020 at 9:30 AM Ian McInerney
>> wrote:
>>
>>> We can't remove the scaling option until we make the VRML importer
>>&g
We can't remove the scaling option until we make the VRML importer handle
proper unit selection. I have routinely run into the case where I go
OpenSCAD -> Wings3D -> KiCad and design a model using mm in OpenSCAD
because it makes for easier computations (all the datasheet values are
nicely given in
e after my old
>> commits. May be something changed with new schematic format and now it
>> really does nothing. I'm away from pc and cannot check it now.
>> I hope it helps.
>>
>> 28 сентября 2020 г. 20:04:27 GMT+03:00, Ian McInerney <
>> ian.s.mciner...@
The upcoming wxWidgets 3.1.5 release has added a new backend supporting
Wayland/EGL to the wxGLCanvas that we use for displaying the OpenGL drawing
canvas. This backend appears to be the new default backend for wxGTK unless
it is explicitly disabled on the wxWidgets configure step (you
pass
What is the option to "Ignore other projects" supposed to be used for in
the back annotation system? Right now it appears that it isn't connected to
anything outside the UI because I am getting an unused variable warning on
it:
[1257/1983] Building CXX object
It is better to post these types of issue to the issue tracer rather than
the developer list so that they can be properly tracked and addressed.
-Ian
On Sat, Sep 12, 2020 at 12:00 PM BERTRAND Joël
wrote:
> Hello,
>
> I'm using kicad 5.99 (built from sources on a Linux
The 3d viewer does have access to the kiway, but I really think we need to
think about how this is done before we just go adding a cross-probing
interface between pcbnew and the 3d viewer. As our kiway is written
currently, adding the cross-probing will basically be adding a brand-new
messaging
I have added xvfb to the docker images used for CI (both the Fedora 31
image and Ubuntu 18.04 image).
-Ian
On Fri, Aug 28, 2020 at 8:36 PM Sylwester Kocjan wrote:
> Hi again,
>
>
> On 23/08/2020 02:12, Ian McInerney wrote:
>
> Are you wanting to add new QA tests that need a
Are you wanting to add new QA tests that need a window to the main unit
test suite? If so, we should not use a VNC server to run them we should use
xvfb to emulate an X server interface.
-Ian
On Sat, Aug 22, 2020 at 11:31 PM Sylwester Kocjan wrote:
> Hello,
>
> Can I ask for updating docker
My solution was not to use anything other than the main kicad launcher ;).
So I never used standalone pcbnew/eeschema/gerbview/... on OSX. I did
figure out what it would take to fix the kiface detection issue, and
started https://gitlab.com/kicad/code/kicad/-/merge_requests/82 to try to
make it so
Found the problem, it was an unintialized enum in the property system
(PCB_LAYER_T). I have just pushed a fix to master for it.
-Ian
On Tue, Aug 18, 2020 at 8:20 PM Ian McInerney
wrote:
> I saw those when I opened up a windows build yesterday, but I haven't had
> time
I saw those when I opened up a windows build yesterday, but I haven't had
time to dig into it (they appear to be windows only).
-Ian
On Tue, Aug 18, 2020 at 8:02 PM Jeff Young wrote:
> I don’t see them.
>
> Can you drop into a debugger and give us the stack trace for each one of
> them?
>
>
It will be `zlib`.
-Ian
On Mon, Aug 17, 2020 at 8:20 PM Roberto Fernández Bautista <
roberto.fer@gmail.com> wrote:
> Hi Seth,
>
> Thanks for heads up. Any ideas what the new dependency might be called in
> VCPKG? Or if not available, could you let me know which CMAKE variable
> should I set
On Tue, Aug 4, 2020 at 1:44 PM Mark Roszko wrote:
> Been working on Phoenix, I basically have something working to build,
> though it'll be awhile before I have anything to share.
>
>
> However, here's the fun part:
> wxPhoenix 4.1.0 only works with wxWidgets 3.1.4 and above
> wxPhoenix 4.0.7
> >>> On 8/3/20 2:37 PM, Wayne Stambaugh wrote:
> >>>> On 8/3/20 2:01 PM, Carsten Schoenert wrote:
> >>>>> Hello Ian,
> >>>>>
> >>>>> Am 03.08.20 um 19:39 schrieb Ian McInerney:
> >>>>>> I have
supported but any ideas what I
>>> could do to install lemon correctly so visual studio/ cmake can recognise
>>> it?
>>>
>>> Thanks
>>>
>>> Roberto
>>>
>>> On Sun, 2 Aug 2020 at 22:06, Ian McInerney
>>> wrote:
&g
hat I could
> do to install lemon correctly so visual studio/ cmake can recognise it?
>
> Thanks
>
> Roberto
>
> On Sun, 2 Aug 2020 at 22:06, Ian McInerney
> wrote:
>
>> Yes, I have a branch on my fork [1] called "im/lemon" that can be used.
>> It
t be found
-Ian
[1] https://gitlab.com/imcinerney/kicad
On Sun, Aug 2, 2020 at 10:01 PM Adam Wolf
wrote:
> Is there a branch packages can use to make sure their lemon integration is
> working?
>
> On Sun, Aug 2, 2020, 4:00 PM Ian McInerney
> wrote:
>
>> Two new build
Two new build-time dependencies are being added to the master branch for v6:
* lemon - The lemon parser generator
* GTK3 (linux only) - the GTK3 libraries (only GTK3, not GTK2 - that is not
supported anymore). This is technically also a runtime dependency, but we
also need GTK3 for wxWidgets, so
The builds will be flagged with dirty until I am able to merge the Lemon MR
that changes the build process to fix it.
-Ian
On Sun, 2 Aug 2020, 6:28 p.m. Mark Roszko, wrote:
> Nick has fixed the Windows nightlies, rejoice
>
> The builds are flagged dirty for what its worth.
>
> On Sat, Aug 1,
We'll have to figure out how to phrase the support requirements for this,
because we have committed to supporting Windows 8.1 until January 10, 2023
apparently - and 8.1 would have 32-bit versions.
-Ian
On Sun, Aug 2, 2020 at 4:58 AM Mark Roszko wrote:
> I am working on a MSVC build chain
No, it isn't a build error that is causing it. If you look at the status
page showing the pipeline results (
https://jenkins.simonrichter.eu/view/KiCad%20Status/job/windows-kicad-msys2-pipeline/),
you'll see the builds finish successfully.
Eeli, can you try downloading the most recent lite build
You need to upgrade to GCC 10.2 to get rid of them - it was a bug with GCC.
I think 10.2 has landed in the Fedora stable repos now.
-Ian
On Thu, Jul 30, 2020 at 9:10 PM Steven A. Falco
wrote:
> I'm running a build of the master branch on Fedora 32, and I'm seeing a
> lot of "note" messages
100. Is it further back that
> that? Given that mingw debug builds are painfully slow, it's going to
> take a while.
>
> On 7/22/2020 6:17 PM, Ian McInerney wrote:
> > Try going right before c0aa6965de7b83ca78dd5c4d700b50a2a03a34e4 first.
> > The errors that Brian showed at
t;>
>
> C:/msys64/mingw64/bin/../lib/gcc/x86_64-w64-mingw32/10.1.0/../../../../x86_64-w64-mingw32/bin/ld.exe:
> >>>>> CMakeFiles/pcbnew_kiface.dir/objects.a(pcbnew_wrap.cxx.obj): in
> >>>>> function `std::_Sp_counted_deleter SWIG_null_deleter,
> >
Ignore all of those notes being printed by the compiler about mismatched
struct/class definition. There is a bug in GCC 10.1 that isn't silencing
those properly, but that is fixed in GCC 10.2/GCC 11 (I think they are
hoping to release 10.2 this week).
The linker errors appear to be Python
- and they
did this in a patch release... that is not going to be fun to switch to.
-Ian
On Mon, Jul 20, 2020 at 11:08 PM Steven A. Falco
wrote:
> On 7/20/20 6:01 PM, Ian McInerney wrote:
> > You should be able to switch the macros:
> > %cmake -> %cmake
> > %ma
There should be no differences with where things get installed to - and if
there are then our "install" targets are incorrect and should be fixed
upstream. The changes to the cmake macros are simply for the build system.
-Ian
On Mon, Jul 20, 2020 at 11:04 PM Steven A. Falco
wrote:
> On 7/20/20
You should be able to switch the macros:
%cmake -> %cmake
%make_build -> %cmake_build
%make_install -> %cmake_install
Then the builds and install will automatically use the proper out of tree
directory. See the change proposal page for more details:
I have just merged the PR that drops GTK2 support for Linux. This means
that the GTK3 version of wxWidgets must be used on all Linux platforms
building and using master (e.g. future v6). This change does not affect
5.1. If you don't use Linux, then this change shouldn't affect you.
-Ian
The core features of KiCad do not rely on GTK, or any specific toolkit
other than wxWidgets currently. GTK is only required if you want to use
KiCad on Linux machines, and it will use the native (Win32 and Appkit)
frameworks on the other operating systems [1]. The core code does have a
dependency
orm to apply in case there are coordinates involved.
>
> I see no benefit into splitting that into two arguments.
>
> On Fri, Jul 10, 2020 at 4:27 PM Ian McInerney
> wrote:
> >
> > Doesn't the origin transform operate on the IU and not use the units
> directly (so th
Doesn't the origin transform operate on the IU and not use the units
directly (so the units are only needed by the actual menu text generation
function)? If they are separate, then we should pass them as two
individual arguments to the GetSelectMenuText function.
-Ian
On Fri, Jul 10, 2020 at
Currently, we support both GTK2 and GTK3 versions of wxWidgets. What I am
proposing here is that we drop the support of GTK2 from our codebase and
only officially support the GTK3 version of wxWidgets on Linux. This would
only apply for master (v6 and beyond), 5.1 would continue to have its same
_annotate.cpp",
> "file":
> "C:/msys64/home/kicad-compile-test2/kicad/eeschema/dialogs/dialog_annotate.cpp"
> }
>
>
>
>
> -Original Message-
> From: Ian McInerney
> To: pjmo...@csi.com
> Cc: ian.s.mciner...@ieee.org ;
or:54:9: note: replace the
> class-key with 'struct'
>54 | class error_code;
>
> There may be more, but these are the ones I see repeating over and over.
> I tried to do a full build a couple of days ago, and the the build crapped
> out part way through. I don't recall the speci
That would probably be because I enabled the warning for
"-Wmismatched-tags" on clang/GCC. This shouldn't be an error though, only a
warning. It warns about declaring things class/struct inconsistently (on
MSVC builds this can cause problems, so I enabled this warning to ensure we
don't have
Hi Adam,
Thanks for your work trying to test this on the new hardware! I have made a
tracker epic for all the fixes we need to make to our code/infrastructure
here: https://gitlab.com/groups/kicad/-/epics/5, so we can more easily keep
track of everything. Currently, it is just tracking a few
Yes, it will use the system file rename command to rename the temp saved
file. If that system command fails, then wx will automatically attempt to
copy the temp saved file instead, and that will preserve the temp saved
file if it fails (so basically, the temp saved file will only be removed if
the
I'm +1 on removing the backup files completely. I can see how they might
have been useful in the past given the save behavior, but I think the
recent changes to that have made them obsolete. I am also someone who puts
all my projects into a git repo, so I have always found them annoying and
never
In general, wishlist items should only be given a milestone if they are
either:
1) Actively being worked on by a developer
2) Currently on the plan for the release they are milestoned against (this
one doesn't need a developer assigned/working on them yet)
Other wishlist items don't get a
The pcb calculator is a kiface (the bitmap 2 component is not though). The
Kiface() symbol is defined in `pcb_calculator.cpp`, and that hasn't changed.
-Ian
On Tue, Jun 23, 2020 at 3:19 PM Jeff Young wrote:
> Did something change regarding how much of common is getting pulled into
>
Welcome.
Yes, if you have some ideas for features to implement, this list is the
best place to start the discussion.
-Ian
On Sat, 20 Jun 2020, 00:20 , wrote:
> Hi,
>
> I just joined this developer's mailing list. I have a feature I'd like to
> add and wanted to know if I should post my
ated.
>
> Then for net class and custom rules, we can just have the severity be a
> rule property, not an error code property.
>
> -Jon
>
> On Thu, Jun 11, 2020, 13:36 Ian McInerney
> wrote:
>
>> Another example I just thought of (not involving costs): differential
&
where the
design would fail as an error.
-Ian
On Thu, Jun 11, 2020 at 6:25 PM Ian McInerney
wrote:
> On Thu, Jun 11, 2020 at 1:13 PM Jeff Young wrote:
>> >
>> > Imagine that violating the micro-via min bumps me up a classification
>> but violating the through-via min
> On Thu, Jun 11, 2020 at 1:13 PM Jeff Young wrote:
> >
> > Imagine that violating the micro-via min bumps me up a classification
> but violating the through-via min drops me out of pooling. There’s a big
> cost difference between those two.
> >
>
This is why I think switching to the severities
Why not make the severity attached to the rule being used instead of the
error code type. I think being able to say "this rule is very important and
should be treated as an error" and "this rule is just a guideline and can
be treated as a warning" while they are both the same error code would make
?
-Ian
On Wed, Jun 10, 2020 at 1:03 PM Ian McInerney
wrote:
> Does anyone know what the sizes for the bitmaps in the cpp-other folder
> are? The bitmaps are:
> stroke_dash
> stroke_dashdot
> stroke_dot
> stroke_solid
> tune_di
Does anyone know what the sizes for the bitmaps in the cpp-other folder
are? The bitmaps are:
stroke_dash
stroke_dashdot
stroke_dot
stroke_solid
tune_diff_pair_length_legend
tune_diff_pair_skew_legend
tune_single_track_length_legend
I noticed that they are not being
There are currently 2 actions on the Eeschema tool framework for changing
the bus entry shape, but they don't have a hotkey assigned by default.
Additionally, bus entries are able to be mirrored and rotated as normal
(using the hotkeys for those actions). So I guess the question is, do we
need to
I am used to autocomplete tools only showing those that match the current
search query. That would also be consistent with the way our hotkey search
control matches when typed as well.
-Ian
On Sat, May 30, 2020 at 4:04 PM Eeli Kaikkonen
wrote:
> Definitely leaving only matching items and
Steve,
Thanks for the update to it, and it looks good, so I have set it to merge
to master when the CI build completes. I will then manually cherry-pick
this to the 5.1 branch.
I do wonder if there is a better way for us to structure this search so
that we don't have to touch the file every time
There are quite a few leaks in the code at present, and every now and then
I go through and try to plug some of them (but I don't find all of them,
that is for sure). My suggestion is that you can either propose a patch in
a merge request if you think you know where the memory should be freed and
Mikolaj,
Sorry for the delay in revisiting the merge request (it seems to have
slipped past me). I just took a look and it seems fine to me, so I just
merged it. Thanks for the contribution, and thanks for the ping to make
sure this got done.
-Ian
On Sun, May 24, 2020 at 3:46 PM Mikołaj Wielgus
kicad search paths and
> > environment variables.
> (from:
> https://gitlab.com/kicad/code/kicad/-/blob/5.1/CMakeLists.txt#L179)
>
> Should I open a ticket for this?
>
> > lør. 23. maj 2020 22.38 skrev Ian McInerney
> > :
> > > Have you tried redefining the en
Have you tried redefining the environment variables to point to the correct
system libraries? Specifically I believe the 4 you need are:
KICAD_TEMPLATE_DIR (the templates)
KICAD_SYMBOL_DIR (the eeschema symbols)
KISYSMOD (the modules)
KISYS3DMOD (the 3d models)
-Ian
On Sat, May 23, 2020 at
We should make sure that both are working though. It is common for people
to put the www in front of the URL.
-Ian
On Sat, May 23, 2020 at 3:31 PM jp charras wrote:
> Le 23/05/2020 à 15:26, Wayne Stambaugh a écrit :
> > I don't know if it was intentional (I'm guessing not) but the KiCad
> >
I think the question is, do the other repos actually need to do "release
candidates." We switched to the release candidate paradigm for the main
code repo because for the initial releases in the 5.1 series it seemed like
as soon as a new version was released a critical regression was found that
If you are going to be compiling the 5.1.6 release against Boost 1.73, you
will probably need to backport the patch given in
https://gitlab.com/kicad/code/kicad/-/merge_requests/207 to the branch. I
have already cherry-picked it to the 5.1 branch, so the head of both 5.1
and master should compile
FYI: I have just merged
https://gitlab.com/kicad/code/kicad/-/merge_requests/198 into 5.1 which
bumps the minimum CMake version required from 3.0 to 3.2 so that we can fix
https://gitlab.com/kicad/code/kicad/-/issues/4209. We discussed the version
bump at length inside
Considering that there are also changes to the library in the releases,
does it make sense to package the light releases for the versioned
(non-testing/nightly) releases. I would think that in this case users would
want the updated libraries.
-Ian
On Fri, May 15, 2020 at 12:33 PM Andrew Lutsenko
If you are using Fedora 32, then you shouldn't be using Python 2 (they
deprecated it in this release with limited exceptions...), and it
definitely doesn't have a Python2 version of wxPython.
-Ian
On Fri, May 15, 2020 at 11:27 AM Simon Richter
wrote:
> Hi,
>
> On 14.05.20 07:28, victor tejada
Wayne Stambaugh
wrote:
> It would make my life easier so I am all for that. I don't know if any
> of our package build scripts still pull from launchpad. That's why I
> asked.
>
> On 5/11/20 2:55 PM, Ian McInerney wrote:
> > I would suggest we switch to GitLab as the s
I would suggest we switch to GitLab as the single source of truth for the
code tarballs for this release.
On Mon, May 11, 2020 at 7:25 PM Wayne Stambaugh
wrote:
> Is there any reason that I need to continue uploading stable source
> archives to launchpad now that you can do this directly from
1 - 100 of 320 matches
Mail list logo
