Re: [Kicad-developers] License CC-BY-SA-4.0 used in KiCad

2019-04-01 Thread Aimylios

Am 31.03.19 um 20:59 schrieb Wayne Stambaugh:

I reverted this commit so this should resolve the Fedora build issue.


Thank you very much! The Fedora nightlies are indeed back to normal.

Am 31.03.19 um 18:59 schrieb Carsten Schoenert:
> agreed, I just wondering why this just happens now. The field of the
> metadata_license was added by Wayne on Oct 24 2018, that is not that
> long ago. And no one has raised their hands as I posted the patch that
> was removing that field. ;-)

This is not correct. The tag was there from the very beginning, Wayne
just changed its value from CC0-1.0 to CC-BY-SA-4.0 [1].
I'm not an AppStream expert myself, just noticed the packaging failure
and tracked it down to your patch. I guess the issue would have been
discovered anyways (sooner or later). Maybe the Fedora tooling is just a
bit stricter than that of other Linux distributions.

Best regards,
Marcus

[1]
https://git.launchpad.net/kicad/commit?id=4db550353e608ac363995cc65d7e9e2251c4ef48

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] License CC-BY-SA-4.0 used in KiCad

2019-03-30 Thread Aimylios

Hi Carsten,

your third patch breaks the Fedora nightly builds. This is the error we
get (see [1] for a full build log):

FAILED:
? tag-missing   :  is not present
Validation of files failed
error: Bad exit status from /var/tmp/rpm-tmp.3E6TT2 (%check)

The reason is that the metadata_license tag is mandatory and must not be
omitted. It indicates the license for the appdata.xml file itself.
Please take a look at the AppStream specification for details [2].

I vote for reverting commit e5de787 [3]. In my opinion, CC-BY-SA-4.0 is
a perfect choice for this kind of file. If the developers prefer a
different license, that is fine for me as well. But please add back the
tag, so that we can continue distributing nightly builds for Fedora.

Best regards,
Marcus

[1]
https://copr-be.cloud.fedoraproject.org/results/@kicad/kicad/fedora-30-x86_64/00875607-kicad/builder-live.log
[2]
https://www.freedesktop.org/software/appstream/docs/chap-Metadata.html#tag-metadata_license
[3]
https://git.launchpad.net/kicad/commit/?id=e5de787f881998ed5882132608b6c15d57334a3e

Am 27.03.19 um 17:36 schrieb Carsten Schoenert:

Hello Wayne,

Am 27.03.19 um 16:34 schrieb Wayne Stambaugh:

Hey Carsten,

I tried merging your patches and patch 1 does not merge cleanly.  Would
you please rebase patch 1 so I can get it merged?


rebased and re-added.


___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp




___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] GAL canvas strategy - testers needed!

2018-09-07 Thread Aimylios

Hi Jeff,

Am 06.09.2018 um 23:08 schrieb Jeff Young:


Supersampling (4x) is still slow on my 3-year-old Mac Powerbook.  But 
Supersampling (2x) renders quickly and is nearly impossible to tell apart on my 
Mac Retina display.  Does 2x not render well on your ThinkPad?


"Supersampling (2x)" is useable at higher zoom levels. At lower zoom 
levels, the thickness of lines is not displayed uniformly, which I find 
somewhat irritating. But I guess I'll just have to get used to that.



Am 07.09.2018 um 00:31 schrieb Jeff Young:

 >> I observed some minor issues (tested on Linux with GTK3):

- buses and graphic lines always have the same colour as wires

I just pushed a fix for this.


The color issue is now fixed, but the style of graphic lines is not 
displayed correctly.



- right mouse click doesn't always work (it seems to me that the context menu 
appears and almost immediately disappears in about 50% of cases)


Hmm… GTK-specific?  (I haven’t seen this at all on OSX.)


To test this, I've compiled your branch with both GTK2 (version 2.24) 
and GTK3 (version 3.22) and it is indeed GTK3-specific.
I thought the whole idea of this work was to make KiCad compatible to 
GTK3, but it seems to me that GTK2 is still the preferred choice.



- the "small crosshair" is rather big (80 pixels instead of 24 pixels)


This is the same size we’ve used in Pcbnew for quite some time.


That is correct, but Pcbnew (GAL) has an option to not always show the 
crosshair, which makes it less annoying. Could such an option be added 
to Eeschema as well? The checkbox is already there, but it has no effect.



- the crosshair is never centered at the position of the mouse cursor in Y 
direction (X direction is OK, see attached screenshot)


The cross-hair jumps to the grid; the mouse pointer does not.  Is this what 
you’re seeing, or is there an offset even when the pointer is on the grid?


Yes, there is always an offset in Y direction, independent of the 
position within the grid. The screenshot shows how close the mouse 
cursor can get to the center of the crosshair. If I move the mouse up 
even one pixel, the crosshair jumps to the next grid point.
But it turns out that this issue is GTK3-specific as well, my GTK2 build 
is not affected.



- adding bitmaps to the schematic does not work (the file selector appears 
again when trying to place the bitmap that was selected before)
- editing bitmaps (mirror, rotate, etc.) doesn't work as expected, the 
transformation is only applied after closing and reopening Eeschema


I’ll look into these next.  (BTW: I hooked up the zoom refresh tool to rebuild 
the graphics state, so you might try it on things like mirror, rotate, etc.)


Zooming doesn't help. But I was talking about the "Image Editor" 
(accessible via "e" hotkey") that has no effect until Eeschema is 
restarted. The usual hotkey actions for rotation, mirroring, etc. do 
work correctly.


Best regards,
Marcus

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] GAL canvas strategy - testers needed!

2018-09-06 Thread Aimylios

Am 06.09.2018 um 17:57 schrieb Tomasz Wlostowski:

On 06/09/18 17:37, Jeff Young wrote:

Hi Tom,

I have committed the completion of the block rotate and mirror fixes (including 
for LibEdit).


Hi Jeff,

Thanks, I played with eeschema for a short while and I really appreciate
all the work you have done!

Dear devs (non-devs too ;-) - whoever is interested in testing this
branch, please compile this and send the feedback here:

https://git.launchpad.net/~jeyjey/kicad/log/?h=tom-eeschema-gal-aug27

Cheers,
Tom


Hi Tom, hi Jeff,

I observed some minor issues (tested on Linux with GTK3):
- buses and graphic lines always have the same colour as wires
- symbol outlines/pins are not drawn correctly (see attached screenshot)
- zoom level is not limited (22 was a reasonable upper value for 
Eeschema in my opinion)
- right mouse click doesn't always work (it seems to me that the context 
menu appears and almost immediately disappears in about 50% of cases)

- the "small crosshair" is rather big (80 pixels instead of 24 pixels)
- the crosshair is never centered at the position of the mouse cursor in 
Y direction (X direction is OK, see attached screenshot)

- global and hierarchical labels only consist of text
- adding bitmaps to the schematic does not work (the file selector 
appears again when trying to place the bitmap that was selected before)
- editing bitmaps (mirror, rotate, etc.) doesn't work as expected, the 
transformation is only applied after closing and reopening Eeschema


Also, I need to set the graphics acceleration to "Supersampling (4x)" to 
get an acceptable display quality, which unfortunately causes Eeschema 
to be rather slow. But that may just be an issue with my old computer 
(Thinkpad T420s).


Other than that, the new implementation looks very promising!

Best regards,
Marcus
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Fix CMakeLists.txt

2018-07-27 Thread Aimylios

Hi Orson,

perfect, thank you very much!

Best regards,
Marcus

Am 27.07.2018 um 09:13 schrieb Maciej Sumiński:

Hi Marcus,

Actually my memory has let my down this time. I have just checked the
patch on msys2 and it worked as expected, so I recommitted the original
version. Sorry for the noise.

Cheers,
Orson

On 07/26/2018 11:36 PM, Aimylios wrote:

Hi Orson,

thank you for taking the time to review and merge my patches!


Thank you for the patches, I have just merged them. I have slightly
modified the last patch - I realize that you used the correct convention
for calling execute_process(), but IIRC it did not work with msys2.
Therefore I simply added the suggested ${wxWidgets_CONFIG_OPTIONS} to
the command line and preserved "sh -c".


Unfortunately, your modified version of my third patch does not work for
me. The reason is that wxWidgets_CONFIG_OPTIONS is of type LIST (not
STRING).

This is the command I use to configure KiCad:
cmake \
     -DKICAD_SCRIPTING_WXPYTHON=OFF \
     -DKICAD_SPICE=OFF \
     -DCMAKE_BUILD_TYPE=Release \
     -DwxWidgets_CONFIG_EXECUTABLE=/usr/bin/wx-config \
     -DwxWidgets_CONFIG_OPTIONS=--toolkit=gtk2 \
     .

With this, the expression:
sh -c "${wxWidgets_CONFIG_EXECUTABLE} ${wxWidgets_CONFIG_OPTIONS}
--query-toolkit"

evaluates to:
sh -c "/usr/bin/wx-config --toolkit=gtk2;--static=no --query-toolkit"

and the semicolon (which separates the elements of the list) terminates
the command line. Consequently, both --static=no and --query-toolkit are
dropped. Instead, you get an error message like "sh: --static=no:
command not found." and the build script incorrectly assumes that GTK2
is used, even if we added --toolkit=gtk3 before.

I don't have an msys2 build environment, but maybe someone who has can
try to check if your memory serves you right and the "correct
convention" does not work.
We might then have to find an alternative solution. But I am not a CMake
guru, so any help is appreciated.

Best regards,
Marcus


___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


Re: [Kicad-developers] Fix CMakeLists.txt

2018-07-26 Thread Aimylios

Hi Orson,

thank you for taking the time to review and merge my patches!


Thank you for the patches, I have just merged them. I have slightly
modified the last patch - I realize that you used the correct convention
for calling execute_process(), but IIRC it did not work with msys2.
Therefore I simply added the suggested ${wxWidgets_CONFIG_OPTIONS} to
the command line and preserved "sh -c".


Unfortunately, your modified version of my third patch does not work for 
me. The reason is that wxWidgets_CONFIG_OPTIONS is of type LIST (not 
STRING).


This is the command I use to configure KiCad:
cmake \
-DKICAD_SCRIPTING_WXPYTHON=OFF \
-DKICAD_SPICE=OFF \
-DCMAKE_BUILD_TYPE=Release \
-DwxWidgets_CONFIG_EXECUTABLE=/usr/bin/wx-config \
-DwxWidgets_CONFIG_OPTIONS=--toolkit=gtk2 \
.

With this, the expression:
sh -c "${wxWidgets_CONFIG_EXECUTABLE} ${wxWidgets_CONFIG_OPTIONS} 
--query-toolkit"


evaluates to:
sh -c "/usr/bin/wx-config --toolkit=gtk2;--static=no --query-toolkit"

and the semicolon (which separates the elements of the list) terminates 
the command line. Consequently, both --static=no and --query-toolkit are 
dropped. Instead, you get an error message like "sh: --static=no: 
command not found." and the build script incorrectly assumes that GTK2 
is used, even if we added --toolkit=gtk3 before.


I don't have an msys2 build environment, but maybe someone who has can 
try to check if your memory serves you right and the "correct 
convention" does not work.
We might then have to find an alternative solution. But I am not a CMake 
guru, so any help is appreciated.


Best regards,
Marcus

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp


[Kicad-developers] Fix CMakeLists.txt

2018-07-21 Thread Aimylios

Hi!

While trying to build KiCad for Fedora 28 I discovered some minor 
problems in the main CMakeLists.txt file. Patches attached.


Best regards,
Marcus
>From cbb85acdd95a6b4f91a8369f405a5ff841a424cf Mon Sep 17 00:00:00 2001
From: Marcus A 
Date: Sat, 21 Jul 2018 08:10:24 +0200
Subject: [PATCH 1/3] Update description of build options

Commit 93683d0021c4f592f929e9785abeda9d169ba03b changed the default build
options but did not update their description.
---
 CMakeLists.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index f2ea761f9..54f235a72 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -60,7 +60,7 @@ option( USE_WX_OVERLAY
 "Use wxOverlay: Always ON for MAC (default OFF).  Warning, this is experimental" )
 
 option( KICAD_SCRIPTING
-"Build the Python scripting support inside KiCad binaries (default OFF)."
+"Build the Python scripting support inside KiCad binaries (default ON)."
 ON )
 
 option( KICAD_SCRIPTING_MODULES
@@ -68,7 +68,7 @@ option( KICAD_SCRIPTING_MODULES
 ON )
 
 option( KICAD_SCRIPTING_WXPYTHON
-"Build wxPython implementation for wx interface building in Python and py.shell (default OFF)."
+"Build wxPython implementation for wx interface building in Python and py.shell (default ON)."
 ON )
 
 option( KICAD_SCRIPTING_ACTION_MENU
-- 
2.17.1

>From 14a749c7a2e2832416a5a5008b992c3f6568393a Mon Sep 17 00:00:00 2001
From: Marcus A 
Date: Sat, 21 Jul 2018 10:53:07 +0200
Subject: [PATCH 2/3] Fix some typos in CMakeLists.txt

---
 CMakeLists.txt | 14 +++---
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 54f235a72..b8cee3cdb 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -72,7 +72,7 @@ option( KICAD_SCRIPTING_WXPYTHON
 ON )
 
 option( KICAD_SCRIPTING_ACTION_MENU
-"Build a tools menu with registred python plugins: actions plugins (default OFF)."
+"Build a tools menu with registered python plugins: actions plugins (default OFF)."
 )
 
 option( KICAD_USE_OCE
@@ -84,14 +84,14 @@ option( KICAD_USE_OCC
 OFF )
 
 option( KICAD_INSTALL_DEMOS
-"Install kicad demos and examples (default ON)"
+"Install KiCad demos and examples (default ON)"
 ON )
 
 # when option KICAD_SCRIPTING OR KICAD_SCRIPTING_MODULES is enabled:
 # PYTHON_EXECUTABLE can be defined when invoking cmake
 # ( use -DPYTHON_EXECUTABLE=/python.exe or python2 )
 # when not defined by user, the default is python.exe under Windows and python2 for others
-# python binary file should be is exec path.
+# python binary file should be in exec path.
 
 # KICAD_SCRIPTING_MODULES requires KICAD_SCRIPTING enable it here if KICAD_SCRIPTING_MODULES is ON
 if ( KICAD_SCRIPTING_MODULES AND NOT KICAD_SCRIPTING )
@@ -107,7 +107,7 @@ endif()
 
 option( BUILD_GITHUB_PLUGIN "Build the GITHUB_PLUGIN for pcbnew." ON )
 
-option( KICAD_SPICE "Build Kicad with internal Spice simulator." ON )
+option( KICAD_SPICE "Build KiCad with internal Spice simulator." ON )
 
 # Global setting: exports are explicit
 set( CMAKE_CXX_VISIBILITY_PRESET "hidden" )
@@ -163,7 +163,7 @@ if( NOT DEFAULT_INSTALL_PATH )
  "Location of KiCad data files." )
 endif()
 
-message( STATUS "Kicad install dir: <${DEFAULT_INSTALL_PATH}>" )
+message( STATUS "KiCad install dir: <${DEFAULT_INSTALL_PATH}>" )
 
 # Generate build system specific header file.
 include( PerformFeatureChecks )
@@ -204,7 +204,7 @@ endif()
 # Unfortunately, the swig autogenerated files have a lot of shadowed variables
 # and -Wno-shadow does not exist.
 # Adding -Wshadow can be made only for .cpp files
-#and will be added later in CMakeLists.txt
+# and will be added later in CMakeLists.txt
 if( CMAKE_COMPILER_IS_GNUCXX OR CMAKE_CXX_COMPILER_ID MATCHES "Clang" )
 include(CheckCXXCompilerFlag)
 
@@ -281,7 +281,7 @@ if( CMAKE_COMPILER_IS_GNUCXX OR CMAKE_CXX_COMPILER_ID MATCHES "Clang" )
 set( CMAKE_EXE_LINKER_FLAGS_RELEASE "-s" )
 
 # _UNICODE definition seems needed under mingw/gcc 4.8
-# (Kicad uses unicode, and on Windows, wxWidgets >= 2.9.4 is mandatory
+# (KiCad uses unicode, and on Windows, wxWidgets >= 2.9.4 is mandatory
 # and uses unicode)
 if( GCC_VERSION VERSION_EQUAL 4.8.0 OR GCC_VERSION VERSION_GREATER 4.8.0 )
 add_definitions(-D_UNICODE)
-- 
2.17.1

>From 74db48d6c46f6469bf8d52f9bd77c7dfafe13d44 Mon Sep 17 00:00:00 2001
From: Marcus A 
Date: Sat, 21 Jul 2018 11:12:46 +0200
Subject: [PATCH 3/3] Fix detection of wxWidgets toolkit

wxWidgets_CONFIG_OPTIONS can be utilised to explicitly select GTK2 or GTK3. It
must be evaluated when querying the toolkit to avoid a mismatch between the
version actually used and the version KiCad assumes to be used.
---
 CMakeLists.txt | 6 --
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index b8cee3cdb..a41740b49 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -555,7 

Re: [Kicad-developers] Packaging question

2018-02-26 Thread Aimylios

Am 26.02.2018 um 19:42 schrieb Wayne Stambaugh:


Can someone please help Steve out here?  I don't know where our fedora
packages are being maintained so I would appreciate someone confirming
this patch.  I don't know that there is much we can do about the
upstream situation.

Thanks,

Wayne






Cheers,

Wayne

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp



Hi,

I'm not involved in Fedora upstream packaging, but like Steve I have 
spent some time to set up a packaging environment for my own, personal 
nightly and release builds, as I am not very happy with the current 
state of the official nightly copr builds.
I based my work on what is published in the fedora-packaging repository 
on GitHub. Unfortunately the SPEC file is in a poor shape and I ended up 
rewriting it almost completely. My private version of both the nightly 
[1] and release [2] packaging is available online.


I also set up an unofficial KiCad release copr (aimylios/kicad-release) 
[3]. The build of 5.0.0-rc1 is still in progress and it might take a 
couple of hours until the package is available.


One of the biggest changes I made is to move the libraries into separate 
packages (similar to what Carsten has done):

kicad
kicad-doc
kicad-templates
kicad-symbols
kicad-footprints
kicad-packages3D
Hopefully this or a similar scheme will be adopted by upstream or at 
least for the nightlies.


I would also like to help improving the nightly builds, but I'd prefer 
if this could be handled via pull requests over at the fedora-packaging 
repository [4]. No need to spam the developers mailing list with all 
this packaging stuff. In the end, we Fedora users are still a minority.


Keep up the great work!

Best regards,
Marcus

[1] https://github.com/aimylios/fedora-kicad-packaging/tree/develop
[2] https://github.com/aimylios/fedora-kicad-packaging/tree/release
[3] https://copr.fedorainfracloud.org/coprs/aimylios/kicad-release
[4] https://github.com/KiCad/fedora-packaging

___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help   : https://help.launchpad.net/ListHelp