-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
Am 18.02.2015 um 02:00 schrieb Lachlan Audas:
Hi guys, I'm running into a problem with your server(s), it's
really really slow, take's for-every to download the source, and
has timed out many time's just trying to download the boost lib's!
hint the user at the grouping feature?
Other than that, I believe the current state is useful already, so if you
feel it should be included, it can already be merged.
Simon
Simon Richter (1):
Add pin overview table
bitmaps_png/CMakeLists.txt | 1 +
bitmaps_png
This adds a new dialog giving an overview over the pins in the current
component. Currently only pin number, name and type are shown. More
functions, and editing support are planned.
---
bitmaps_png/CMakeLists.txt | 1 +
bitmaps_png/cpp_26/pin_table.cpp
For part with multiple subunits, the unit number is prepended to the pin
name. Common/shared pins are prefixed with com.
---
eeschema/dialogs/dialog_lib_edit_pin_table.cpp | 45 --
eeschema/dialogs/dialog_lib_edit_pin_table.h | 2 +-
eeschema/lib_pin.h
This adds a new dialog giving an overview over the pins in the current
component. Currently only pin number, name and type are shown. More
functions, and editing support are planned.
---
bitmaps_png/CMakeLists.txt | 1 +
bitmaps_png/cpp_26/pin_table.cpp
Hi,
On 21.03.2015 19:45, jp charras wrote:
Please, attach the patch file.
Usually, for many reasons, patches inside a mail are not very usable.
Sure, can do. I use git format-patch and git send-email usually, which
should make sure that nothing untoward happens to the patch before sending.
Hi,
On 21.03.2015 19:45, jp charras wrote:
To be usable, you have also (this is mandatory) to manage the unit
number (perhaps also the De Morgan representation). Otherwise the pin
table has no sense for multi parts per package components.
I've added handling for the unit number. For the NOR
Hi,
On 06.03.2015 06:31, Mark Roszko wrote:
I really don't agree with this patch with marking all the subclasses
as virtual just because if the base class parameters ever change for a
method the subclasses end up making entirely new functions. And you
won't get any errors because it'll just
Hi,
On 06.03.2015 13:13, Wayne Stambaugh wrote:
Please do *not* commit any patches for C++ 0x11. Using C++ 0x11 is not
an option at this point. We still have to support older compilers that
do not support 0x11. I do not want to abandoned a large segment of our
user base.
Yes, those
---
CMakeLists.txt | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 1d7b8f3..15ab933 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -155,7 +155,7 @@ if( CMAKE_COMPILER_IS_GNUCXX OR CMAKE_CXX_COMPILER_ID
MATCHES Clang )
#
---
common/fp_lib_table.cpp| 2 +-
common/project.cpp | 2 +-
eeschema/lib_export.cpp| 8
pcbnew/gpcb_plugin.cpp | 2 +-
pcbnew/legacy_plugin.cpp | 2 +-
pcbnew/router/pns_topology.cpp | 8
6 files changed, 12 insertions(+), 12
Hi,
On 06.03.2015 10:02, jp charras wrote:
I committed your patches 1, 2 and 3.
Excellent.
I am not sure patches 4 and 5 are very useful.
That's why I put them up for debate. #5 is more useful together with
override markers in C++11 mode.
Simon
signature.asc
Description: OpenPGP
---
pcbnew/router/pns_diff_pair.cpp | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/pcbnew/router/pns_diff_pair.cpp b/pcbnew/router/pns_diff_pair.cpp
index 2af6f3e..c08af7c 100644
--- a/pcbnew/router/pns_diff_pair.cpp
+++ b/pcbnew/router/pns_diff_pair.cpp
@@ -623,9 +623,8
Hi,
On 26.03.2015 11:21, Martijn Kuipers wrote:
If the program is expressly designed to accept user requests and send
responses over a network, then it meets these criteria. Common examples of
programs that would fall into this category include web and mail servers,
interactive
Hi,
Am 27.03.2015 um 13:04 schrieb jp charras:
- add position of pins
Noted.
- add sorting by pin type (mainly to check is all GND or VCC pins are
existing)
You can group by pin type by right clicking the column header, which
allows you to keep the sorting by pin number for a quicker
The GCC/LLVM specific handling for adding -fvisibility options is later in
this file, so these options would only be added on the second invocation
without clearing the cache.
---
CMakeLists.txt | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/CMakeLists.txt
Hi,
On 31.03.2015 13:29, Brian Sidebotham wrote:
delete m_model;
should really be:
delete m_model;
m_model = NULL;
Other functions (namely read_DEF) make assignment from m_model. It's
too complicated to try and rely on knowing the logical order of
functions in a file format
This moves most coordinates exactly in the middle of a pixel, allowing a
good mapping to pixels during rasterization, which gives a sharper image.
Thanks to Nick Østergaard.
---
bitmaps_png/cpp_26/pin_table.cpp | 35 ---
bitmaps_png/sources/pin_table.svg | 189
Thanks to John Beard
---
bitmaps_png/cpp_26/pin_table.cpp | 42 +++
bitmaps_png/sources/pin_table.svg | 258 --
2 files changed, 128 insertions(+), 172 deletions(-)
diff --git a/bitmaps_png/cpp_26/pin_table.cpp b/bitmaps_png/cpp_26/pin_table.cpp
index
---
eeschema/dialogs/dialog_lib_edit_pin_table.cpp | 28 +-
1 file changed, 23 insertions(+), 5 deletions(-)
diff --git a/eeschema/dialogs/dialog_lib_edit_pin_table.cpp b/eeschema/dialogs/dialog_lib_edit_pin_table.cpp
index c66594f..d72b447 100644
---
The GTK implementation occasionally queries the model even during
destruction of the internal representation, so notify the widget before
modifying the model.
---
eeschema/dialogs/dialog_lib_edit_pin_table.cpp | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Am 22.02.2015 um 17:40 schrieb Bernhard Stegmaier:
Wouldn’t it make more sense or be more consistent to have it in
$KISYSMOD where all models are?
No, the fp-lib-table introduces $KISYSMOD -- ideally there shouldn't
be any hard dependencies on
Hi,
On 06.03.2015 02:53, Mark Roszko wrote:
/home/mroszko/kicad/include/math/vector2d.h:412:54: warning:
‘*((void*)( m)+4).VECTOR2int::y’ may be used uninitialized in this
function [-Wmaybe-uninitialized]
return VECTOR2T ( x - aVector.x, y - aVector.y );
That is an interesting warning
This replaces a few tabs with four spaces. No code changes.
---
common/dialogs/dialog_page_settings.h | 2 +-
cvpcb/dialogs/dialog_config_equfiles.h | 2 +-
cvpcb/dialogs/fp_conflict_assignment_selector.h | 8
This function is only used for builds with -DDEBUG, so hide it in derived
classes as well.
---
include/worksheet_viewitem.h| 2 ++
pcbnew/ratsnest_viewitem.h | 2 ++
pcbnew/router/router_preview_item.h | 2 ++
pcbnew/tools/bright_box.h | 2 ++
pcbnew/tools/edit_points.h
is massive, and synchronizes the entire codebase to explicitly
virtual functions. It might be a bit boring to read though.
Simon Richter (5):
Indentation cleanup
Implement GetClass() throughout the project
Make Show() dependent on DEBUG preprocessor symbol
Use platform safe storage for pin
This function is really only used for diagnostics.
---
3d-viewer/3d_material.h | 8
3d-viewer/3d_struct.h | 8
include/base_struct.h | 5 +
include/worksheet_viewitem.h | 8
The uint32_t type has exactly 32 bits, while the definition of long is up
to the compiler (although it happens to be 32 bits almost everywhere).
---
eeschema/lib_pin.cpp | 4 ++--
eeschema/lib_pin.h | 7 ---
2 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/eeschema/lib_pin.cpp
Hi,
Am 23.03.2015 um 18:56 schrieb Chris Pavlina:
KiCad does not currently support standalone vias, and I think that would
be a necessity before a *clean* implementation of this could be done.
I think these vias would be placed inside a copper pour, so there is
something to connect them to.
This function takes an UR_* flag to denote the operation, not an item
status. As these constants have the same value, this doesn't change
generated code.
---
eeschema/dialogs/dialog_edit_libentry_fields_in_lib.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
This is really an abstract class and should not be instantiated directly.
---
include/base_struct.h | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/include/base_struct.h b/include/base_struct.h
index 360e4ff..5db45f4 100644
--- a/include/base_struct.h
+++
Hi,
I'm cleaning up a bit in the codebase, and found that the ERC silently
treats the ElectricPinType and TypeSheetLabel enums the same.
typedef enum { enum ElectricPinType {
NET_INPUT, PIN_INPUT,
NET_OUTPUT,
Hi,
On 22.04.2015 04:16, Ed Johns wrote:
I just re-implemented the calculations for pin offsets when inserting pins
in the library editor in eeschema. They are no longer dependent on global
settings. They are only dependent on the currently selected grid size and
the size of the text. The
Hi,
On 22.04.2015 08:25, jp charras wrote:
https://github.com/KiCad/kicad-library/blob/master/KiCad_Library_Convent
ion.txt
Automatic insertion of pins have to comply with that.
Good point. Would it make sense to leave this to the user still, but
somehow warn them (e.g. in the config
---
CMakeModules/FindOpenSSL.cmake | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/CMakeModules/FindOpenSSL.cmake b/CMakeModules/FindOpenSSL.cmake
index bbf431a..de91787 100644
--- a/CMakeModules/FindOpenSSL.cmake
+++ b/CMakeModules/FindOpenSSL.cmake
@@ -147,7 +147,7 @@
Hi,
On 17.05.2015 23:46, Mário Luzeiro wrote:
I know that this was a recent topic here in the mailing list, but so, what
are the plans about updating the boost library?
I was looking for the boost::geometry and it is very outdated with current
kicad version (1.54).
Outdated, yes. Problem
---
eeschema/dialogs/dialog_lib_edit_pin.cpp | 13 -
eeschema/dialogs/dialog_lib_edit_pin.h | 1 -
eeschema/pinedit.cpp | 2 --
eeschema/widgets/pin_type_combobox.cpp | 15 +++
4 files changed, 15 insertions(+), 16 deletions(-)
diff --git
---
eeschema/class_netlist_object.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/eeschema/class_netlist_object.h b/eeschema/class_netlist_object.h
index 0b51714..cd3ff48 100644
--- a/eeschema/class_netlist_object.h
+++ b/eeschema/class_netlist_object.h
@@ -154,7 +154,7 @@
This allows us to use compiler warnings for enum coverage in switch
statements.
---
eeschema/lib_pin.cpp | 6 +++---
eeschema/pin_type.h | 7 +--
2 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/eeschema/lib_pin.cpp b/eeschema/lib_pin.cpp
index a47d815..b974438 100644
---
---
eeschema/lib_pin.cpp | 28 +-
eeschema/lib_pin.h | 8 ---
eeschema/pin_type.cpp | 43 ++
eeschema/pin_type.h| 2 ++
eeschema/widgets/pin_type_combobox.cpp | 4
---
eeschema/CMakeLists.txt | 1 +
eeschema/erc.cpp| 8 +++---
eeschema/lib_pin.cpp| 27 ++
eeschema/lib_pin.h | 9 +-
eeschema/pin_type.cpp | 73 +
eeschema/pin_type.h | 5
6 files changed, 86
---
eeschema/class_netlist_object.cpp| 6 +++---
eeschema/class_netlist_object.h | 8 ++--
eeschema/dialogs/dialog_lib_edit_pin.cpp | 2 +-
eeschema/dialogs/dialog_lib_edit_pin.h | 6 +++---
eeschema/erc.cpp | 10 +-
eeschema/lib_pin.cpp
---
eeschema/CMakeLists.txt | 1 +
eeschema/lib_pin.cpp| 29 +++--
eeschema/pin_shape.cpp | 67 +
eeschema/pin_shape.h| 5
4 files changed, 76 insertions(+), 26 deletions(-)
create mode 100644 eeschema/pin_shape.cpp
---
eeschema/lib_pin.h| 21 +--
eeschema/pin_shape.h | 50 +++
eeschema/widgets/pin_shape_combobox.h | 2 +-
3 files changed, 52 insertions(+), 21 deletions(-)
create mode 100644 eeschema/pin_shape.h
diff --git
This removes the list generation, and simply iterates over the allowed
values.
---
eeschema/lib_pin.cpp| 11 ---
eeschema/lib_pin.h | 7 ---
eeschema/widgets/pin_shape_combobox.cpp | 11 +++
3 files changed, 7 insertions(+), 22
The bitmask requires a mapping to a different type used for the combobox,
and only few of the enumeration values actually make sense.
This removes the mapping, and uses a the same numbering throughout the
entire code.
---
eeschema/dialogs/dialog_lib_edit_pin.cpp | 2 +-
---
eeschema/dialogs/dialog_lib_edit_pin_base.cpp | 8
eeschema/dialogs/dialog_lib_edit_pin_base.h | 2 +-
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/eeschema/dialogs/dialog_lib_edit_pin_base.cpp b/eeschema/dialogs/dialog_lib_edit_pin_base.cpp
index 48b41f8..42d508a
file mode 100644
index 000..0c69bb2
--- /dev/null
+++ b/eeschema/widgets/pin_shape_combobox.cpp
@@ -0,0 +1,43 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2015 Simon Richter simon.rich...@hogyros.de
+ *
+ * This program is free
These overrides provide typed access to the current selection, overriding
the regular integer-based accessor/mutator.
---
eeschema/dialogs/dialog_lib_edit_pin.h | 2 +-
eeschema/widgets/pin_shape_combobox.cpp | 12
eeschema/widgets/pin_shape_combobox.h | 5 +
3 files
---
eeschema/lib_pin.h | 20 +---
eeschema/pin_type.h | 49 +
2 files changed, 50 insertions(+), 19 deletions(-)
create mode 100644 eeschema/pin_type.h
diff --git a/eeschema/lib_pin.h b/eeschema/lib_pin.h
index f4b21a8..bc84be6
---
eeschema/dialogs/dialog_lib_edit_pin_table.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/eeschema/dialogs/dialog_lib_edit_pin_table.cpp b/eeschema/dialogs/dialog_lib_edit_pin_table.cpp
index cfc3ee8..fdb5375 100644
--- a/eeschema/dialogs/dialog_lib_edit_pin_table.cpp
This makes the widget universally usable without special initialisation.
---
eeschema/dialogs/dialog_lib_edit_pin.cpp | 12
eeschema/dialogs/dialog_lib_edit_pin.h | 1 -
eeschema/pinedit.cpp | 1 -
eeschema/widgets/pin_shape_combobox.cpp | 12
4
These enums are silently treated as equivalent in the ERC, because the
values are casted to integers. This works for the most part, because the
values are similar, but this is in no way enforced, and there is a small
inconsistency (NET_UNSPECIFIED vs PIN_PASSIVE).
This turns the storage into a
.
Simon
Simon Richter (27):
Move feature check before dependent tests
Use Link Time Optimization with GCC in Release builds
Make NETLIST_OBJECT::GetConnectionType() const
Remove superfluous cast
Separate ElectricPinType and TypeSheetLabel
Regenerate Edit Pin dialog with newer
This significantly reduces executable file size, and should also improve
performance.
---
CMakeLists.txt | 15 +++
CMakeModules/PerformFeatureChecks.cmake | 1 +
2 files changed, 16 insertions(+)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index
---
eeschema/dialogs/dialog_lib_edit_pin_table.cpp | 50 +++---
1 file changed, 45 insertions(+), 5 deletions(-)
diff --git a/eeschema/dialogs/dialog_lib_edit_pin_table.cpp b/eeschema/dialogs/dialog_lib_edit_pin_table.cpp
index 28b4938..cfc3ee8 100644
---
---
eeschema/class_netlist_object.h | 2 +-
eeschema/dialogs/dialog_edit_label.cpp | 3 ++-
eeschema/dialogs/dialog_sch_edit_sheet_pin.h | 8 ++--
eeschema/edit_label.cpp | 8
eeschema/sch_sheet_pin.cpp | 2 +-
---
eeschema/dialogs/dialog_lib_edit_pin.h | 2 +-
eeschema/widgets/pin_type_combobox.cpp | 12
eeschema/widgets/pin_type_combobox.h | 5 +
3 files changed, 18 insertions(+), 1 deletion(-)
diff --git a/eeschema/dialogs/dialog_lib_edit_pin.h
---
eeschema/lib_pin.cpp | 11 ---
eeschema/lib_pin.h | 6 --
eeschema/widgets/pin_type_combobox.cpp | 5 +++--
3 files changed, 3 insertions(+), 19 deletions(-)
diff --git a/eeschema/lib_pin.cpp b/eeschema/lib_pin.cpp
index b0852ef..0783110
---
eeschema/dialogs/dialog_erc.cpp | 14 +++---
eeschema/dialogs/dialog_erc.h | 6 +++---
eeschema/erc.cpp| 6 +++---
eeschema/lib_pin.cpp| 6 +++---
eeschema/pin_type.h | 2 +-
5 files changed, 17 insertions(+), 17 deletions(-)
diff --git
Simon Richter simon.rich...@hogyros.de
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version
Hi,
On 08.04.2015 22:23, Wayne Stambaugh wrote:
32 and 64 bit builds are completely separate so configuring the correct
default install path should not be an issue.
I'm not sure that the path is the same when installing a 32 bit build on
64 bit Windows as when installing it on 32 bit Windows.
application.
+ *
+ * Copyright (C) 2015 Simon Richter simon.rich...@hogyros.de
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License
Hi,
On 04.04.2015 17:23, Lorenzo Marcantonio wrote:
On Sat, Apr 04, 2015 at 01:36:31AM +0200, Simon Richter wrote:
2. Keep strings/bitmaps outside of LIB_PIN
These are UI elements.
Also, AFAIK the bitmaps are only used in the pin dialog...
So far. I plan to add the same widget
Hi,
On 08.04.2015 14:38, Wayne Stambaugh wrote:
It will have to be fixed by modifying config.h.in so that you can
specify the DEFAULT_INSTALL_PATH when configuring the build instead of
always defining it as ${CMAKE_INSTALL_PREFIX}.
That is likely to break, because the default installation
Hi,
Am 21.06.2015 um 03:06 schrieb Cirilo Bernardo:
a. Abstracting a stream class which can magically support
file retrieval by http, local files, and perhaps github (I know
next to nothing about how github works) and local file
writing.
std::istream?
b. The stream class can automatically
Hi,
That is a legitimate bug. Not sure if it is on the tracker yet, but
it has been discussed on the list.
Can you create bugs on the bug tracker, so these don't get dropped again?
We definitely need extensive testing on MacOS. On reasonably modern
MacBooks, OpenGL is the only canvas with
Hi,
this fixes a bug in the pin table, where a break; between two case
statements is indeed incorrect, despite what Coverity claims.
Simon
From 2e0b3d621520478ea0baae9d19d0ef8d7627652a Mon Sep 17 00:00:00 2001
From: Simon Richter simon.rich...@hogyros.de
Date: Thu, 25 Jun 2015 00:22:11
Hi,
I've just tried to get context sensitive help using the F1 key, and
found that instead it zoomed in.
The menu claims that the hotkey for zooming in is Alt-F1, while the
hotkey list accessible with the '?' key lists F1 without qualifiers.
On Mac, Ctrl-'+' is used instead, which is fairly
Hi,
On 04.07.2015 02:29, Miguel Angel Ajo wrote:
bzr: ERROR: Invalid http response for
http://bazaar.launchpad.net/~kicad-product-committers/kicad/product/.bzr/branch-format:
Unable to handle http code 503: Service Temporarily Unavailable
ERROR: Failed to pull
Infrastructure issue, either
Hi,
On 30.06.2015 01:20, Chris Pavlina wrote:
I'd happily implement the net tagging, though it'd require a bit of
brainstorming from everyone on how those tags should be stored and
shared with pcbnew.
I think it belongs in the netlist.
Ideally, I'd like this as attributes on the component
---
include/tool/tool_base.h| 4 +++-
include/tool/tool_interactive.h | 2 +-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/tool/tool_base.h b/include/tool/tool_base.h
index 584a979..caec3b7 100644
--- a/include/tool/tool_base.h
+++ b/include/tool/tool_base.h
@@
---
pcbnew/autorouter/autorout.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/pcbnew/autorouter/autorout.h b/pcbnew/autorouter/autorout.h
index 348191f..b863039 100644
--- a/pcbnew/autorouter/autorout.h
+++ b/pcbnew/autorouter/autorout.h
@@ -73,7 +73,7 @@ extern int
---
pcbnew/exporters/export_idf.cpp | 2 +-
utils/idftools/idf_common.cpp | 4 ++--
utils/idftools/idf_common.h | 13 +
utils/idftools/idf_parser.cpp | 2 +-
4 files changed, 5 insertions(+), 16 deletions(-)
diff --git a/pcbnew/exporters/export_idf.cpp
.
Simon
Simon Richter (4):
Avoid bringing own definition of M_PI (MSVC warning C4005)
Clean up missing DLL export (MSVC warning C4275)
Clean up truncation of constants (MSVC warning C4309)
Clean up ambiguous default ctor (MSVC warning C4520)
gerbview/class_am_param.h | 2 +-
include
---
gerbview/class_am_param.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gerbview/class_am_param.h b/gerbview/class_am_param.h
index deeb93e..3d95b7c 100644
--- a/gerbview/class_am_param.h
+++ b/gerbview/class_am_param.h
@@ -162,7 +162,7 @@ public:
m_dvalue =
construction that the object is indeed complete.
Simon
commit 8d88ec9ff1e401c15bb1af662a0fe0b7475b2401
Author: Simon Richter simon.rich...@hogyros.de
Date: Mon Jul 6 00:53:05 2015 +0200
pin table: do not set vscroll by default to avoid crash on MacOS
diff --git a/eeschema/dialogs
Hi,
On 04.07.2015 12:53, Camille 019 wrote:
-if( layer == NO_AVAILABLE_LAYERS )
+if( layer == static_castint(NO_AVAILABLE_LAYERS) )
Nope, that relies on undefined overflow semantics, and compilers that do
constant propagation through casts would miscompile this code.
Hi,
On 23.05.2015 20:18, jp charras wrote:
Sorry to say that, but what is the interest of this work:
Now that is a good question. My initial goal was to basically shave a
few percents off in text size and get a bit of extra speed, just by
setting a few compiler flags.
I'm happy to put this on
Hi,
On 23.05.2015 18:48, Bernhard Stegmaier wrote:
I guess this is due to LTO not even being used on OS X, if I understand your
cmake change correctly:
find_program(CMAKE_GCC_AR NAMES
${_CMAKE_TOOLCHAIN_PREFIX}gcc-ar${_CMAKE_TOOLCHAIN_SUFFIX} HINTS
${_CMAKE_TOOLCHAIN_LOCATION})
Hi,
I've submitted a branch for inclusion at
https://code.launchpad.net/~sjr/kicad/lto
that seeks to use link-time optimization when possible.
We've already established that this is broken on MSYS, so I'm going to
disable it there, while it does some good on Linux. This leaves MacOS.
Could
Hi,
Will that speed up the building process also?
It might, a tiny bit, but in my experience it was mostly neutral, as the
savings from the shorter compiler invocations were eaten up because now
the linker took a lot more time.
Simon
signature.asc
Description: OpenPGP digital signature
Hi,
I know it is not official supported, but I believe I read here that
someone is able to build it, is it stable? could that person (persons)
help try build that on VS?
Yes, it works. You need a few patches, I keep these in the
lp:~sjr/kicad/msvc branch on Launchpad, and merge them in the
Hi,
Am 01.08.2015 um 20:40 schrieb Wayne Stambaugh:
Does anyone have any major objections to this
solution. If not, I will commit this patch.
On the contrary, I'd like to see this happen.
Simon
signature.asc
Description: OpenPGP digital signature
Hi,
On 30.06.2015 19:29, Andy Peters wrote:
Ideally, I'd like this as attributes on the component pins, and then
either inherited directly or at least verified by the ERC.
One argument against making this a pin attribute rather than a net attribute:
FPGA pins can be single-ended or
Hi,
On 30.06.2015 20:58, Andy Peters wrote:
Some micro vendors offer peripheral configuration tools which set up the chip
per user needs. If that configuration could be imported, then the user
doesn’t really need to do any work at all.
For FPGAs, I wanted to go the route of exporting the
Hi,
On 29.07.2015 21:00, jp charras wrote:
I removed a few conditional compilations in rev 6014.
This bug already known, but not fixed, should be fixed in this rev.
There is an older patch from Blair Bonnett removing all support for
wxWidgets 3.0. I have that in my repository, and it got
Hi,
three small patches for include files cleanup, each of them pretty
self-explanatory.
Simon
Simon Richter (3):
Remove superfluous includes
Include iso646.h for not, and and or keywords
add missing C++ stdlib headers
common/gal/graphics_abstraction_layer.cpp | 2 ++
common
These are unnecessary:
- in math_util.cpp, including stdint.h is unnecessary because math_util.h
already requires it
- in the other two headers, no definitions from stdint.h are used.
---
common/math/math_util.cpp | 1 -
include/layers_id_colors_and_visibility.h | 1 -
These source files use definitions from the standard library without making
sure they are included. It happens to work on current gcc because parts of
the standard library implicitly include these other headers, but that is
not guaranteed.
---
common/gal/graphics_abstraction_layer.cpp | 2
These are not actually keywords in the C++ standard, but macros. Unless
iso646.h is included, these are not guaranteed to be present.
---
include/common.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/common.h b/include/common.h
index b881444..8be1e25 100644
---
This macro is defined in stddef.h, so cstddef is needed.
---
include/gal/opengl/vertex_common.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/gal/opengl/vertex_common.h b/include/gal/opengl/vertex_common.h
index 9ec8d9b..0d8eb9d 100644
--- a/include/gal/opengl/vertex_common.h
+++
fabs(int) is not defined in the standard, so this is ambiguous. Moving the
cast inside the argument clarifies which overload we want, and gives the
correct return type.
---
common/geometry/shape_poly_set.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Hi,
this repairs const correctness for some STL container accesses:
- The result of .begin() and .end() on a const reference is a const_iterator.
- The .erase() function requires a mutable iterator.
The GNU C++ stdlib actually uses the same type here, because set members
may not be manipulated
Hi,
that function returns a bool, but common/geometry/shape_poly_set.cpp
line 704-705 has
if( !d )
return -1;
That smells fishy and gives a compiler warning.
Simon
signature.asc
Description: OpenPGP digital signature
Hi,
the MSVC build complains:
pcbnew\router\pns_tool_base.h(37) : warning C4275: non dll-interface
class 'TOOL_INTERACTIVE' used as base for dll-interface class
'PNS_TOOL_BASE'
include\tool/tool_interactive.h(35) : see declaration of 'TOOL_INTERACTIVE'
pcbnew\router\pns_tool_base.h(36) : see
Hi,
On 28.07.2015 06:32, Mark Roszko wrote:
What? Those keywords are part of the C++ standard from 2003(at least).
Its only MSVC that is backwards and broken in this regard.
Ah okay, then I misremembered that. Well, I've only used them once in
'99, and back then I needed the header.
Simon
---
common/displlst.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/common/displlst.cpp b/common/displlst.cpp
index 77631fc..afea210 100644
--- a/common/displlst.cpp
+++ b/common/displlst.cpp
@@ -236,7 +236,7 @@ void EDA_LIST_DIALOG::onClose( wxCloseEvent event )
/*
Hi,
The sorting function signature is supposed to use wxIntPtr, which is
guaranteed to be large enough to hold a pointer. The Windows ABI defines
long to be 32 bits, even on 64 bit architectures.
I have no idea how that ever compiled.
Simon
---
common/displlst.cpp | 2 +-
1 file changed, 1
Hi,
The way I see it implemented is to have an extra icon on the KiCad main
window next to the 'PI editor' for the library tool. (The images below
are just a mock-up, surely will follow the KiCad design once moving to
the real thing)
We have already some work on the way here -- in the
1 - 100 of 898 matches
Mail list logo
