* Create new GEDA PCB plug in.
* Add support for opening GEDA PCB footprints with module editor.
* Make import footprint file dialog remember last selected footprint
type during current session.
* Update module editor file import to use new GEDA PCB plug in.
* Let IO_MGR know about GEDA PCB plug in.
* Create a WHITESPACE_FILTER_READER to simplify parsing GEDA PCB footprint
files.
Because now there are 2 formats, board backup files extensions are .brd-bak or .kicad_pcb-bak instead of the old .000 extension, depending on the initial file format.
their type, by using the more abstract PLUGIN::Footprint*() functions.
This is an intermediate phase only, other changes will be necessary as library table support comes in.
Encapsulate usage of library path searching, since that will go away as library table support comes in.
Add FOOTPRINT_EDIT_FRAME::{get,set}LibPath() and FOOTPRINT_EDIT_FRAME::{get,set}LibNickName() functions
to provide this encapsulation.
some functions used in wxWidgets ( like _strdup and few other) are not defined when this option is set.
but the equivalent option -std=gnu++0x works fine so it is used under mingw instead of -std=c++0x.
(Under Linux/OSX the option -std=c++0x is used)
This required the additional compiler command line option "-std=c++0x".
2) Switch to unique_ptr from auto_ptr which is C++ deprecated.
3) Change to new English layer names per mailing list discussion, see class_board.cpp.
4) When saving to *.kicad_pcb or *.kicad_mod, identify opportunities to use wildcard
layer sets, for pads so far.
5) Switch to two std::string hashtables in pcb_parser.cpp, away from one wxString based one
for layer names and layer sets mapping. One hashtable holds the mask, the other the index.
Layer sets are only in the mask table.
6) Move "LOCALE_IO toggle" into PCB_IO::Format() since it is a public API function and
caller should find it as convenient as possible to use. LOCALE_IO should handle
nesting OK in the case where public Format() is called from one of the Footprint*()
functions.
7) Drop support for "export module to new library". Creating new libraries will have to
be handled in concert with library table entries, and we've talked recently about
creating a better footprint library manager, one that resides in a DLL/DSO. So
this kind of functionality needs to be handled in there in the near future.
8) Change name of exported PCB_IO module/footprint to *.kicad_mod and not *.emp.
* If this param is set to 0, the previous behavior is used.
* Else the solder mask is plotted as polygon, with shapes near than this min width value are merged.
Pcbnew in nanometer version: fix a serious rounding issue due to the fact dialogs were using 4 digits for mantissa.
* This is not enough for nanometer coordinates ( 8 are needed to display a value in inches),
* so sometimes after closing a dialog, some coordinates were modified, although no new value was entered.
Minor other bug fixed.
* Add footprint methods to PCB_IO.
* Add FP_CACHE and FP_CACHE_ITEM for handling new footprint library design.
* Add code to save legacy libraries in new format.
* Change behavior of BOARD_ITEM::GetLayerName() to return the default layer
name when the item does not have a BOARD as a parent.
* Minor changes to the module output formatter when writing to module library
files (no BOARD as parent).
* Add new (and some that I forgot along the way) CMake flags to
EDA_BASE_FRAME::CopyVersionInfoToClipboard().
* Add -Wno-narrowing to GCC flags to stop GCC 4.7 from complaining about
a conversion from int to unsigned in the Boost polygon library.
* Add INPUT_STREAM_READER to richio.cpp to allow using any object derived
from wxInputStream as a LINE_READER (needs to be validated).
clean duplicate code.
Simplify HPGL dialog: plot offet removed and replace by only one option (plot origin to center of sheet or not)
This is surely enought.
Mainly to plot drill maps, but can be used to plot boards, for documentation.
The print svg still exists, but the plot SVG has more options (mirroring, holes in pads),
however print svg allows color print, and full board printing, and plot does not.
and therefore tracks are now dragged when a end point is inside a pad, not necessary on the pad position.
However, drag functions still need more cleanup.
Kicad manager: shows now svg files in tree files.
Pcbnew: fix incorrect printing (board out of page) when creating drill map in postscript format ( noticeable mainly for small boards)
* serious code cleanup (remove duplicate code)
* shows (option in 3D preference menu) the copper items (tracks, zones...) in 3D, using 35 microns copper thickness.
However, because there are a lot more3D data to show (roughly 4 times more), this is slower.
connection, no matter if it is originating locally or from the outside
of the host. There is no check in place to discard non-local
connections. The only security against a malicious attack would be
provided by a local firewall, which is not guaranteed to be installed on
every workstation kicad is used on.
I tested this, and a host running eeschema accepts connections on TCP
port 4243 from other hosts on the internet.
A patch to remedy this potentially serious security hole is attached. It
creates the listener on localhost instead. A flag is provided to allow
the creation of sockets on 0.0.0.0 instead, if required. localhost is
the default.
All: press ctrl+shift key when moving the mouse allows the graphic cursor to be moved outside the grid.
(useful to place graphic objects, texts in any position, regardless the current grid)
This is *a work in progress*, so some features are missing, and/or could be modified.
Mainly keepout zones are not yet exported to autorouters, and pads are not taken in account.
Some code cleanup in polygon.*
* Add s-expression file parser object and keyword files.
* Fix minor issues with s-expression file formatting.
* Fix a minor bug the zone container fill state parsing in the legacy plugin.
* Move EDA_TEXT visibility definition to eda_text.h.
* Add minor BOARD_ITEM object improvements to support s-expression file
parser.
This may have been an undocumented bug from more than a year ago, which manifested itself in *.mod files containing duplicate names.
LEGACY_PLUGIN loads those now quietly, but appends "_v2", "_v3", etc. to each succeeding duplicate sharing the same name.
base on expected text and current window font.
Expand the virtual world to 2.14 meters in the nanometer build of PCBNEW.
This seems to be holding up for now.
Delete LEGACY_PLUGIN cache after deleting library of same name.
Remove automatically generted freerouter help header.
Add Normalize() to BOX2.
Clamp some values in the virtual space of drawframe.cpp
Should allow Pcbnew code easier to change and Gerbview code more understandable and easier to maintain.
Code cleaning (remove dead code, add comments).
Minor other enhancements.
Other scaling factors (MILS_TO_IU_SCALING_FACTOR and DECIMILS_TO_IU_SCALING_FACTOR)
also defined only in convert_to_biu.h.
Allows different scaling value for Gerbview.
Needs more tests.
solution to see that it was not pre-computing constants when used in static initializers.
So we have two use cases, and need two solutions to be optimal, since the compiler
is too stupid to do the right thing.
I think we need something else for compile time computable constants, to be optimal
in both use cases. There is quite a bit of code savings by using a macro for that
situation from my testing. To fully capitalize on this, we need to go back and make
Mm2mils() and what not macros also, or have MACRO versions of them too.
as a starting point
* Make double Distance() take double arguments and remove internal range checking.
* Start on EDA_DRAW_FRAME::AdjustScrollBars() and use "double" for most all calculations
in anticipation of setting INT_MAX INT_MIN limits eventually.
* Move board item object Format() functions into PCB_IO object.
* Change file format to use layer names instead of numbers.
* Change file extension to kicad_pcb.
// This provides better project control over rounding to int from double
// than wxRound() did. This scheme provides better logging in Debug builds
// and it provides for compile time calculation of constants.
#include <stdio.h>
#include <assert.h>
#include <limits.h>
//-----<KiROUND KIT>------------------------------------------------------------
/**
* KiROUND
* rounds a floating point number to an int using
* "round halfway cases away from zero".
* In Debug build an assert fires if will not fit into an int.
*/
#if defined( DEBUG )
// DEBUG: a macro to capture line and file, then calls this inline
static inline int KiRound( double v, int line, const char* filename )
{
v = v < 0 ? v - 0.5 : v + 0.5;
if( v > INT_MAX + 0.5 )
{
printf( "%s: in file %s on line %d, val: %.16g too ' > 0 ' for int\n", __FUNCTION__, filename, line, v );
}
else if( v < INT_MIN - 0.5 )
{
printf( "%s: in file %s on line %d, val: %.16g too ' < 0 ' for int\n", __FUNCTION__, filename, line, v );
}
return int( v );
}
#define KiROUND( v ) KiRound( v, __LINE__, __FILE__ )
#else
// RELEASE: a macro so compile can pre-compute constants.
#define KiROUND( v ) int( (v) < 0 ? (v) - 0.5 : (v) + 0.5 )
#endif
//-----</KiROUND KIT>-----------------------------------------------------------
// Only a macro is compile time calculated, an inline function causes a static constructor
// in a situation like this.
// Therefore the Release build is best done with a MACRO not an inline function.
int Computed = KiROUND( 14.3 * 8 );
int main( int argc, char** argv )
{
for( double d = double(INT_MAX)-1; d < double(INT_MAX)+8; d += 2.0 )
{
int i = KiROUND( d );
printf( "t: %d %.16g\n", i, d );
}
return 0;
}
*) for footprint access into *.mod files and
*) BOARD save/load
The item_io.cpp and ioascii.cpp have been set off to the side for reference
as *.notused, for awhile.
The CMake options USE_NEW_PCBNEW_LOAD and USE_NEW_PCBNEW_SAVE are gone,
this is now the mandatory usage of the LEGACY_PLUGIN. This should reduce
code maintenance for awhile until the s-expression plugin comes into
play. But at least for legacy format, there is not two code bodies
to maintain any more.
A new LEGACY_PLUGIN footprint library caching scheme is in place which
needs some testing. It should not be any faster, but might give better
results in a networked environment if there is *.mod files on the server.
* Remove internal units from BASE_SCREEN and it's derivatives.
* Remove internal units from EDA_DRAW_FRAME and it's derivatives.
* Use build time code to replace internal units conversions.
* Fix scaling bug in page layout sample window that I created in my
last commit.
* Move all convert from user to internal units into base_units.cpp.
* Remove internal units parameters from all moved conversion functions.
* Revise all source code that calls the moved conversion functions.
* Remove internal units from all dialog text control helper classes.
* Move all convert from internal to user units functions into separate file.
* Remove internal units parameter from all moved conversion functions.
* Revise all source code that calls the moved conversion functions.
* Compile these conversion routines separately for the appropriate pcb or
schematic internal units.
* Move internal units specific status bar update code into the appropriate
application for updating the status bar.
* Move millimeter user units rounding function to common.cpp.
* Move EDA_TEXT object into separate header and source file.
* Compile EDA_TEXT class separately for BOARD_ITEM and SCH_ITEM units.
* Compile PAGE_INFO class separately for BOARD_ITEM and SCH_ITEM units.
* Minor formatting tweaks to Pcbnew s-expression file.
* Move internal unit formatting functions into BOARD_ITEM and SCH_ITEM.
* Save dialog now supports saving boards to new file format.
* Add CMake option to build s-expression file save.
* Add check to main CMakeList.txt file to make sure nanometers are
enables when the new file format is built.
* Minor tweaks to object format functions for improved output.
* Rename kicad_plugin.h/cpp to legacy_plugin.h/cpp.
* Add s-expression Format() function to all objects derived from
BOARD_ITEM.
* Add s-expression Format() function to base objects as required.
* Add functions to convert coordinates from base internal units
(nanometers) to millimeter string for writing to s-expression
file.
* Add temporary dummy conversion functions to prevent link errors
until schematic and board object and action code can be separated
into DSO/DLL.
* Add CMake build option to build Pcbnew with nanometer internal
units.
* BLOCK_SELECTOR class is not longer derived from EDA_ITEM.
* Encapsulate BLOCK_SELECTOR class member variables and add access methods.
* Move HandleBlockBegin() function from block_commande.cpp to drawframe.cpp.
* Remove virtual methods from top level derived objects per future
coding policy change.
* Remove Doxygen copydoc statement from objects derived from EDA_ITEM
since the comments are automatically copied to the derived object.
* Removed copy and pasted Doxygen comments from objects derived from
EDA_ITEM.
issue in cast to wxRect fixed (EDA_RECT accepts negative size, but not wxRect, so the wxRectcast is normalized now).
This fixes some bugs.
A minor other bug fix in Pcbnew.
This works, but in my version of wxformbuilder there is a bug which does not properly show the
subclass property, even though it is still in play. This happens after saving then loading the *.fbp file.
So it is a nuisance bug, but does not affect functionality.
* All objects derived from EDA_ITEM now have consistent hit test method
definitions.
* Remove double function calls from all classes derived from SCH_ITEM.
* Lots of Doxygen comment fixes.
* make 2 bitmaps more visible.
* fix incorrect screen redraw when rotating a component.
* netlist : the pcbnew advanced format can be now stored (persistant option) in project config.
Pcbnew:
* graphic_item_properties_base.fbp * dialog pad edit: make some strings more easier to translate.
* global pad edition: fix a bug that change pad position instead of pad size (round pas only).
Eeschema:
* fix a bug: new created wires were not alwaes shows until the screen was redrawn.
wxWidgets 2.9 knows wxSVGFILEDC device context.
Unfortunately, arcs are drawn as pies (this is a feature, not a bug).
This is not correct for Kicad.
So Kicad has its own wxSVGFILEDC (named KicadSVGFileDC) that is basically the same as wxSVGFILEDC,
but with arcs drawn as arcs, not pies.
Note also under wxWidgets 2.8 does not know wxSVGFILEDC , so kicad had already its own SVG device context.
* Convert Eeschema from manually linked list to DLIST for storing
SCH_ITEM objects.
* Add helper functions to SCH_SCREEN for appending list of SCH_ITEM
objects.
* Fix bug in wire editing code for accurate undo/redo behavior.
* Add member to DLIST to append another DLIST.
* Other minor code cleaning.
derived classes' implementations of these functions become non virtual and
can be truly _inlined_ for speed! All GetPosition() in derived classes were also
changed to return const wxPoint&, that is, a reference rather than a
full copy of the position wxPoint. There was no need for polymorphism in
{Get,Set}Position() since we never call these functions via generic pointer.
* Remove BOARD::{Get,Set}Position() since they were only there to satisfy
the pure virtuals established in BOARD_ITEM, which are now gone.
* Added const wxPoint& CPolyLine::GetPos(), made CPolyLine::Get{X,Y}() inline.
* Derive CPolyPt from wxPoint so we can return "const wxPoint&" fromt
CPolyLine::GetPos().
Eeschema: Viewlib is now accessible from Libedit (to browse libs or load a component to edit)
Libedit: uses now the same dialog as the schematic editor to load a component
Some code cleaning.
which is in turn already within BOARD.
* encapsulate class D_PAD with accessors, making data private.
* make D_PAD::GetBoundingRadius() do its longer calculation lazily, based on
m_boundingRadius == -1.
* Added a dialog to select options: one or 2 files, units (mm or inch), and force INSERT option.
* code cleaning
* Modify corresponding icon in menu to show a footprint, not a schematic component.
* Eeschema can generate this netlist format.
* Pcbnew can use (automatic identification) the current format or the new format.
* Cvpcb does not use yet the new format.
sizes. Tested with postscript output only. Required minor file format changes
to reflect the "portrait" setting. common/dialogs/dialog_page_settings.cpp
uses a checkbox but its name is "Landscape", which is inverted from portrait,
but since it is the more common choice, I used that rather than portrait.
The tooltip for that checkbox makes it clear. No portrait mode is supported
for "User" paper size.
* Remove unnecessary copy constructors from board and module library
objects.
* Add doClone() method to board and library objects.
* Add comment to class definitions where the default copy constructor
generated by the compiler was adequate.
* Replace copy method with clone method where applicable.
* Remove DuplicateStruct() function.
* Remove track object copy function.
* Remove unnecessary copy constructors from schematic and component
library objects.
* Add comment to class definitions where the default copy constructor
generated by the compiler was adequate.
* Add assignment operator to EDA_ITEM, SCH_ITEM, and all schematic
objects where the default assignment operator generated by the
compiler would not be adequate.
This feature is very useful for electronics hobbyists who use DIY PCB technology
(both toner transfer and photoresist methods).
Also width correction may be useful for PCB designers who take care of track width etching.
This patch also fixes some minor PS plotting issues.
* Move SCH_ITEM::Place() code into SCH_EDIT_FRAME so schematic item
objects no longer have any knowledge of the schematic frame window
or undo/redo containers.
* EDA_DRAW_PANEL completely encapsulated.
* Moved OSX m_overlay member from EDA_DRAW_PANEL to EDA_DRAW_FRAME where
it is used.
* Doxygen comment warning fixes.
* Add check box and flag to ignore replacing the component reference
designator field.
* Fix a resizing bug when switching between the find and replace modes of
the find/replace dialog.
* Fix replace component reference designator field bug.
* Force rebuild of search list when schematic is modified and the replace
dialog is closed.
* Prevent finding the next item when the find/replace dialog is closed.
* Respect warp mouse pointer setting when closing find/replace dialog.
* Fix duplicate mnemonic control accelerator keys in find/replace dialog.
* EDA_DRAW_FRAME completely encapsulated except for DrawFrame member.
* Moved members specific to Pcbnew from EDA_DRAW_FRAME to PCB_BASE_FRAME
or PCB_EDIT_FRAME as appropriate.
* Replace EDA_TOOLBAR with wxAuiToolBar as EDA_TOOL bar provided no
additional functionality and made code less readable.
* Remove EDA_TOOLBAR class definition from wxstruct.h and delete file
wineda_toolbar.cpp.
* Rename tool bar members to something more descriptive since the
horizontal and vertical references wont mean anything once the
tool bars are movable.
* Lots of dead code removal.
* changed all the RotatePoint() functions in trigo.{h,cpp} to take a double as the
angle, which is still in tenths of degrees for now.
* EDA_TEXT::m_Orient is now double, still in tenths of degrees (for a short while),
but the double means parsing files and saving them use a different format string.
++pcbnew
* DRAWSEGMENT::GetStart() and GetEnd() do not operate for S_ARC like they used to.
They are now simply accessors for m_Start and m_End. Use DRAWSEGMENT::GetArcStart()
and GetArcEnd() and GetCenter() for arcs. specctra_export.cpp was the only
source file dependent on the old behavior.
* DIMENSION::m_Text is now contained, not dynamically allocated.
* more kicad_plugin work.
* Enable replace toolbar button and menu entry.
* Improve find replace logging granularity.
* Fix find and replace dialog control hiding and disabling.
* Minor improvements to the SCH_FIND_REPLACE_DATA object.
* Move find collector list iterator into the collector object.
* Add visibility override flag to EDA_ITEM to temporarily show items that
are hidden during find and replace.
* added BOARD::GetPad(int), made BOARD::m_NetInfo private, and contained
within the BOARD, not dynamically allocated and owned via pointer.
* added BOARD::GetPadCount()
* changed NETINFO_LIST::GetCount() to GetNetCount()
* added BOARD::GetNetCount()
* more kicad_plugin work.
* Fix compil issue about automatically created pcb_plot_params.h :
pcb_plot_params.* moved from pcbnew to common because item_io.cpp (using this file) is compiled in common.
In common.cpp GetTimeStamp is renamed GetNewTimeStamp (a better name).
Pcbnew: prepare work to calculate connections between pads that inteserct and therefore can be connected without any track (composite pads).
* Removed Pcb_Frame argument from BOARD() constructor, since it precludes
having a BOARD being edited by more than one editor, it was a bad design.
And this meant removing m_PcbFrame from BOARD.
* removed BOARD::SetWindowFrame(), since BOARD::m_PcbFrame gone.
* Removed the global BOARD_DESIGN_SETTINGS which was in class_board.cpp
* Added BOARD_DESIGN_SETTINGS to the BOARD class, a full instance
* A couple of dialogs now only change BOARD_DESIGN_SETTINGS when OK is pressed,
such as dialog_mask_clearance, dialog_drc, etc.
* Removed common/pcbcommon.cpp's int g_CurrentVersionPCB = 1 and replaced it
with build_version.h's #define BOARD_FILE_VERSION, although there may be a
better place for this constant.
* Made the public functions in PARAM_CFG_ARRAY be type const.
void SaveParam(..) const and void ReadParam(..) const
* PARAM_CFG_BASE now has virtual destructor since we have various ways of
destroying the derived classes and boost::ptr_vector must be told about this.
* Pass const PARAM_CFG_ARRAY& instead of PARAM_CFG_ARRAY so that we can use
an automatic PARAM_CFG_ARRAY which is on the stack.
* PCB_EDIT_FRAME::GetProjectFileParameters() may no longer cache the array,
since it has to access the current BOARD and the BOARD can change.
Remember BOARD_DESIGN_SETTINGS are now in the BOARD.
* Made the m_BoundingBox member private, this was a brutally hard task,
and indicative of the lack of commitment to accessors and object oriented
design on the part of KiCad developers. We must do better.
Added BOARD::GetBoundingBox, SetBoundingBox(), ComputeBoundingBox().
* Added PCB_BASE_FRAME::GetBoardBoundingBox() which calls BOARD::ComputeBoundingBox()
* Removed Pcb_Frame argument from BOARD() constructor, since it precludes
having a BOARD being edited by more than one editor, it was a bad design.
And this meant removing m_PcbFrame from BOARD.
* removed BOARD::SetWindowFrame(), and BOARD::m_PcbFrame
* Removed the global BOARD_DESIGN_SETTINGS which was in class_board.cpp
* added BOARD_DESIGN_SETTINGS to the BOARD class, a full instance
* a couple dialogs now only change BOARD_DESIGN_SETTINGS when OK is pressed,
such as dialog_mask_clearance, dialog_drc, etc.
* Removed common/pcbcommon.cpp's int g_CurrentVersionPCB = 1 and replaced it
with build_version.h's #define BOARD_FILE_VERSION, although there may be a
better place for this constant.
* Made the public functions in PARAM_CFG_ARRAY be type const.
void SaveParam(..) const and void ReadParam(..) const
* PARAM_CFG_BASE now has virtual destructor since we have various way of
destroying the derived class and boost::ptr_vector must be told about this.
* Pass const PARAM_CFG_ARRAY& instead of PARAM_CFG_ARRAY so that we can use
an automatic PARAM_CFG_ARRAY which is on the stack.\
* PCB_EDIT_FRAME::GetProjectFileParameters() may no longer cache the array,
since it has to access the current BOARD and the BOARD can change.
Remember BOARD_DESIGN_SETTINGS are now in the BOARD.
* Made the m_BoundingBox member private, this was a brutally hard task,
and indicative of the lack of commitment to accessors and object oriented
design on the part of KiCad developers. We must do better.
Added BOARD::GetBoundingBox, SetBoundingBox(), ComputeBoundingBox().
* Added PCB_BASE_FRAME::GetBoardBoundingBox() which calls BOARD::ComputeBoundingBox()
Enhanced algorithms to calculate board connections:
Previously, a track was seen connected to a pad only if the track end is
exactly on the pad position.
Now a track is seen connected to a pad if the track end is inside the pad shape.
Algorithm to calculate pads connections to track is still very fast.
However some other functions (drag pads, track len calculation ...)
still need the track end exactly on the pad position.
Dead code removal.