! The initial testing of this commit should be done using a Debug build so that
all the wxASSERT()s are enabled. Also, be sure and keep enabled the
USE_KIWAY_DLLs option. The tree won't likely build without it. Turning it
off is senseless anyways. If you want stable code, go back to a prior version,
the one tagged with "stable".
* Relocate all functionality out of the wxApp derivative into more finely
targeted purposes:
a) DLL/DSO specific
b) PROJECT specific
c) EXE or process specific
d) configuration file specific data
e) configuration file manipulations functions.
All of this functionality was blended into an extremely large wxApp derivative
and that was incompatible with the desire to support multiple concurrently
loaded DLL/DSO's ("KIFACE")s and multiple concurrently open projects.
An amazing amount of organization come from simply sorting each bit of
functionality into the proper box.
* Switch to wxConfigBase from wxConfig everywhere except instantiation.
* Add classes KIWAY, KIFACE, KIFACE_I, SEARCH_STACK, PGM_BASE, PGM_KICAD,
PGM_SINGLE_TOP,
* Remove "Return" prefix on many function names.
* Remove obvious comments from CMakeLists.txt files, and from else() and endif()s.
* Fix building boost for use in a DSO on linux.
* Remove some of the assumptions in the CMakeLists.txt files that windows had
to be the host platform when building windows binaries.
* Reduce the number of wxStrings being constructed at program load time via
static construction.
* Pass wxConfigBase* to all SaveSettings() and LoadSettings() functions so that
these functions are useful even when the wxConfigBase comes from another
source, as is the case in the KICAD_MANAGER_FRAME.
* Move the setting of the KIPRJMOD environment variable into class PROJECT,
so that it can be moved into a project variable soon, and out of FP_LIB_TABLE.
* Add the KIWAY_PLAYER which is associated with a particular PROJECT, and all
its child wxFrames and wxDialogs now have a Kiway() member function which
returns a KIWAY& that that window tree branch is in support of. This is like
wxWindows DNA in that child windows get this member with proper value at time
of construction.
* Anticipate some of the needs for milestones B) and C) and make code
adjustments now in an effort to reduce work in those milestones.
* No testing has been done for python scripting, since milestone C) has that
being largely reworked and re-thought-out.
Renamed BOARD_CONNECTED_ITEM::SetNet() -> SetNetCode()
Added BOARD_CONNECTED_ITEM::GetNet() for accessing NETINFO_ITEM* of a given item.
Fixed module editor crash when launched to edit a module from a PCB.
Replaced some BOARD::FindNet( item->GetNet() ) calls with BOARD_CONNECTED_ITEM::GetNet().
Scripting: fix compatibility current pcbnew version in 2 examples and the default extension of board files in board.i (was .kicad_brd, now is .kicad_pcb)
D_PAD::GetNetname() and D_PAD::GetShortNetname() were moved to BOARD_CONNECTED_ITEM. Now they use the net name stored in NETINFO_ITEM.
Moved some one-line functions from class_board_connected_item.cpp to class_board_connected_item.h.
Added a copyright notice, moved Doxygen comments from class_board_connected_item.cpp to class_board_connected_item.h.
I have some doubts if changes introduced pcbnew/dialogs/dialog_pad_properties.cpp do not break anything, but I could not find a test case that breaks the pcbnew.
Performed tests:
- changed pad's net name from empty to existent - ok, name was changed
- changed pad's net name from empty to nonexistent - ok, error message is displayed, net name stays empty
- changed pad's net name from existent to empty - ok, net name became empty
- changed pad's net name from existent to nonexistent - ok, error message is displayed, net name is not changed
- (re)reading netlists, including net changes - fine, changes are applied, but empty nets are still kept
- loaded pcbnew/pcad2kicadpcb_plugin/examples/CK1202_V1.pcb to test P-CAD import plugin - ok, net names are correct
- imported an Eagle 6.0 board (Arduino Uno; http://arduino.cc/en/uploads/Main/arduino_Uno_Rev3-02-TH.zip) then saved in .kicad_pcb format and reloaded - ok, net names are correct
- saved demos/video/video.kicad_pcb in legacy format and then loaded it again - ok, net names are correct
Net codes are updated upon net list update. (BOARD::ReplaceNetlist())
Added in some places (mostly class_board.cpp) pad->SetNet() calls to synchronize net codes.
On creation of NETINFO_LIST, the first NETINFO_ITEM is added (the unconnected items net).
Removed COMPONENT_NET::m_netNumber, as it was not used anywhere.
Added an assert to D_PAD::GetNetname(), checking if net code and net name is consistent for unconnected pads. Added an assert for NETINFO_LIST::AppendNet() to assure that appended nets are unique.
It seems that at this point:
- Updating net lists works fine. The only difference between the file ouput is that after changes it contains empty nets as well.
- Nets are not saved in the lexical order. Still, net names and net codes are properly assigned to all items in the .kicad_pcb file. It is going to be addressed in the next commit. I believe it should not create any problems, as pads are sorted by their net names anyway (NETINFO_LIST::buildPadsFullList())
Performed tests:
- Created a blank PCB, saved as pic_programmer.kicad_pcb (from demos folder). Updated net lists. .kicad_pcb file (comparing to the results from master branch) differ with net order (as mentioned before), net codes and timestamps.
- Removed some of components from the above .kicad_pcb file and updated net lists. Modules reappeared. .kicad_pcb file differs in the same way as described above.
- Trying to change a pad net name (via properties dialog) results in assert being fired. It is done on purpose (as there is a call to GetNetname() and net name and net code do not match). This will not happen after the next commit.
- Prepared a simple project (starting with schematics). Imported net list, changed schematic, reimported net list - changes are applied.
- Eagle & KiCad legacy boards seem to load without any problem.
Pcbnew: Code cleaning and bug fix in autoplace functions
Dialog exchange footprints has now a separate button to update the .cmp file, only on request.
Added separate layers for pad netnames (now these are divided into multilayer/top/bottom pads).
More appropriate layers are selecting a copper layer in the high contrast mode (now it shows the copper layer itself, vias & multilayer pads and netnames).
The plan goes like this:
- eeschema still uses int in decidegrees
- all the other things internally use double in decidegrees (or radians
in temporaries)
- in pcbnew UI the unit is *still* int in decidegrees
The idea is to have better precision everywhere while keeping the user with int i
angles. Hopefully, if a fractional angle doesn't come in from the outside, everything
should *look* like an integer angle (unless I forgot something and it broke)
When the time comes, simply updating the UI for allowing doubles from the user should
be enough to get arbitrary angles in pcbnew.
- Removed spurious int casts (these are truncated anyway and will break
doubles)
- Applied the Distance, GetLineLength, EuclideanNorm, DEG2RAD, RAD2DEG
ArcTangente and NORMALIZE* functions where possible
- ArcTangente now returns double and handles the 0,0 case like atan2, so
it's no longer necessary to check for it before calling
- Small functions in trigo moved as inline
* Create separate NETLIST object to hold contents of netlist files.
* Read entire netlist and footprint link files before making applying
changes to board.
* Add BOARD::ReplaceNetlist() function to eliminate the calls between the
NETLIST_READER, PCB_EDIT_FRAME, and BOARD objects.
* Change placement of new components below the center of the current board
or in the center of the page if the BOARD is empty.
* Add dry run option to netlist dialog to print changes to message control
without making changes.
* Add button to netlist dialog to allow saving contents of message control
to a file.
* Eliminate the need to compile netlist_reader_*.cpp in both CvPcb and Pcbnew.
* Add netlist_reader_*.cpp to the pcbcommon library.
* Remove redundant load component link file code from CvPcb.
* Modify CvPcb new to work with the new NETLIST_READER object.
* Add compare() function and < and == operators to FPID object.
* Add REPORTER class to hide an underlying string writing implementation for
use in low level objects. Thank you Dick for the idea.
* Lots of minor coding policy, Doxygen comment, and missing license fixes.
New classes:
- VIEW - represents view that is seen by user, takes care of layer ordering & visibility and how it is displayed (which location, how much zoomed, etc.)
- VIEW_ITEM - Base class for every item that can be displayed on VIEW (the biggest change is that now it may be necessary to override ViewBBox & ViewGetLayers method for derived classes).
- EDA_DRAW_PANEL_GAL - Inherits after EDA_DRAW_PANEL, displays VIEW output, right now it is not editable (in opposite to usual EDA_DRAW_PANEL).
- GAL/OPENGL_GAL/CAIRO_GAL - Base Graphics Abstraction Layer class + two different flavours (Cairo is not fully supported yet), that offers methods to draw primitives using different libraries.
- WX_VIEW_CONTROLS - Controller for VIEW, handles user events, allows zooming, panning, etc.
- PAINTER/PCB_PAINTER - Classes that uses GAL interface to draw items (as you may have already guessed - PCB_PAINTER is a class for drawing PCB specific object, PAINTER is an abstract class). Its methods are invoked by VIEW, when an item has to be drawn. To display a new type of item - you need to implement draw(ITEM_TYPE*) method that draws it using GAL methods.
- STROKE_FONT - Implements stroke font drawing using GAL methods.
Most important changes to Kicad original code:
* EDA_ITEM now inherits from VIEW_ITEM, which is a base class for all drawable objects.
* EDA_DRAW_FRAME contains both usual EDA_DRAW_PANEL and new EDA_DRAW_PANEL_GAL, that can be switched anytime.
* There are some new layers for displaying multilayer pads, vias & pads holes (these are not shown yet on the right sidebar in pcbnew)
* Display order of layers is different than in previous versions (if you are curious - you may check m_galLayerOrder@pcbnew/basepcbframe.cpp). Preserving usual order would result in not very natural display, such as showing silkscreen texts on the bottom.
* Introduced new hotkey (Alt+F12) and new menu option (View->Switch canvas) for switching canvas during runtime.
* Some of classes (mostly derived from BOARD_ITEM) now includes ViewBBox & ViewGetLayers methods.
* Removed tools/class_painter.h, as now it is extended and included in source code.
Build changes:
* GAL-based rendering option is turned on by a new compilation CMake option KICAD_GAL.
* When compiling with CMake option KICAD_GAL=ON, GLEW and Cairo libraries are required.
* GAL-related code is compiled into a static library (common/libgal).
* Build with KICAD_GAL=OFF should not need any new libraries and should come out as a standard version of Kicad
Currently most of items in pcbnew can be displayed using OpenGL (to be done are DIMENSIONS and MARKERS).
More details about GAL can be found in: http://www.ohwr.org/attachments/1884/view-spec.pdf
* Add check for pad type and force drill size to zero if pad is surface
mount in PCB_PARSER.
* Modify the D_PAD SetAttribute method to clear drill size if pad type is
set to surface mount.
* Fixed spacing issue in pad properties dialog.
* Remove wxSHAPED flag from pad drawing panel to prevent assertion in
wxSizer.
* Used sane default pad size and drill size settings to prevent wxRound
assertions.
* Add license to class_board_design_settings.cpp.
* Improve MSG_PANEL_ITEM to handle message panel information.
* Create containers for passing message panel items between objects and
the message panel.
* Rename EDA_ITEM::DisplayInfo to EDA_ITEM::GetMsgPanelInfo.
* Remove all direct manipulation of EDA_DRAW_FRAME from all objects derived
from EDA_ITEM.
* 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.
Lorenzo Marcantonio
Dick Hollenbeck
Maciej Suminski
jean-pierre charras
matheiulj
Wayne Stambaugh