It's too expensive to fetch the timestamps when the github
server is busy. See Dick Hollenbeck's comments at the top of
github_plugin.cpp for more info.
Also adds some safety to the other caching algorithms after
seeing github_plugin's wild-west usage of the kicad_plugin.
Fixes: lp:1753143
* https://bugs.launchpad.net/kicad/+bug/1753143
Allow partial library reads in libraries that support footprint per file.
This includes the KiCad and GEDA footprint libraries.
Allow for partially cached libraries rather than ignoring all valid files
when an error occurs.
Currently, kicad_pcb files have a (version ...) tag, but it is ignored. This
commit does the following:
1. Parse the version. If it's later than the last supported version, present an
alternative error message suggesting an upgrade if the file does not load
correctly. The version will be interpreted as an integer YYYYMMDD to suggest
a KiCad release date.
2. Accept a (version ...) tag also in kicad_mod files, but do not write them
yet. If no version tag is present in these files, assume the version to be that
of the current format version at the time of this commit.
This is meant to be merged to the 4.x stable series, and preps for KiCad 5
which will start emitting version tags also in footprints - users with what
will then be 'old stable' will not get a parse error when we introduce this for
footprints, and we can safely increment the format version later.
Eagle plugin: if a board has a odd number of copper layers, the converted board copper layer count is rounded to the next even count (a odd count creates issues in .kicad_pcb files).
2) Change from legacy Cu stack to counting down from top=(F_Cu or 0).
The old Cu stack required knowing the count of Cu layers to make
sense of the layer number when converting to many exported file types.
The new Cu stack is more commonly used, although ours still gives
B_Cu a fixed number.
3) Introduce class LSET and enum LAYER_ID.
4) Change *.kicad_pcb file format version to 4 from 3.
5) Change fixed names Inner1_Cu-Inner14_Cu to In1_Cu-In30_Cu and their
meanings are typically flipped.
6) Moved the #define LAYER_N_* stuff into legacy_plugin.cpp where they
can die a quiet death, and switch to enum LAYER_ID symbols throughout.
7) Removed the LEGACY_PLUGIN::Save() and FootprintSave() functions.
You will need to convert to the format immediately, *.kicad_pcb and
*.kicad_mod (=pretty) since legacy format was never going to know
about 32 Cu layers and additional technical layers and the reversed Cu
stack.
Most of the changes are just adding GetDesignSettings() before every occurence of m_NetClasses.
More complex changes:
class_netclass.cpp - NETCLASS does not store the pointer to the parent BOARD anymore. Added function SetParams( BOARD_DESIGN_SETTINGS& ).
class_netclass.h - Removed GetTrackMinWidth(), GetViaMinDiameter(), GetViaMinDrill(), GetuViaMinDiameter(), GetuViaMinDrill() as they were refering to BOARD_DESIGN_SETTINGS anyway (they are not net class specific).
kicad_plugin.cpp - filters out empty nets (that are anyway not saved) when storing net class information. Previously it was done in NETCLASS::Format() function.
Now, nets are saved with consecutive net codes (both modern & legacy plugins).
Zones are saved together with their nets, without depending on the fact if there are any pads with such net. Therefore validation of zone net names was removed (pcbnew/class_board.cpp).
Performed tests:
- Changed a pad's net name from empty to existent - ok, name was changed.
- Changed a pad's net name from empty to nonexistent - ok, error message is displayed, net name stays empty.
- Changed a pad's net name from existent to empty - ok, net name became empty
- Changed a pad's net name from existent to nonexistent - ok, error message is displayed, net name is not changed.
- Drawn a zone that belongs to a net, then modified schematics so the net does not exist anymore. After reloading the net list, all pads/tracks are updated. Zones still belongs to the net that does not exist in the schematic (but still exists in .kicad_pcb file). After running DRC, the zone becomes not filled.
- Undo & redo affects assignment of a polygon to a specific net (you may change net of a polygon, refill it and undo/redo the changes).
- KiCad s-expr & legacy, Eagle, P-CAD boards seem to load without any problem (they also contain correct net names assigned to the appropriate pads). All types of board file formats were loaded, then saved in sexpr format and reopened with a KiCad built from the master branch (without my modifications).
- A few boards were also saved using the legacy format and were opened with the master KiCad without any issues.
- Change a net name for a pad, restore with undo/redo - ok
- Remove everything, restore with undo - ok
- Remove everything, reload netlist - ok
Differences observed between files saved by the master branch KiCad and this one:
- list of nets are not saved in any particular order, so net codes may differ
- the default net class does not contain the unconnected net
comment for GITHUB_PLUGIN which should flow into Doxygen output.
*) Rewrote:
PCB_BASE_FRAME::Save_Module_In_Library(): now uses fp-lib-table and PROPERTIES.
PCB_EDIT_FRAME::ArchiveModulesOnBoard(): now can archive to any writable library type.
PCB_BASE_FRAME::SelectLibrary(): is now generic for selecting a library, not just the active library.
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.
Tracks should be loaded in the same order as they were saved, no exceptions.
This is for the version control system user.
But even with this objective now implemented in LEGACY_PLUGIN, the TRACKS
are still going through a food processor after they are loaded, and the poor
version control system user has no relief, even with this patch.
TRACKS are always modified after being loaded, and I think this needs to be fixed still.
* 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.
Jeff Young
Wayne Stambaugh
Jon Evans
Maciej Suminski
Simon Richter
Chris Pavlina
jean-pierre charras
Dick Hollenbeck
Wayne Stambaugh