* Fixed issue #704. TODO: replace algorithm by De Boor, check if accuracy is still maintained.
* Switch spline interpolation algorithm to De Boor for better numerical stability.
* Updated tests with DXF accuracy test, provide a warning for unsupported SPLINE types.
* Fixed a compiler warning
* Fixed#646 (crash on PCell Library "delete")
The issue was mainly caused by a "cleanup" call on
the library. Cleanup is supposed to remove excess
top level PCell variants. For libraries this is not
possible, as the library does not know which variants
are used and which are not.
In addition, some hardening against segfaults in
case of defect layouts has been applied.
* * Updated the build system for Mac
* Catalina env. uses Python 3.8 in MacPorts, Homebrew, and Anaconda3.
* Use Ruby 2.7 in MacPorts.
* Test build for macOS "Big Sur" (11.0) on Intel Mac "Catalina" (10.15.7).
* Updated the build system for macOS, including the future support for Big Sur (11.0)
* Changing indentation to 4 spaces
* squash
* removing more global variables (squash)
* | tee always exits with 0
* parameter is a dictionary now
* Changing default Makefile to HB38 and adding -rbinc2 option for Catalina
* Catalina building...
* macQAT fixed
* run_build_command depends on parameteres dict alone
* Adding options to macbuild/makeDMG4mac.py
In particular, we can now specify a -u (unsafe) option, and
specify the target dmg file with -t target.dmg
This patch introduces a slightly more elaborate scheme for building
menus on MacOS. Under normal circumstances, this scheme should maintain
the menu item order as far as possible when Qt is trying to sync with
Mac's menu bar.
Co-authored-by: Matthias <klayoutmatthias>
* WIP: added test case, fixed dup problem and '-' shortcut
* WIP: updated DRC doc and could not resist the temptation to fix 'it's' vs. 'its'
* Deep mode also working with select now. Updated tests.
In addition, the "destroyed" and "objectNameChanged" signals
were added (specifically to QObject).
The API binding for Qt5 was updated which adds some events.
* Password encoding to UTF-8 for Basic HTTP authentication - this is basically a hack. Needs to be reverted as soon as Qt switches to UTF-8 encoding.
* Workaround for issue #615