According to what I found in the docs[1], there is no "DIRECTORY" available as <type>.
Therefore, correcting it to "PATH", which is according to the docs correct.
At the moment I'm getting the following error[2]:
-----------
CMake Warning (dev) at CMakeLists.txt:8 (set):
implicitly converting 'DIRECTORY' to 'STRING' type.
This warning is for project developers. Use -Wno-dev to suppress it.
CMake Warning (dev) at CMakeLists.txt:9 (set):
implicitly converting 'DIRECTORY' to 'STRING' type.
This warning is for project developers. Use -Wno-dev to suppress it.
---------------
[1] https://cmake.org/cmake/help/v3.6/command/set.html
[2] https://launchpadlibrarian.net/464670600/buildlog_ubuntu-focal-amd64.cura_1%3A4.5~202002121434~rev4433~pkg330~ubuntu20.04.1_BUILDING.txt.gz
Some inconsistencies with the ellipses, mostly.
There was also one line that was missing from every language. This happened because the Lionbridge import script I made doesn't properly process those.
Smartling also displayed the wrong texts to the translator sometimes. The translator didn't understand that they had to use singular for some texts, so those had to be corrected. I could correct them for every language except Turkish, so those lines are now wrong.
Contributes to issue CURA-7201.
For BAM and filled materials the default would give a warning position since for a filament switch it needs to retract more than for the normal print.
Done as a 5 minute fix.
Added additional option for disabling all retract options
Added more comments.
Added support for firmware based retractions.
Added support for outputting changes to LCD (untested).
Added type hints to most functions and variables.
Created GCodeCommand class for better detection of G1 vs G10 or G11 commands, and accessing arguments.
Fixed issue of properly restoring previous values in single layer option.
Improved wording of Single Layer vs Keep Layer to better reflect what was happening.
Moved most GCode methods to GCodeCommand class.
