Windows won't allow long keys in the backend the keyring python package uses as a backend. This means the access_token part can't be stored in the obvious way. Timeboxed some attempts at working around this limitation, but couldn't make it work within the time set. As this is mostly an extra precaustion protecting users that share config folders around against themselves (in other words, if this goes wrong it's not unreasonable to blame the user) it's not top critical, and the important part of that (the refresh_token) can proceed, giving any potential attacker only a 10 minute window from the moment any user shares their %appdata%/cura files (again, this is not how we intent for users to behave, but they can and will do it this way).
CURA-7180
When there is only one file provider (i.e. the local file provider), the Open File(s) will be a
simple item in the File menu.
When there are more than one file providers, the Open File(s) will become a submenu in the File
menu, which will contain all the file providers as submenu items.
CURA-7868
Although it is not entirely sure yet, we are fairly certain that libArachne release will have
a setting version of 18, so it is changed by this commit. With this change, the VersionUpgrader
will be properly triggered.
Contributes to CURA-7683 and CURA-7656
If for some reason the print sequence is set to one-at-a-time be default in the printer definition
and the number of extruders of the printer is set to >=2, then the print_sequence won't show up in
neither the quality changes nor the user changes, yet it will still have "one-at-a-time" as a
value. In a such case, it will still need to be set to "all-at-once" in the user changes.
This is a theoretical case, as it is very unlikely for a printer to have "one-at-a-time" set by
default in its definition (.def.json) file and still be able to support more than one extruders.
But it is a world full of possibilities out there, so you never know...
CURA-7827
Previously, the print sequence would be always set to all-at-once in the user container whenever
the second extruder would be enabled. This was causing an interface issue, where the circular
reset-setting arrow would appear next to the setting, even if the quality changes already had the
correct value ("all-at-once").
This is now fixed by checking the following:
- If print_sequence == "one_at_a_time" in the quality (so it is already saved in a quality profile)
then set the print_sequence to "all_at_once" in the user changes so that it will be displayed as
an unsaved change
- If print_sequence == "one_at_a_time" in the user changes container only, meaning that it is not
saved in any quality profiles, then just reset the print_sequence in the user changes, so that it
will have the correct value ("all-at-once") without the reset-setting arrow appearing next to it.
CURA-7827
These were floats. Rounding numpy-floats still results in floats so I'm just using int(). The result is rounded down instead of nearest, but the difference is at most 50 nanometres anyway.
Contributes to issue CURA-7501.
The user doesn't know what a global stack is. This is a message that more clearly specifies what is really happening as far as the user can see.
The global stack may be None if this code is executed while there is no active printer. Normally the only time there is no active printer is:
- during first launch, while in the welcome screen
- during start-up until the profiles have been loaded
- if there was a profile corruption of some kind with the active printer.
So this function should not get called during those times. It probably isn't called except for the last case (in which case it's acceptable I guess).
This change is made after a request from a translator what the hell this means.
Contributes to issue CURA-7783.
It needs to use the double brim size, as it needs to ensure that
it has enough room for it's own brim and any object brim.
The resolution of the circle is increased so that the shadow matches
more with what is actually being printed
CURA-7643
The addMachineExtruders function assumes that this is a new printer (first time the printer is activated since a restart of Cura) and it needs to add the extruder stacks to it. However most of the time when we're switching, it's not to a new printer. If it's not a new printer, this function would get called anyway and we'd be adding the extruders multiple times. This then does a lot of unnecessary work and eventually fails a couple of calls deep.
This change prevents it from adding the extruders if there are already extruders associated with the printer, thus preventing this unnecessary work (switch is faster) and preventing the error from happening.
Discovered during work on CURA-7501.
