Does VS necessarily have to close open unsaved files to be able to check for an build update?
-
I used VS yesterday and while I saved the design at some point, did not save the most recent edits (also did not close the file) I was working on at that time, but did leave it open in VS overnight. When I awakened the M1 Mac from sleep this morning, VS immediately wanted to close my open files and automatically gave me the warning option to save my changed edited files before VS closed them. See screenshot below:
I assume this was because the new update 1.2.032 was available (I was using 1.2.031 last night) and VS wants to be able to close open files to check for a new update automatically.
If this is the case, I understand what it is doing and why, but wondering if there's a better way for VS to handle this kind of situation. The user does not necessarily know that VS wants to quit open files to check for an update but one wrong click and unsaved work goes away for good.
-
@Boldline Yes, the update consists of a restart.
-
@VectorStyler said in Does VS necessarily have to close open unsaved files to be able to check for an build update?:
@Boldline Yes, the update consists of a restart.
Yes I understand this but what about reversing the order of how VS presents this to the user? Why not show there is an update and then if the user decides to update, then show the warning panel to save open documents? This would still allow for an auto-update option in VS so the user is aware of the update, but if they decline the update in that moment, they can keep working o their open projects without being told to save their files.
On a related side note, it gives the same warning if I go to manually check for an update - VS first then asks me to save my open files before checking for the update. I tested this with files I did not care about and left them unsaved and because there was no new update at the time I manually checked, nothing happened to the files I was warned about being unsaved
-
@Boldline I will try doing that. I think the problem was with the type of exit used by the Sparkle updater, I have to check that.
