Replies: 7 comments 5 replies
-
I recently got informed about FontSetViewer. |
Beta Was this translation helpful? Give feedback.
-
I was also asked to include https://fonts.google.com/. |
Beta Was this translation helpful? Give feedback.
-
Hey all, I'm one of the reached-out-to-persons. 😄 I work on font rendering for Windows Terminal and will try to help wherever I can. When it comes to listing fonts, I would probably recommend listing them grouped by their typographic family name ( My recommendation is to install fonts for all users, always (= The problem with that approach is that access to the directory is restricted to elevated processes. Explorer can install fonts there without an UAC prompt (if the user is an Admin account), because they use |
Beta Was this translation helpful? Give feedback.
-
There is a comment from @RobCannon with 16 👍at current count not to "require" elevation. Unfortunately, packaged applications (MSIX based installers) require the font to be installed for the machine. Since WinGet does have the context of scope, we should be able to figure something out such that fonts can be installed as user or for the machine. I'm expecting we'll want to have some kind of informational message to help users understand the caveats with packaged applications. |
Beta Was this translation helpful? Give feedback.
-
My initial thoughts on the interaction here is to treat fonts as a separate vertical for logical containment. A separate repo might be too much (aka On winget.exe I would continue with the category separation. An entire Programmatic interfaces (COM API and PowerShell) would have a new installed fonts catalog, as well as a few new properties to expose categories. Beyond the catalog selection, the existing API surface would not need to be changed (except to potentially add properties specific to fonts if needed). I would expect that my recent work supporting side-by-side might play in here to better allow grouping of font families while still allowing for fine control if desired. There are many more implementation details that would need to be considered, but much of that is not necessarily important to end users/consumers of our interfaces. |
Beta Was this translation helpful? Give feedback.
-
I believe this is a cool idea. (1) Use cases
The last time I had to do this for a customer CI objective we end up using a small tool for thats.
@denelon your link to install fonts is still correct but overly improved by Windows Settings Uri ms-settings:fonts (2) Remote related, default fonts and protection
I wished Cascadia Code would be replacing consolas and Courier New as a new default for all fixed fonts in Windows.
(3) Benefits of WinGet and font management
Currently deployment of fonts is not that easy. We ended up to use a helper tool as Windows native Tools didn't do the trick for remote deployment of fonts at scale. (4) potential (legal) issues Fonts are often underlying licensing and certainly copyright. (5) performance and clutter Winget search should not be affected by adding fonts in terms of results, consistenty and if possible speed, when searching for apps. Could this be solved by using a new repo for fonts, which is searched last or adding tags? With 4 and 5, I support the comments and thoughts of @JohnMcPMS . |
Beta Was this translation helpful? Give feedback.
-
WinGet should throw an error if the fonts are being used by other programs. |
Beta Was this translation helpful? Give feedback.
-
Hey all, I've started digging into what it would look like to support managing fonts using WinGet. I've reached out to a few folks to get some feedback on this, and I wanted to keep the "noise" out of the Issue:
I don't think we want to treat them the same way we do with portable packages due to the way WinGet creates entries for portable apps like Microsoft.NuGet in Apps & Features.
I've been looking at a few different references like:
I've also been thinking about the implications of adding them to the WinGet Community Repository and how that has implications on the size of the PreIndexed cache.
One of the more popular sites (Nerd Fonts) already has a decent collection of fonts for us to potentially evaluate when we get to the point of supporting an experimental feature for managing fonts.
Since we're likely to be enumerating over a directory "C:\Windows\Fonts" and each file can have its own version, I'm expecting some additional complexity when an upgrade to a font could represent downloading a set of files.
Thoughts?
Beta Was this translation helpful? Give feedback.
All reactions