I like you suggestion very much, I'll add this to my suggestion list and would implement this one for version 2.0.
Thank you.
How usable are svg favicons nowadays? Do I still need a dozen link rels in my <head> with various icons in different formats/sizes? Currently I have one .ico file as rel="shortcut icon", 4 pngs in various sizes as "icon", 4 pngs as "apple-touch-icon" and 1 svg as "mask icon"
I use https://opentype.js.org/ to convert a font to path.
As you can see in https://caniuse.com/link-icon-svg all modern browsers support SVG favicon right now (88% of users globally). You can also download the PNG version to support wider audience. In the next version, I'll include support to download PNG version for all required sizes, with a sample html page and manifest.json file.
Also, support for dark mode is in my todo list.
There's a good article on this here https://medium.com/swlh/are-you-using-svg-favicons-yet-a-gui...
I just use https://realfavicongenerator.net/ which keeps things relatively easy.
https://caniuse.com/link-icon-svg
Supported by Edge, Firefox, Chrome, and Opera. Safari and iOS in a rather limited way, but you're still looking at about 70% of users with SVG icons alone. Not so sure how to look up if they're supported directly in the link element.
I just tested the "emoji" version on Chrome and it rendered some total junk as the icon, so I'm not sure about support for that either.
You can check the browser supports on https://caniuse.com/link-icon-svg
You can download the PNG version to create favicon for older browsers, but it's something we'll cover in the next version. There would be a "Download All" option with all required PNG sizes, and a sample html file with higher priority for the SVG version. So modern browsers would use the SVG version, and old browsers uses the PNG files.
One seemingly major feature missing: ability to change the font size.
So, just a thought: pre-render and save versions of each common character in each available font. ([a-z,A-Z,0-9] and watch for other common selections.) When a user says, "I want an 'R'", you display a grid of pre-rendered SVG icons of R in all available fonts (so you don't have to load all of the actual fonts at the same time). As SVG, they could be viewed at any scale. They choose the font by clicking the icon instead of the font name. Then they choose size, color, etc., to customize it, and that one is rendered live.
Good work.
For font-picker, we have enough horizontal space. I might add another sidebar where you can preview and search font-families. I'll redesign the UI for the next version.
For example, the laughing with tears emoji is in full color with an open mouth, but selecting it results in the outlined version without color and is showing teeth.
When the colored emoji do work, they will sometimes be cropped at the bottom or will leave some artifacts behind from a previous selection. These seem to disappear when switching to another tab and then coming back, though.
Some of the letters will extend beyond the boundaries of the image, too. For example, the left and right ends of the combinations "me" or "ww" get clipped off.
Despite those issues, this looks promising. I really like the interface and overall simplicity of the tool.
Emojis are using the native emoji list from your machine. I should check the emoji-picker documentation to see why this happening. I'll see if I can find an emoji picker that returns SVG. That would solve this issue definitely.
Regards sizing, yeah, the text size is fixed at 80px. I'll add a control to change the font-size in upcoming updates.
Glad you like it so far. Will try to improve the tool soon.