Edit: or make it moveable/collapsible

Some obvious features which I missed:

- nodes tool should want to snap begin/end nodes together when editing (and the snap distance could be larger when initially creating, esp. w/ the pencil tool)

- rather than a Nodes tool for editing, something like to Wick Editor's Flash/Futurewave Smarksetch pull/push deforming would be much more mobile friendly

- if corner rounding doesn't work on polygons it should be hidden/disabled

(I presume it’s not written in Hy/Hylang, the cute little Lisp that compiles to Python.)

Granted, I wish it somehow could had the functionalities SVG Path Editor[0] has, which I haven't been able to find in any other editor (like, for example, converting absolute `<path>` coordinates to relative, editing each line, curve, and so on for each `<path>`).

Another feature that would be great to have but would require a monumental amount of work, or maybe even be its own project, would be animation.

Don't know if your spline editing mechanism (i.e. the ability to drag the line around from an aribtrary point rather than tweaking the individual handles) is something you came up with yourself or borrowed from elsewhere, but I really really like it.

Only thing I noticed is that touch scrolling doesn't work on the tree on the left. Otherwise it's very smooth.

Have you implemented any keyboard shortcuts?

Also what I always thought would be a real killer feature would be something similar to blender modifiers (array, mirror, etc.) but in a vector editor, allowing for non-destructive editing.

Couple of questions for OP

What is it written in? What will the license be?

Context: Currently evaluating the venerable SVGEdit (MIT, JavaScript) for a project

https://github.com/SVG-Edit/svgedit

I'll write my first impressions of https://github.com/hyvectorapp so it helps usability and improvement.

Oh it's a freemium app.

hyvectorapp starts off looking like figma, penpot style which is a good sign.

I can export svg [x]

There's no align to grid system.

vector tracing is not generally solved except via the vectormagic product and machine learning research prototypes. I wonder how you solved it.

Can't interact more today so I'll end with this note of hopefulness.

I might be slightly biased as I am more used to even/odd infill mode compared to non zero mode, but having user constantly keep track of which paths are clockwise and which are counterclockwise seems like unnecessary hassle, especially for paths without infill where it shouldn't matter. At least there is a UI indicator for it, otherwise I would be very confused.

I usually fall back to Illustrator and then clean up the resulting markup, or a text-based editor such as https://www.svgviewer.dev/

Your UX is quite polished, and your tool already supports more features than other ones I've found, good work!

For reference this is my project, https://github.com/jwmcglynn/donner, which has a web-based "editor" (currently just code-based editing) prototype here: https://jwmcglynn.github.io/donner-editor/

SVG is one of those things that has lots of potential but has been impacted by not-so-great tooling, it's my passion and I'm glad to see innovation in the space.

A couple First Impression Experience pieces of feedback:

* when it first starts, how about a blank untitled document to play with, and maybe select a fun tool like the Bezier pencil as the initially selected tool. I was able and motivated to click around, realize there was no open document (probably the biggest stumbling block), create a new document, change the tool, and start playing, but many users won't be.

* it seems like a small thing, but please make the default canvas a bit larger (maybe 512 or 500 square). Again just more fun in that critical 10-20 second window, which is all most people will realistically give when checking out a new thing. If you can't hook 'em in that timeframe, they simply browse away and you lose a ton of folks.

Looks very good, thank you for sharing.

- The Pen tool feels intuitive.

- I like how the Objects panel shows previews of each layer.

- It is nice to see clip-path support integrated well, and features such as converting text to outlines, and boolean operations - these seem like rare features.

Here's some critical feedback after using it:

- Keybindings to switch between tools would be useful, and they should be annotated on the tooltips

- I'd like to see zoom/panning supported with Ctrl +/- and panning when holding space and clicking/dragging.

- When editing groups of shapes, it is hard to select individual shapes in the group. Double-clicking switches to the Nodes tool, which is not exactly what I want. I'd like something where double-clicking a group goes into some sort of "Group Isolation" editing mode.

- When using the Pen tool, I'd like to be able to go back to tweak the previous points, such as adjusting the control points, etc.

- Undo in Pen mode does not undo the insertion of points in the path, instead of Undoes operations from before the shape was started.

- Text support seems limited, for example an embedded @font-face did not render correctly. Similarly for filter support. If the editor doesn't support these features I'd expect them to at least be rendered correctly by the browser, but it seems like SVG rendering here is independent.

One thing I’m curious about is whether you’ve considered a plugin or API layer for generative workflows? There’s a growing niche of users who blend SVG editing with scripted or AI-assisted visuals, and having a programmable layer (maybe like a mini language or JS bridge) could make Hyvector even more powerful.

Regardless, congrats on shipping this! Always great to see fresh tools that lower the barrier to thoughtful vector work.

- I really like the level of fine-grained control, even for changes that don't visibly affect the output, like reversing a path. For that case, some visual feedback would help, since it's hard to tell if it worked

- Copying a selection to the clipboard seems to include all fonts used in the scene (even if not selected) as base64 in the SVG, leading to a bloated output

Do you plan to keep the strict 1:1 mapping with SVG, or are you open to some abstractions? I still use Figma for creating SVGs because of one killer feature I can't work without anymore: Vector Networks [1]. They'd need some internal conversion to valid path data, but it's an abstraction that could add real value.

On the technical side, I'm surprised the editor uses SVG for rendering. Although of course it makes sense, I half expected it to use canvas under the hood. Lately I've been experimenting with rendering simple 3D objects using SVG and ran into performance issues pretty quickly, though I'm not yet sure what exactly' is causing them.

Did you do any stress testing yet, especially on older devices?

[1] https://www.figma.com/blog/introducing-vector-networks/

Is it just me though, or is the saving functionality wonky? I seem to be able to save but then "save as" doesn't prompt for a new filename / location. I wanted to fiddle with an svg, add additional layers, save a few versions locally - but seem unable to do this?

<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="20" height="20" viewBox="0 0 20 20" stroke-width="2" stroke="black"> <circle cx="10" cy="10" r="10" fill="currentColor" mask="url(#maskArrow)"></circle> <mask id="maskArrow"> <circle cx="10" cy="10" r="10" fill="white"></circle> <path d="M15.41 16.59L10.83 12l4.58-4.59L14 6l-6 6l6 6l1.41-1.41z"></path> </mask> </svg>