TODO:

• [x] Ugly stroking: e-image-003 (especially S), e-image-017, e-image-022 (clearly visible) (raqote bug)
• [x] a-stroke-dasharray-004 (raqote bug)
• [x] a-stroke-linecap-004 (should be done on the resvg size)
• [x] a-stroke-linecap-005 (should be done on the resvg size)
• [x] e-clipPath-019
• [x] e-feBlend-002 (raqote bug, BlendMode::Multiply produce nothing)
• [x] e-image-010 (raqote bug? Looks like some issue with a pattern.)
• [x] e-mask-007 (~~raqote bug? Looks like something is wrong with clip~~)
• [x] e-path-013 (raqote bug)
• [x] e-radialGradient-011 (raqote bug)
• [x] e-radialGradient-029
• [x] ~~e-stop-016 (raqote bug)~~
• [x] Markers are very slow, aka multiple sublayers.
• [x] Optional anti-aliasing.

Closes #134

Hello, I tried to use the function in my C++ project as I've done before, but when rendering the svg, text is not displayed after providing a font and calling the function with the font family. This used to work in releases before v0.28.0.

I see in the commit history that this function was removed then added back, so maybe something was missed when reverting that change?

The C++ project I'm referring to is: https://github.com/ANSH3LL/Graphics-Extensions-for-Solar2D and the line where I'm using the function is: https://github.com/ANSH3LL/Graphics-Extensions-for-Solar2D/blob/23151db1257915cf0aaa47df7728f6c0c4291b30/shared/PluginGfxe.cpp#L772

Loading fonts can be a pretty expensive operation, especially on targets like JS/Wasm, where there is no mmap-like feature, and one must load fonts asynchronously before invoking resvg.

To optimise this case, it would be handy if usvg provided a way to get set of fonts used in the SVG document before rendering it to the final form. That way, resvg-js users, as well as native ones, could only load fonts that will be definitely necessary for rendering the document.

I'm not sure how feasible this is given that usvg downcasts text to shapes as part of its processing, but maybe it could preserve text nodes in some intermediate form to give user the chance to do this async loading?

usvg currently works great in wasm environments as long as the optional "text" feature is disabled. I'd love to get text support working too. From my understanding so far, there are at least 3 issues:

• a dependency on memmap2, used for reading fonts
• reading fonts from system directories (#202)
• a dependency on harfbuzz

The first two are relatively straightforward. As part of #202, design an API to feed a FontDB into usvg that can be instantiated many different ways, one of them just being raw bytes of a font file. (include_bytes! is a low-effort way to get this working in a wasm environment.)

The harfbuzz dependency seems harder. My requirements are limited to simple Latin left-to-right text; harfbuzz is overkill. I found font2svg and think it might be possible to make something much simpler. Before I go down this rabbit hole, I wanted to ask for your advice.

Background

I'm using usvg in A/B Street to turn both normal SVG files and text into paths for lyon, which then give me polygons. This does the usvg->lyon->my simple Polygon abstraction path, based off this example from lyon. I'm using this same pipeline for text as well. (There are many inefficient and silly things that the text pipeline does right now, like constantly re-load fonts.) In a few weeks, I'm hoping to properly split off my drawing, geometry, and GUI libraries as new crates for other people to use. These libraries attempt to give exactly the same experience on native and web -- not using system UI controls/style -- without any changes to user code needed for different platforms.

There is recent work to define an SVG Native spec, which is a subset of SVG 1.1. Adobe recently published a SVG Native Viewer library (C++ 11), and the FreeType project is considering resvg among the candidates as the backend for rendering OpenType+SVG fonts.

I wonder how resvg relates to SVG Native, and if you have any plans to provide full compatibility to the SVG Native profile?

• http://lists.nongnu.org/archive/html/freetype-devel/2019-05/threads.html#00064
• https://w3c.github.io/svgwg/specs/svg-native/index.html
• https://github.com/adobe/svg-native-viewer
• https://docs.microsoft.com/en-us/typography/opentype/spec/svg
• https://helpx.adobe.com/fonts/using/ot-svg-color-fonts.html

There are some tools that allow people to make OpenType+SVG fonts

• https://www.fontlab.com/font-editor/fontlab-vi/
• https://github.com/microsoft/OpenType-SVG-Font-Editor
• https://github.com/adobe-type-tools/opentype-svg

Benchmarking on icons only is not particularly representative of performance in general.

I've tried rendering Inkscape about screens as they're bigger than icons and represent a wide variety of SVG feature combinations, using more advanced features over time as Inkscape itself evolved.

On my machine resvg takes 1 second to render them all while librsvg takes 5 seconds. I've blacklisted 0.48 splash screen as it has some unsupported filters. librsvg also renders the 0.92 splash screen incorrectly, while resvg works fine.

What

I.e. I only want the lower left quadrant of the image I would specify a ViewBox with a Rect::new(0.5, 0.5, 0.5, 0.5). FitTo would then pertain to the size of that region, not the total canvas.

Why

I am looking into adding support for SVG textures to a 3D renderer. During render time the renderer will request texture tiles at certain resolutions. These tiles are the aforementioned sub images.

Another use case is a viewer that e.g. allows zooming. When you zoom in 10000 times you do not want to be forced to render the whole canvas but only the visible portion.

How

~~It would be best if this worked by creating some sort of persistent representation of the SVG that can be queried for such 'sub' Pixmaps to keep any initial set-up/overhead of parsing etc. a one time cost.~~

~~I.e. you would create an SVG struct and call render_region() or the like on it, repeatedly.~~

I guess this already works efficiently when querying the same Tree/Node repeatedly?

TODO:

• [x] Sync with master.
• [ ] e-pattern-18
• [ ] e-marker-22 is very slow
• [x] Support older Skia.
• [x] Update BUILD.adoc
• [x] Test on all platforms.
• [ ] ~~Test on CI.~~

Closes #44

I would like to provide my own set of fonts with my application (as binary objects) and not reply on any system fonts at all. Is there a way to skip all system font resolution and to provide a font manually?

Hey, I've made some changes to allow you to use Skia's Canvas wherever possible. This allows you to use resvg with PDFs 🎉.

https://skia.org/user/api/skcanvas_creation

The SkPDF backend uses SkDocument instead of SkSurface, since a document must include multiple pages.

Something like this:

let pdf_doc = skia::PDF::make_document(&pdf_stream);
let pdf_canvas = pdf_doc.begin_page(width, height);
resvg::backend_skia::render_to_canvas(&tree, &opt, size, pdf_canvas);

Hi, this is really more of a question than a bug probably. I am setting up a new Windows 10 build server for OpenShot, and noticed that resvg.dll does not render any text. It does in fact render other shapes, gradients, and lines just fine... but anywhere "text" is supposed to be, it renders nothing.

git clone https://github.com/RazrFalcon/resvg
cd resvg/capi
QT_DIR="C:\\msys64\\mingw64\\" cargo build --verbose --release --features="qt-backend"

I compiled resvg on MSYS2 (using the latest version of Qt 5.12.4). No errors were reported during the build. I was also able to link to it with no issues. When OpenShot renders a SVG file though, all text is missing.

So, just curious if this rings a bell or if this is something silly on my part, missing a dependency, etc... If not, what do you suggest as the best way to troubleshoot an issue like this? Thank again!

Please consider adding build instructions to resvg and usvg showing people how to build a binary using rust/Cargo. (I tried for about 30 minutes before giving up.)

Since usvg converts markers into regular nodes, we cannot render them between path's fill and stroke. Which is allowed by the spec. We can render them only below or above the path.

The only way to implement this is by rewriting markers support. Basically removing this simplification step from usvg, which is unfortunate.

Hallo, I've noticed that patterns result in some anti-aliasing issues at the edges, if they have non-integer dimensions and the shapes inside the pattern go right against the edge.

To illustrate the case I made a large white rectangle covered by a rectangle on top, which has a pattern of 20.13x20.13 containing just a black rectangle filling the pattern exactly. In the PNG result you can see the white shining through, but the whole area should be black. (Note: the viewBox is only 0 0 200 200, but the image has to be larger to see the effect, so its width and height are set to 1000)

Any of these workarounds make it render correctly:

• the pattern has a width of 20x20
• the rectangle inside the pattern extends past the pattern dimensions, e.g. from -1,-1 to 21,21
• the rectangle inside the pattern uses shape-rendering "crispEdges"

but I believe the result should be the same without any of them.

Perhaps the pattern should be copied on all four sides while rendering its content?

Given the following input:

<svg viewBox="0 0 10 10" xmlns="http://www.w3.org/2000/svg">
  <g id="grp" transform="translate(2,2)" fill="blue">
    <rect id="sq" x="1" y="1" width="4" height="4"/>
  </g>
</svg>

Running usvg resolves attributes as expected:

<svg width="10" height="10" viewBox="0 0 10 10" xmlns="http://www.w3.org/2000/svg">
    <defs/>
    <path id="sq" fill="#0000ff" stroke="none" transform="matrix(1 0 0 1 2 2)" d="M 1 1 L 5 1 L 5 5 L 1 5 Z"/>
</svg>

However, when running usvg --keep-named-groups the transform from the group is translated to its matrix form but not propagated to the descendants, leaving the path element with an implicit/inherited transform attribute:

<svg width="10" height="10" viewBox="0 0 10 10" xmlns="http://www.w3.org/2000/svg">
    <defs/>
    <g id="grp" transform="matrix(1 0 0 1 2 2)">
        <path id="sq" fill="#0000ff" stroke="none" d="M 1 1 L 5 1 L 5 5 L 1 5 Z"/>
    </g>
</svg>

It feels like we would expect the following output in order to comply with the Micro SVG spec:

<svg width="10" height="10" viewBox="0 0 10 10" xmlns="http://www.w3.org/2000/svg">
    <defs/>
    <g id="grp">
        <path id="sq" fill="#0000ff" stroke="none" transform="matrix(1 0 0 1 2 2)" d="M 1 1 L 5 1 L 5 5 L 1 5 Z"/>
    </g>
</svg>