From 2e1e64ae8c5a5ff997eda11b0b2f2482563bcc56 Mon Sep 17 00:00:00 2001
From: liquidev <liquidev@tutanota.com>
Date: Sat, 30 Nov 2024 20:41:04 +0100
Subject: [PATCH] add SVG size computation to ImageSizeCache

based on a very simple reading of the SVG's XML to find out its `width`, `height` and `viewBox`
no further parsing is required or will be provided
and you are an excellent test subject
---
 Cargo.lock                                   |  7 ++
 crates/treehouse/Cargo.toml                  |  1 +
 crates/treehouse/src/vfs/image_size_cache.rs | 72 ++++++++++++++++++--
 3 files changed, 75 insertions(+), 5 deletions(-)

diff --git a/Cargo.lock b/Cargo.lock
index 4305391..86dc7c6 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -2154,6 +2154,7 @@ dependencies = [
  "tracing-subscriber",
  "treehouse-format",
  "ulid",
+ "xmlparser",
 ]
 
 [[package]]
@@ -2491,6 +2492,12 @@ version = "0.5.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1e9df38ee2d2c3c5948ea468a8406ff0db0b29ae1ffde1bcf20ef305bcc95c51"
 
+[[package]]
+name = "xmlparser"
+version = "0.13.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "66fee0b777b0f5ac1c69bb06d361268faafa61cd4682ae064a171c16c433e9e4"
+
 [[package]]
 name = "yoke"
 version = "0.7.5"
diff --git a/crates/treehouse/Cargo.toml b/crates/treehouse/Cargo.toml
index d58ae13..509b003 100644
--- a/crates/treehouse/Cargo.toml
+++ b/crates/treehouse/Cargo.toml
@@ -31,3 +31,4 @@ tracing.workspace = true
 tracing-chrome = "0.7.2"
 tracing-subscriber = { version = "0.3.18", features = ["env-filter"] }
 ulid = "1.0.0"
+xmlparser = "0.13.6"
diff --git a/crates/treehouse/src/vfs/image_size_cache.rs b/crates/treehouse/src/vfs/image_size_cache.rs
index 48bb5f7..ec0fcab 100644
--- a/crates/treehouse/src/vfs/image_size_cache.rs
+++ b/crates/treehouse/src/vfs/image_size_cache.rs
@@ -29,11 +29,16 @@ where
     fn compute_image_size(&self, path: &VPath) -> anyhow::Result<Option<ImageSize>> {
         if path.extension().is_some_and(config::is_image_file) {
             if let Some(content) = query::<Content>(&self.inner, path) {
-                let reader = image::ImageReader::new(Cursor::new(content.bytes()))
-                    .with_guessed_format()
-                    .context("cannot guess image format")?;
-                let (width, height) = reader.into_dimensions()?;
-                return Ok(Some(ImageSize { width, height }));
+                if path.extension() == Some("svg") {
+                    return Ok(svg_size(&content.string()?));
+                } else {
+                    let _span = info_span!("raster_image_size").entered();
+                    let reader = image::ImageReader::new(Cursor::new(content.bytes()))
+                        .with_guessed_format()
+                        .context("cannot guess image format")?;
+                    let (width, height) = reader.into_dimensions()?;
+                    return Ok(Some(ImageSize { width, height }));
+                }
             }
         }
 
@@ -77,3 +82,60 @@ where
         write!(f, "ImageSizeCache({:?})", self.inner)
     }
 }
+
+/// Quickly determine the size of an SVG without parsing it into a DOM.
+///
+/// This method is a tentative check; the point is to return an image size that's _good enough
+/// default_ rather than what the size is going to be in the user's web browser.
+#[instrument(skip(svg))]
+fn svg_size(svg: &str) -> Option<ImageSize> {
+    let mut tokenizer = xmlparser::Tokenizer::from(svg);
+
+    fn parse_view_box(s: &str) -> Option<[u32; 4]> {
+        let mut iter = s.split_whitespace();
+        let min_x = iter.next()?.parse().ok()?;
+        let min_y = iter.next()?.parse().ok()?;
+        let width = iter.next()?.parse().ok()?;
+        let height = iter.next()?.parse().ok()?;
+        Some([min_x, min_y, width, height])
+    }
+
+    let mut in_svg = false;
+    let mut width: Option<u32> = None;
+    let mut height: Option<u32> = None;
+    let mut view_box: Option<[u32; 4]> = None;
+    while let Some(Ok(token)) = tokenizer.next() {
+        if let xmlparser::Token::ElementStart { local, .. } = &token {
+            if local == "svg" {
+                in_svg = true;
+                continue;
+            }
+        }
+
+        if in_svg {
+            // If another element starts, we're no longer in the root <svg>.
+            if let xmlparser::Token::ElementStart { .. } = &token {
+                break;
+            }
+
+            if let xmlparser::Token::Attribute { local, value, .. } = &token {
+                match local.as_str() {
+                    "width" => width = value.parse().ok(),
+                    "height" => height = value.parse().ok(),
+                    "viewBox" => {
+                        view_box = parse_view_box(value);
+                    }
+                    _ => (),
+                }
+                continue;
+            }
+        }
+    }
+
+    match (width, height, view_box) {
+        (Some(width), Some(height), _) | (_, _, Some([_, _, width, height])) => {
+            Some(ImageSize { width, height })
+        }
+        _ => None,
+    }
+}