Automation: doc_info reports annotation and link rects in displayed-page coordinates (#129) (#242)

doc_info (and the MCP document-info resource built on it) serialised
annotation and link rectangles straight from the PDF /Rect, in user
space, while every other tool reports and accepts rectangles with the
origin at the top-left of the displayed page. Feeding a doc_info rect
to link_edit therefore moved the link, and on rotated pages the two
conventions disagreed in more than a y flip.

Convert both through the page's user_to_view, the same conversion
comment_list and link_list use, so the three agree. Pages that an
annotation cannot be matched to keep the raw rect.

Co-authored-by: wstock <1394687+wstock@users.noreply.github.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>

crates/automation/src/comments.rs

@@ -49,7 +49,9 @@ fn rect_to_user(p: &PageInfo, r: [f64; 4]) -> [f64; 4] {
     [a[0].min(b[0]), a[1].min(b[1]), a[0].max(b[0]), a[1].max(b[1])]
 }
 
-fn rect_to_view(p: &PageInfo, r: [f32; 4]) -> [f32; 4] {
+/// A user-space rectangle as the displayed-page rectangle every tool reports and accepts:
+/// origin at the top-left after `/Rotate`, rounded to 1/100 pt.
+pub(crate) fn rect_to_view(p: &PageInfo, r: [f32; 4]) -> [f32; 4] {
     let (a, b) = (p.user_to_view(r[0], r[1]), p.user_to_view(r[2], r[3]));
     let round = |v: f32| (v * 100.0).round() / 100.0;
     [round(a[0].min(b[0])), round(a[1].min(b[1])), round(a[0].max(b[0])), round(a[1].max(b[1]))]

crates/automation/src/lib.rs

@@ -1578,16 +1578,19 @@ fn info(d: &Document) -> Value {
             "page": n + 1, "label": p.label, "width": p.width, "height": p.height, "rotation": p.rotation,
         })).collect::<Vec<_>>(),
         "outline": outline(&i.outline),
+        // Rectangles use the tools' convention (top-left of the displayed page, like
+        // comment_list and link_list), not raw PDF user space, so they can be fed back to
+        // geometry-taking tools (#129).
         "annotations": i.annotations.iter().map(|a| json!({
             "page": page1(a.page), "type": a.subtype, "author": a.author, "contents": a.contents,
-            "modified": a.modified, "name": a.name, "in_reply_to": a.in_reply_to, "rect": a.rect,
+            "modified": a.modified, "name": a.name, "in_reply_to": a.in_reply_to, "rect": view_rect(i, a.page, a.rect),
         })).collect::<Vec<_>>(),
         "fields": i.fields.iter().map(|f| json!({
             "name": f.name, "kind": format!("{:?}", f.kind), "value": f.value, "page": f.page.map(page1),
             "tooltip": f.tooltip, "has_actions": f.has_actions,
         })).collect::<Vec<_>>(),
         "links": i.links.iter().map(|l| json!({
-            "page": page1(l.page), "rect": l.rect,
+            "page": page1(l.page), "rect": view_rect(i, l.page, l.rect),
             "target": match &l.target {
                 pdfcraft_render::LinkTarget::Page(p) => json!({ "page": page1(*p) }),
                 pdfcraft_render::LinkTarget::Uri(u) => json!({ "uri": u }),
@@ -1604,6 +1607,12 @@ fn info(d: &Document) -> Value {
     })
 }
 
+/// A user-space rectangle on 0-based `page` in displayed-page coordinates; unchanged when the
+/// page is unknown (a malformed annotation that points at no page).
+fn view_rect(i: &pdfcraft_render::DocInfo, page: usize, rect: [f32; 4]) -> [f32; 4] {
+    i.pages.get(page).map_or(rect, |p| comments::rect_to_view(p, rect))
+}
+
 fn outline(items: &[pdfcraft_render::OutlineItem]) -> Value {
     Value::Array(items.iter().map(|o| json!({ "title": o.title, "page": o.page.map(|p| p + 1), "children": outline(&o.children) })).collect())
 }

crates/automation/tests/automation.rs

@@ -1427,6 +1427,53 @@ fn links_through_tools() {
     assert!(matches!(a.call("link_add", &json!({ "doc": doc, "page": 1, "rect": [0, 0, 50, 20] })), Err(ToolError::InvalidArgs(_))));
 }
 
+/// doc_info reports annotation and link rectangles in the tools' displayed-page convention,
+/// the same values comment_list and link_list give, so they can be fed back to link_edit (#129).
+#[test]
+fn doc_info_rects_are_displayed_page_coordinates() {
+    let dir = workdir("info-rects");
+    let mut a = auto(&dir);
+    // Page 2 is rotated so an unrotated-only y flip cannot pass.
+    let doc = ok(&mut a, "doc_create", json!({ "from": "blank", "width": 612, "height": 792, "pages": 2 }))["doc"].as_u64().unwrap();
+    ok(&mut a, "page_rotate", json!({ "doc": doc, "pages": [2], "degrees": 90 }));
+    for page in 1..=2 {
+        ok(
+            &mut a,
+            "comment_add",
+            json!({ "doc": doc, "page": page, "type": "note", "at": [72, 72], "author": "Example", "contents": "Fixture note" }),
+        );
+        ok(&mut a, "link_add", json!({ "doc": doc, "page": page, "rect": [72, 100, 200, 120], "url": "https://example.com" }));
+    }
+    ok(&mut a, "doc_save", json!({ "doc": doc, "path": "fixture.pdf" }));
+    let re = ok(&mut a, "doc_open", json!({ "path": "fixture.pdf" }))["doc"].as_u64().unwrap();
+    let info = ok(&mut a, "doc_info", json!({ "doc": re }));
+    let comments = ok(&mut a, "comment_list", json!({ "doc": re }));
+    let links = ok(&mut a, "link_list", json!({ "doc": re }));
+    let close = |a: &Value, b: &Value| {
+        let (a, b) = (a.as_array().unwrap(), b.as_array().unwrap());
+        a.len() == 4 && a.iter().zip(b).all(|(x, y)| (x.as_f64().unwrap() - y.as_f64().unwrap()).abs() < 0.01)
+    };
+    for page in 1..=2 {
+        let page = json!(page);
+        let find = |items: &Value| items.as_array().unwrap().iter().find(|x| x["page"] == page).unwrap()["rect"].clone();
+        let (info_note, note) = (find(&info["annotations"]), find(&comments["comments"]));
+        assert!(close(&info_note, &note), "page {page}: doc_info note {info_note} vs comment_list {note}");
+        if page == 1 {
+            assert!(close(&note, &json!([72.0, 72.0, 92.0, 92.0])), "note {note}");
+        }
+        let (info_link, link) = (find(&info["links"]), find(&links["links"]));
+        assert!(close(&info_link, &link), "page {page}: doc_info link {info_link} vs link_list {link}");
+        assert!(close(&link, &json!([72.0, 100.0, 200.0, 120.0])), "page {page}: link {link}");
+    }
+    // Reusing the doc_info rectangle in a geometry-taking edit leaves the link where it is.
+    let rotated = links["links"].as_array().unwrap().iter().find(|l| l["page"] == 2).unwrap().clone();
+    let from_info = info["links"].as_array().unwrap().iter().find(|l| l["page"] == 2).unwrap()["rect"].clone();
+    ok(&mut a, "link_edit", json!({ "doc": re, "page": 2, "index": rotated["index"], "rect": from_info }));
+    let after = ok(&mut a, "link_list", json!({ "doc": re }));
+    let moved = after["links"].as_array().unwrap().iter().find(|l| l["page"] == 2).unwrap()["rect"].clone();
+    assert!(close(&moved, &rotated["rect"]), "link moved: {moved} vs {}", rotated["rect"]);
+}
+
 #[test]
 fn comment_checkmarks_locks_hiding_and_summaries_through_tools() {
     let dir = workdir("comment-polish");