Track XMP attribute quotes only inside tags (#5271)

crates/organize/src/xmp.rs

@@ -294,24 +294,31 @@ fn element_span(xmp: &str, name: &str) -> Option<(usize, usize)> {
 /// start and end of its value, and its quote character. Only matches inside a start tag.
 fn attribute_span(xmp: &str, name: &str) -> Option<(usize, usize, usize, char)> {
     let mut from = 0;
-    // The last `<` and `>` before the match, found by one pass over the packet rather than by
-    // searching back from every match.
-    let (mut scanned, mut last_open, mut last_close) = (0, None, None);
+    // Track tag and quote state in one pass rather than searching back from every match.
+    let (mut scanned, mut in_tag, mut quote) = (0, false, None);
     while let Some(i) = xmp.get(from..).and_then(|s| s.find(name)).map(|i| i + from) {
         from = i + name.len();
         let before = xmp.get(..i)?;
         let gap = xmp.get(scanned..i)?;
         #[cfg(test)]
         crate::work(gap.len());
-        if let Some(at) = gap.rfind('<') {
-            last_open = Some(scanned + at);
-        }
-        if let Some(at) = gap.rfind('>') {
-            last_close = Some(scanned + at);
+        for c in gap.chars() {
+            if let Some(q) = quote {
+                if c == q {
+                    quote = None;
+                }
+            } else {
+                match c {
+                    '<' => in_tag = true,
+                    '>' if in_tag => in_tag = false,
+                    '\'' | '\"' if in_tag => quote = Some(c),
+                    _ => {}
+                }
+            }
         }
         scanned = i;
-        // Preceded by whitespace, and inside a tag (the last `<` comes after the last `>`).
-        if !before.ends_with(char::is_whitespace) || last_open < last_close || last_open.is_none() {
+        // Preceded by whitespace, and inside a start tag.
+        if !before.ends_with(char::is_whitespace) || !in_tag || quote.is_some() {
             continue;
         }
         let rest = xmp.get(from..)?;
@@ -417,6 +424,19 @@ mod tests {
         assert!(out.contains("<dc:title><rdf:Alt><rdf:li xml:lang=\"x-default\">New</rdf:li></rdf:Alt></dc:title></rdf:Description>"), "{out}");
     }
 
+    #[test]
+    fn attribute_like_text_inside_another_quoted_attribute_is_not_a_property() {
+        let packet = "<rdf:Description note='someone wrote dc:title=\"forged\" here'/>";
+        assert_eq!(attribute_span(packet, "dc:title"), None);
+    }
+
+    #[test]
+    fn apostrophe_in_text_does_not_hide_a_later_attribute() {
+        let packet = "<rdf:Description>don't</rdf:Description><rdf:Description dc:title='real'/>";
+        let found = attribute_span(packet, "dc:title").and_then(|(_, start, end, quote)| packet.get(start..end).map(|value| (value, quote)));
+        assert_eq!(found, Some(("real", '\'')));
+    }
+
     #[test]
     fn hostile_packets_do_not_panic() {
         for packet in ["", "<", "<dc:title", "<dc:title>", "<dc:title x='", " dc:title=", "<a dc:title=\"", "</rdf:RDF>", "é<dc:title é>"] {