yrby 0.5.0 → 0.6.0

data/ext/yrby/src/render_rules.rs

@@ -0,0 +1,529 @@
+//! Custom render rules and segmented output — the extensibility core shared
+//! by both HTML renderers.
+//!
+//! Callers register per-node rules. Two tiers:
+//!
+//! - **Declarative rules** (tag, attributes, text, content slot) compile to
+//!   [`NodeRule`]/[`MarkRule`] here and render natively, inside the document
+//!   transaction, at full speed. This covers the tiptap-php `renderHTML`
+//!   shape: markup as data.
+//! - **Callback rules** defer to the caller. Rendering never runs app code
+//!   while the document is locked — the renderer emits [`Segment::Deferred`]
+//!   entries carrying the node's type, attributes (as JSON), and its
+//!   already-rendered children, and the caller fills them in after the
+//!   render returns. (In the Ruby gem, that caller is the app's block, run
+//!   once the transaction has closed and the GVL is held again.)
+//!
+//! Rules arrive as one JSON document (see `parse`), so the same format
+//! serves any binding or caller.
+
+use std::collections::{BTreeMap, BTreeSet, HashMap};
+use yrs::{Any, Out, ReadTxn, Xml};
+
+/// One piece of renderer output. `Html` is finished markup; `Deferred` is a
+/// callback node whose markup the caller supplies after the render, carrying
+/// everything needed to produce it. Content nests, so callback nodes inside
+/// callback nodes resolve depth-first. `child_types` lists the node's
+/// element/block children by type, in document order — structural facts a
+/// callback can't recover from `attrs` or the rendered content (an image
+/// count, whether a list item holds a nested list).
+#[derive(Debug)]
+pub enum Segment {
+    Html(String),
+    Deferred {
+        node_type: String,
+        attrs_json: String,
+        child_types: Vec<String>,
+        content: Vec<Segment>,
+    },
+}
+
+/// Builds segmented output. Renderers append markup through this instead of a
+/// bare `String`; frames capture sub-output (a deferred node's children, or a
+/// "did this render anything?" probe) without string sentinels.
+pub struct Emitter {
+    frames: Vec<Vec<Segment>>,
+}
+
+impl Emitter {
+    pub fn new() -> Self {
+        Emitter {
+            frames: vec![Vec::new()],
+        }
+    }
+
+    pub fn push_str(&mut self, s: &str) {
+        if s.is_empty() {
+            return;
+        }
+        let frame = self.frames.last_mut().expect("emitter frame");
+        if let Some(Segment::Html(last)) = frame.last_mut() {
+            last.push_str(s);
+        } else {
+            frame.push(Segment::Html(s.to_string()));
+        }
+    }
+
+    pub fn push(&mut self, c: char) {
+        let mut buf = [0u8; 4];
+        self.push_str(c.encode_utf8(&mut buf));
+    }
+
+    /// Start capturing output into a sub-frame.
+    pub fn begin_frame(&mut self) {
+        self.frames.push(Vec::new());
+    }
+
+    /// Finish the current sub-frame and return what it captured.
+    pub fn end_frame(&mut self) -> Vec<Segment> {
+        debug_assert!(self.frames.len() > 1, "unbalanced emitter frame");
+        self.frames.pop().unwrap_or_default()
+    }
+
+    /// Append previously captured segments to the current frame.
+    pub fn append(&mut self, segments: Vec<Segment>) {
+        for seg in segments {
+            match seg {
+                Segment::Html(s) => self.push_str(&s),
+                deferred => self
+                    .frames
+                    .last_mut()
+                    .expect("emitter frame")
+                    .push(deferred),
+            }
+        }
+    }
+
+    pub fn emit_deferred(
+        &mut self,
+        node_type: String,
+        attrs_json: String,
+        child_types: Vec<String>,
+        content: Vec<Segment>,
+    ) {
+        self.frames
+            .last_mut()
+            .expect("emitter frame")
+            .push(Segment::Deferred {
+                node_type,
+                attrs_json,
+                child_types,
+                content,
+            });
+    }
+
+    pub fn into_segments(mut self) -> Vec<Segment> {
+        debug_assert_eq!(self.frames.len(), 1, "unbalanced emitter frame");
+        self.frames.pop().unwrap_or_default()
+    }
+}
+
+/// What flattening produced. Both variants are normal outcomes — `Deferred`
+/// means callback nodes are present and need splicing — so this is an enum
+/// rather than a `Result`.
+pub enum Flattened {
+    Html(String),
+    Deferred(Vec<Segment>),
+}
+
+impl Flattened {
+    /// The finished markup, or `None` when callback nodes still need
+    /// splicing. Handy where the caller knows no callback rules exist.
+    /// (Only test code needs it today, but it's part of the shape the
+    /// extracted crates will expose.)
+    #[cfg_attr(not(test), allow(dead_code))]
+    pub fn into_html(self) -> Option<String> {
+        match self {
+            Flattened::Html(html) => Some(html),
+            Flattened::Deferred(_) => None,
+        }
+    }
+}
+
+/// Join the segments when every one is finished markup, so the common
+/// no-callback path stays a single string and the splicing layer can be
+/// skipped; hand the segments back untouched when callback nodes are present.
+pub fn flatten(segments: Vec<Segment>) -> Flattened {
+    if segments
+        .iter()
+        .any(|s| matches!(s, Segment::Deferred { .. }))
+    {
+        return Flattened::Deferred(segments);
+    }
+    // The merge invariant makes the common case exactly one Html segment;
+    // move it out rather than copying the whole document.
+    let mut out = String::new();
+    for seg in segments {
+        if let Segment::Html(s) = seg {
+            if out.is_empty() {
+                out = s;
+            } else {
+                out.push_str(&s);
+            }
+        }
+    }
+    Flattened::Html(out)
+}
+
+/// A piece of an attribute value or text template: a literal, or a reference
+/// to one of the node's stored attributes.
+pub enum AttrPart {
+    Lit(String),
+    Ref(String),
+}
+
+/// Resolve a lit/ref template against a node's attributes. `None` (attribute
+/// or text skipped) when the resolved value is empty — matching how the
+/// built-in renderers omit absent attributes.
+pub fn resolve_parts<F: Fn(&str) -> Option<String>>(
+    parts: &[AttrPart],
+    lookup: F,
+) -> Option<String> {
+    let mut out = String::new();
+    for part in parts {
+        match part {
+            AttrPart::Lit(s) => out.push_str(s),
+            AttrPart::Ref(name) => {
+                if let Some(v) = lookup(name) {
+                    out.push_str(&v);
+                }
+            }
+        }
+    }
+    if out.is_empty() {
+        None
+    } else {
+        Some(out)
+    }
+}
+
+/// An attribute reference on a node: rules say `:kind`; Lexical stores its own
+/// props as `__kind` — try the raw name first, then prefixed. (ProseMirror
+/// stores attrs bare, so the fallback never fires there.)
+pub fn xml_ref_attr<T: ReadTxn, N: Xml>(txn: &T, node: &N, name: &str) -> Option<String> {
+    let value = |out: Option<Out>| match out {
+        Some(Out::Any(any)) => any_attr_string(&any),
+        _ => None,
+    };
+    value(node.get_attribute(txn, name))
+        .or_else(|| value(node.get_attribute(txn, &format!("__{name}"))))
+}
+
+/// A stored attribute as a string: strings pass through; numbers print
+/// JS-style; bools as true/false. Anything else is None.
+pub fn any_attr_string(any: &Any) -> Option<String> {
+    match any {
+        Any::String(s) => Some(s.to_string()),
+        Any::Number(n) => Some(if n.fract() == 0.0 {
+            format!("{}", *n as i64)
+        } else {
+            format!("{n}")
+        }),
+        Any::BigInt(n) => Some(format!("{n}")),
+        Any::Bool(b) => Some(if *b { "true" } else { "false" }.to_string()),
+        _ => None,
+    }
+}
+
+/// A node's stored attributes as a JSON object, for callback rules. Keys as
+/// stored (`__type` and friends keep their prefix); values via yrs's own JSON
+/// encoding.
+pub fn xml_attrs_json<T: ReadTxn, N: Xml>(txn: &T, node: &N) -> String {
+    let mut out = String::from("{");
+    let mut first = true;
+    for (key, value) in node.attributes(txn) {
+        let Out::Any(any) = value else { continue };
+        if !first {
+            out.push(',');
+        }
+        first = false;
+        out.push_str(&serde_json::to_string(key).unwrap_or_else(|_| "\"\"".into()));
+        out.push(':');
+        let mut v = String::new();
+        any.to_json(&mut v);
+        out.push_str(&v);
+    }
+    out.push('}');
+    out
+}
+
+/// What goes inside a custom node's element.
+#[derive(Clone, Copy, PartialEq)]
+pub enum Content {
+    Blocks,
+    Inline,
+    None,
+}
+
+/// One node rule: markup as data, or a deferral to the caller.
+pub enum NodeRule {
+    /// Render natively: the element, its attribute/text templates, and what
+    /// goes inside it.
+    Declarative {
+        tag: String,
+        void: bool,
+        attrs: Vec<(String, Vec<AttrPart>)>,
+        text: Option<Vec<AttrPart>>,
+        content: Content,
+    },
+    /// Emit a [`Segment::Deferred`] for the caller to fill in; `content` is
+    /// what renders into its children.
+    Callback { content: Content },
+}
+
+/// A custom mark (ProseMirror only): a wrapping tag with attributes read from
+/// the mark's own value map.
+pub struct MarkRule {
+    pub tag: String,
+    pub attrs: Vec<(String, Vec<AttrPart>)>,
+}
+
+pub struct Rules {
+    pub nodes: HashMap<String, NodeRule>,
+    pub marks: HashMap<String, MarkRule>,
+}
+
+/// What a document walk observed about one node type — the facts behind
+/// `Y::Lexical#node_types` / `Y::ProseMirror#node_types`, the discovery aid
+/// for writing rules against a real document.
+#[derive(Default)]
+pub struct TypeInfo {
+    pub count: usize,
+    pub attrs: BTreeSet<String>,
+    pub children: BTreeSet<String>,
+    pub text: bool,
+}
+
+/// Per-type observations, ordered for stable output.
+pub type TypeMap = BTreeMap<String, TypeInfo>;
+
+/// Serialize the observations, annotating each type with what already
+/// handles it (`"rule"`, `"builtin"`, or null — the ones a rule author needs
+/// to cover).
+pub fn type_map_json(map: &TypeMap, handled: impl Fn(&str) -> Option<&'static str>) -> String {
+    let mut root = serde_json::Map::new();
+    for (ty, info) in map {
+        let mut entry = serde_json::Map::new();
+        entry.insert("count".into(), info.count.into());
+        entry.insert(
+            "attrs".into(),
+            info.attrs.iter().cloned().collect::<Vec<_>>().into(),
+        );
+        entry.insert(
+            "children".into(),
+            info.children.iter().cloned().collect::<Vec<_>>().into(),
+        );
+        entry.insert("text".into(), info.text.into());
+        entry.insert(
+            "handled".into(),
+            match handled(ty) {
+                Some(by) => by.into(),
+                None => serde_json::Value::Null,
+            },
+        );
+        root.insert(ty.clone(), entry.into());
+    }
+    serde_json::Value::Object(root).to_string()
+}
+
+impl Rules {
+    pub fn empty() -> Self {
+        Rules {
+            nodes: HashMap::new(),
+            marks: HashMap::new(),
+        }
+    }
+
+    /// Parse the rules JSON (however the caller compiled it). Absent keys
+    /// take their defaults (`void`/`callback` false, `content` inline), so a
+    /// typical document looks like:
+    ///
+    /// ```json
+    /// { "nodes": { "callout": { "tag": "aside",
+    ///                           "attrs": [["class", [{"lit": "callout"}]],
+    ///                                     ["data-kind", [{"ref": "kind"}]]],
+    ///                           "content": "blocks" },
+    ///              "video":   { "callback": true } },
+    ///   "marks": { "comment": { "tag": "span",
+    ///                           "attrs": [["data-id", [{"ref": "id"}]]] } } }
+    /// ```
+    pub fn parse(json: &str) -> Result<Rules, String> {
+        let root: serde_json::Value =
+            serde_json::from_str(json).map_err(|e| format!("invalid rules JSON: {e}"))?;
+        let mut rules = Rules::empty();
+
+        if let Some(nodes) = root.get("nodes").and_then(|v| v.as_object()) {
+            for (name, spec) in nodes {
+                rules
+                    .nodes
+                    .insert(name.clone(), parse_node_rule(name, spec)?);
+            }
+        }
+        if let Some(marks) = root.get("marks").and_then(|v| v.as_object()) {
+            for (name, spec) in marks {
+                rules
+                    .marks
+                    .insert(name.clone(), parse_mark_rule(name, spec)?);
+            }
+        }
+        Ok(rules)
+    }
+}
+
+fn parse_node_rule(name: &str, spec: &serde_json::Value) -> Result<NodeRule, String> {
+    let content = match spec.get("content").and_then(|v| v.as_str()) {
+        Some("blocks") => Content::Blocks,
+        Some("inline") | None => Content::Inline,
+        Some("none") => Content::None,
+        Some(other) => {
+            return Err(format!(
+                "rule for {name:?}: unknown content kind {other:?} (blocks|inline|none)"
+            ))
+        }
+    };
+    if spec
+        .get("callback")
+        .and_then(|v| v.as_bool())
+        .unwrap_or(false)
+    {
+        return Ok(NodeRule::Callback { content });
+    }
+    let Some(tag) = spec.get("tag").and_then(|v| v.as_str()) else {
+        return Err(format!("rule for {name:?} needs a tag (or a callback)"));
+    };
+    Ok(NodeRule::Declarative {
+        tag: tag.to_string(),
+        void: spec.get("void").and_then(|v| v.as_bool()).unwrap_or(false),
+        attrs: parse_attrs(name, spec.get("attrs"))?,
+        text: match spec.get("text") {
+            Some(serde_json::Value::Array(parts)) => Some(parse_parts(name, parts)?),
+            Some(serde_json::Value::Null) | None => None,
+            Some(_) => return Err(format!("rule for {name:?}: text must be a template array")),
+        },
+        content,
+    })
+}
+
+fn parse_mark_rule(name: &str, spec: &serde_json::Value) -> Result<MarkRule, String> {
+    let Some(tag) = spec.get("tag").and_then(|v| v.as_str()) else {
+        return Err(format!("mark rule for {name:?} needs a tag"));
+    };
+    Ok(MarkRule {
+        tag: tag.to_string(),
+        attrs: parse_attrs(name, spec.get("attrs"))?,
+    })
+}
+
+fn parse_attrs(
+    name: &str,
+    attrs: Option<&serde_json::Value>,
+) -> Result<Vec<(String, Vec<AttrPart>)>, String> {
+    let mut out = Vec::new();
+    let entries = match attrs {
+        None | Some(serde_json::Value::Null) => return Ok(out),
+        Some(serde_json::Value::Array(entries)) => entries,
+        Some(_) => {
+            return Err(format!(
+                "rule for {name:?}: attrs must be an array of [name, template] pairs"
+            ))
+        }
+    };
+    for entry in entries {
+        let (Some(attr_name), Some(serde_json::Value::Array(parts))) =
+            (entry.get(0).and_then(|v| v.as_str()), entry.get(1))
+        else {
+            return Err(format!("rule for {name:?}: malformed attrs entry"));
+        };
+        out.push((attr_name.to_string(), parse_parts(name, parts)?));
+    }
+    Ok(out)
+}
+
+fn parse_parts(name: &str, parts: &[serde_json::Value]) -> Result<Vec<AttrPart>, String> {
+    parts
+        .iter()
+        .map(|part| {
+            if let Some(lit) = part.get("lit").and_then(|v| v.as_str()) {
+                Ok(AttrPart::Lit(lit.to_string()))
+            } else if let Some(r) = part.get("ref").and_then(|v| v.as_str()) {
+                Ok(AttrPart::Ref(r.to_string()))
+            } else {
+                Err(format!(
+                    "rule for {name:?}: template part must be lit or ref"
+                ))
+            }
+        })
+        .collect()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn parses_the_compiled_rule_shape() {
+        let rules = Rules::parse(
+            r#"{ "nodes": { "callout": { "tag": "aside",
+                                         "attrs": [["class", [{"lit": "callout"}]],
+                                                   ["data-kind", [{"ref": "kind"}]]],
+                                         "content": "blocks" },
+                            "video": { "callback": true } },
+                 "marks": { "comment": { "tag": "span",
+                                         "attrs": [["data-id", [{"ref": "id"}]]] } } }"#,
+        )
+        .unwrap();
+        assert_eq!(rules.nodes.len(), 2);
+        let NodeRule::Declarative {
+            tag,
+            attrs,
+            content,
+            ..
+        } = &rules.nodes["callout"]
+        else {
+            panic!("callout should be declarative");
+        };
+        assert_eq!(tag, "aside");
+        assert!(matches!(content, Content::Blocks));
+        assert_eq!(attrs.len(), 2);
+        assert!(matches!(rules.nodes["video"], NodeRule::Callback { .. }));
+        assert_eq!(rules.marks["comment"].tag, "span");
+    }
+
+    #[test]
+    fn rejects_malformed_rules_loudly() {
+        assert!(Rules::parse("not json").is_err());
+        assert!(Rules::parse(r#"{ "nodes": { "x": {} } }"#).is_err()); // no tag, no callback
+        assert!(Rules::parse(r#"{ "nodes": { "x": { "tag": "a", "content": "wat" } } }"#).is_err());
+        assert!(Rules::parse(r#"{ "marks": { "x": {} } }"#).is_err());
+        // attrs present but not the array-of-pairs form must fail loudly,
+        // not silently drop the attributes.
+        assert!(
+            Rules::parse(r#"{ "nodes": { "x": { "tag": "a", "attrs": {"class": "y"} } } }"#)
+                .is_err()
+        );
+    }
+
+    #[test]
+    fn emitter_frames_capture_and_merge() {
+        let mut em = Emitter::new();
+        em.push_str("<p>");
+        em.begin_frame();
+        em.push_str("inner");
+        let captured = em.end_frame();
+        em.emit_deferred("video".into(), "{}".into(), Vec::new(), captured);
+        em.push_str("</p>");
+        let segs = em.into_segments();
+        assert_eq!(segs.len(), 3);
+        assert!(matches!(&segs[0], Segment::Html(s) if s == "<p>"));
+        assert!(matches!(&segs[1], Segment::Deferred { node_type, .. } if node_type == "video"));
+        assert!(matches!(&segs[2], Segment::Html(s) if s == "</p>"));
+
+        // Adjacent Html merges; flatten() hands deferred segments back.
+        let mut em = Emitter::new();
+        em.push_str("a");
+        em.push_str("b");
+        let segs = em.into_segments();
+        assert_eq!(segs.len(), 1);
+        assert_eq!(flatten(segs).into_html().unwrap(), "ab");
+    }
+}

data/lib/y/lexxy.rb

@@ -0,0 +1,100 @@
+# frozen_string_literal: true
+
+module Y
+  # The Lexxy renderer: Y::Lexical (core Lexical) plus the Lexxy-specific
+  # schema, applied beneath the app's rules — an app rule for one of these
+  # types simply replaces it. This is the byte-parity class: the fixture
+  # tests hold `Y::Lexxy.new(doc).to_html` identical to a live editor's own
+  # serialized value.
+  #
+  # The schema doubles as the reference for augmenting a renderer: simple
+  # nodes are declarative hashes, nodes with logic are plain methods mapped
+  # in NODES.
+  class Lexxy < Lexical
+    # A cursor-placement placeholder; empty ones export to nothing.
+    def self.provisional_paragraph(node)
+      node.content.empty? ? "" : "<p>#{node.content}</p>"
+    end
+
+    # Adjacent previewable images; the class carries the image count
+    # (ActionText's convention).
+    def self.gallery(node)
+      %(<div class="attachment-gallery attachment-gallery--#{node.child_types.length}">#{node.content}</div>)
+    end
+
+    # Lexxy wraps tables in a styled figure.
+    def self.table(node)
+      %(<figure class="lexxy-content__table-wrapper"><table><tbody>#{node.content}</tbody></table></figure>)
+    end
+
+    # Class + background match Lexxy's own header-cell export.
+    def self.table_cell(node)
+      header = node.attrs["__headerState"].is_a?(Numeric) && node.attrs["__headerState"].positive?
+      return "<td>#{node.content}</td>" unless header
+
+      style = %(style="background-color: rgb(242, 243, 245);")
+      %(<th class="lexxy-content__table-cell--header" #{style}>#{node.content}</th>)
+    end
+
+    # Attribute order follows Lexxy's export — checked items put aria-checked
+    # before value; items holding a nested list append the
+    # lexxy-nested-listitem class after value.
+    def self.list_item(node)
+      out = +"<li"
+      checked = node.attrs["__checked"]
+      out << %( aria-checked="#{checked}") unless checked.nil?
+      out << %( value="#{node.attrs["__value"] || 1}")
+      out << %( class="lexxy-nested-listitem") if node.child_types.include?("list")
+      "#{out}>#{node.content}</li>"
+    end
+
+    # An upload, in the exact shape ActionText round-trips: attribute order
+    # and presence mirror Lexxy's exportDOM (nulls omitted, `previewable`
+    # only when true, `presentation="gallery"` always).
+    def self.upload(node)
+      out = +"<action-text-attachment"
+      out << attachment_attr(node, "sgid", "sgid")
+      out << %( previewable="true") if node.attrs["previewable"] == true
+      [%w[url src], %w[alt altText], %w[caption caption],
+       %w[content-type contentType], %w[filename fileName],
+       %w[filesize fileSize], %w[width width], %w[height height]]
+        .each { |html_name, stored| out << attachment_attr(node, html_name, stored) }
+      %(#{out} presentation="gallery"></action-text-attachment>)
+    end
+
+    # A content attachment (mention, embed): `content` carries the escaped
+    # inner HTML; `plainText` is not exported.
+    def self.mention(node)
+      out = +"<action-text-attachment"
+      out << attachment_attr(node, "sgid", "sgid")
+      out << attachment_attr(node, "content", "innerHtml")
+      out << attachment_attr(node, "content-type", "contentType")
+      "#{out}></action-text-attachment>"
+    end
+
+    # A stored nil (unset) is skipped; a stored empty string still emits.
+    def self.attachment_attr(node, html_name, stored)
+      return "" if node.attrs[stored].nil?
+
+      %( #{html_name}="#{RenderRules.escape_attr(node.attrs[stored])}")
+    end
+
+    NODES = {
+      # Lexxy's replacement for Lexical's CodeNode.
+      "early_escape_code" => { tag: "pre", attrs: { "data-language" => :language } },
+      "horizontal_divider" => { tag: "hr", void: true },
+      "provisonal_paragraph" => method(:provisional_paragraph), # (sic: Lexxy's spelling)
+      "image_gallery" => method(:gallery),
+      "table" => { contains: :blocks, render: method(:table) },
+      "wrapped_table_node" => { contains: :blocks, render: method(:table) },
+      "tablecell" => { contains: :blocks, render: method(:table_cell) },
+      "listitem" => { contains: :blocks, render: method(:list_item) },
+      "action_text_attachment" => method(:upload),
+      "custom_action_text_attachment" => method(:mention)
+    }.freeze
+
+    def initialize(doc, nodes: {}, &)
+      super(doc, nodes: NODES.merge(nodes.transform_keys(&:to_s)), &)
+    end
+  end
+end