phase 6: LSP foundation — didOpen/didChange, diagnostics, outline

`tower-lsp` server (`nuwiki-lsp::run` + `run_stdio`) wired up by the
`nuwiki-ls` binary. Backend holds an `Arc<DashMap<Url, DocumentState>>`
per SPEC §6.10 plus an `Arc<SyntaxRegistry>` pre-loaded with
`VimwikiSyntax`. Re-parse is full per change; incremental parsing
deferred post-v1 per spec.

Capabilities declared:
- `text_document_sync = FULL`
- `document_symbol_provider = true`
- `position_encoding = UTF-8` when the client opts into LSP 3.17+
  encodings (SPEC §6.11). Otherwise we keep the legacy UTF-16 default
  and translate spans through a per-line UTF-16 code-unit walk that
  handles BMP and surrogate pairs alike.

Handlers: `initialize` (negotiate encoding), `initialized` (log it),
`shutdown`, `did_open` / `did_change` (full re-parse +
`publishDiagnostics`), `did_close` (drop state, clear diagnostics),
`document_symbol` (nested outline).

Diagnostics: every `BlockNode::Error` (from §6.7's resilient parser)
emits one `DiagnosticSeverity::ERROR` with `source = "nuwiki"`. The
walker descends into Blockquotes and ListItems so errors nested inside
those don't slip through.

Document outline: nested `DocumentSymbol` tree built recursively from
heading levels (`= H1 =` → root, `== H2 ==` → child of preceding H1,
etc.). Titles flatten inline markers via `inline_to_text` so
`*world*` shows up as `world` in the outline.

Pure helpers (`to_lsp_position`, `to_lsp_range`, `ast_diagnostics`,
`headings_to_symbols`) are `pub` so the integration tests can verify
conversion correctness without spinning up the async server. End-to-
end LSP message exercises will land in Phase 8 alongside navigation.

Tests (11): UTF-8 passthrough, UTF-16 conversion for é + 🦀, single
and multi-error diagnostic emission, blockquote descent, flat outline,
nested outline by levels, inline-marker stripping in titles, empty-
heading fallback, registry-dispatch parity.

Dep pin: `idna_adapter` held at 1.2.0 — 1.2.2 needs Cargo's
edition2024 feature (Rust 1.85), our MSRV is 1.83.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

crates/nuwiki-lsp/tests/lsp_helpers.rs

@@ -0,0 +1,255 @@
+//! Tests for the pure helpers that the LSP backend uses to translate
+//! between `nuwiki-core` ASTs and the LSP wire format. The async
+//! request-handling loop is exercised at integration time in Phase 8+
+//! (navigation), so this file stays focused on conversion correctness.
+
+use nuwiki_core::ast::{
+    inline::InlineNode, BlockNode, BlockquoteNode, DocumentNode, ErrorNode, ParagraphNode,
+    Position, Span, TextNode,
+};
+use nuwiki_core::syntax::vimwiki::VimwikiSyntax;
+use nuwiki_core::syntax::SyntaxPlugin;
+use nuwiki_lsp::{ast_diagnostics, headings_to_symbols, to_lsp_position, to_lsp_range};
+use tower_lsp::lsp_types::{DiagnosticSeverity, DocumentSymbolResponse, SymbolKind};
+
+// ===== Position / Range conversion =====
+
+#[test]
+fn position_passthrough_in_utf8_mode() {
+    let pos = Position {
+        line: 3,
+        column: 7,
+        offset: 0,
+    };
+    let lsp = to_lsp_position(&pos, "ignored", true);
+    assert_eq!(lsp.line, 3);
+    assert_eq!(lsp.character, 7);
+}
+
+#[test]
+fn position_translates_to_utf16_for_multibyte_chars() {
+    // "héllo" in UTF-8: h(1) é(2 bytes) l(1) l(1) o(1) — total 6 bytes.
+    // In UTF-16 each char is one code unit (BMP), so byte col 3 ("after é")
+    // maps to UTF-16 char col 2.
+    let line = "héllo\nworld\n";
+    let pos = Position {
+        line: 0,
+        column: 3,
+        offset: 3,
+    };
+    let lsp = to_lsp_position(&pos, line, false);
+    assert_eq!(lsp.line, 0);
+    assert_eq!(lsp.character, 2);
+}
+
+#[test]
+fn position_handles_astral_plane_in_utf16() {
+    // 🦀 is U+1F980, two UTF-16 code units (surrogate pair), four UTF-8 bytes.
+    let text = "🦀x\n";
+    // After the crab + x, byte col is 5, UTF-16 col is 3 (2 surrogates + 1 BMP).
+    let pos = Position {
+        line: 0,
+        column: 5,
+        offset: 5,
+    };
+    let lsp = to_lsp_position(&pos, text, false);
+    assert_eq!(lsp.character, 3);
+}
+
+#[test]
+fn range_passthrough_in_utf8_mode() {
+    let span = Span::new(
+        Position {
+            line: 0,
+            column: 0,
+            offset: 0,
+        },
+        Position {
+            line: 0,
+            column: 5,
+            offset: 5,
+        },
+    );
+    let r = to_lsp_range(&span, "hello\n", true);
+    assert_eq!(r.start.line, 0);
+    assert_eq!(r.start.character, 0);
+    assert_eq!(r.end.character, 5);
+}
+
+// ===== Diagnostics =====
+
+#[test]
+fn ast_diagnostics_yields_one_per_error_node() {
+    let span_a = Span::new(
+        Position {
+            line: 0,
+            column: 0,
+            offset: 0,
+        },
+        Position {
+            line: 0,
+            column: 4,
+            offset: 4,
+        },
+    );
+    let span_b = Span::new(
+        Position {
+            line: 2,
+            column: 0,
+            offset: 10,
+        },
+        Position {
+            line: 2,
+            column: 3,
+            offset: 13,
+        },
+    );
+    let doc = DocumentNode {
+        span: Span::default(),
+        metadata: Default::default(),
+        children: vec![
+            BlockNode::Error(ErrorNode {
+                span: span_a,
+                raw: "first".into(),
+                message: "first parse failure".into(),
+            }),
+            BlockNode::Paragraph(ParagraphNode {
+                span: Span::default(),
+                children: vec![InlineNode::Text(TextNode {
+                    span: Span::default(),
+                    content: "ok".into(),
+                })],
+            }),
+            BlockNode::Error(ErrorNode {
+                span: span_b,
+                raw: "second".into(),
+                message: "second parse failure".into(),
+            }),
+        ],
+    };
+
+    let diags = ast_diagnostics(&doc, "ignored", true);
+    assert_eq!(diags.len(), 2);
+    assert_eq!(diags[0].message, "first parse failure");
+    assert_eq!(diags[0].severity, Some(DiagnosticSeverity::ERROR));
+    assert_eq!(diags[0].source.as_deref(), Some("nuwiki"));
+    assert_eq!(diags[0].range.start.line, 0);
+    assert_eq!(diags[1].range.start.line, 2);
+}
+
+#[test]
+fn ast_diagnostics_descends_into_blockquotes() {
+    let span = Span::new(
+        Position {
+            line: 1,
+            column: 2,
+            offset: 0,
+        },
+        Position {
+            line: 1,
+            column: 6,
+            offset: 4,
+        },
+    );
+    let inner_error = BlockNode::Error(ErrorNode {
+        span,
+        raw: "x".into(),
+        message: "nested".into(),
+    });
+    let bq = BlockNode::Blockquote(BlockquoteNode {
+        span: Span::default(),
+        children: vec![inner_error],
+    });
+    let doc = DocumentNode {
+        span: Span::default(),
+        metadata: Default::default(),
+        children: vec![bq],
+    };
+
+    let diags = ast_diagnostics(&doc, "", true);
+    assert_eq!(diags.len(), 1);
+    assert_eq!(diags[0].message, "nested");
+}
+
+// ===== Document symbols =====
+
+fn parse(src: &str) -> DocumentNode {
+    VimwikiSyntax::new().parse(src)
+}
+
+#[test]
+fn flat_outline_when_all_headings_same_level() {
+    let doc = parse("= A =\n= B =\n= C =\n");
+    let syms = headings_to_symbols(&doc, "", true);
+    assert_eq!(syms.len(), 3);
+    let names: Vec<&str> = syms.iter().map(|s| s.name.as_str()).collect();
+    assert_eq!(names, vec!["A", "B", "C"]);
+    for s in &syms {
+        assert!(s.children.is_none() || s.children.as_ref().unwrap().is_empty());
+        assert_eq!(s.kind, SymbolKind::STRING);
+    }
+}
+
+#[test]
+fn nested_outline_follows_heading_levels() {
+    let src = "\
+= Top =
+== Sub A ==
+=== Deeper ===
+== Sub B ==
+= Other =
+";
+    let doc = parse(src);
+    let syms = headings_to_symbols(&doc, "", true);
+    assert_eq!(syms.len(), 2);
+
+    // Top contains Sub A and Sub B; Sub A contains Deeper.
+    let top = &syms[0];
+    assert_eq!(top.name, "Top");
+    let top_kids = top.children.as_ref().expect("Top should have children");
+    assert_eq!(top_kids.len(), 2);
+    assert_eq!(top_kids[0].name, "Sub A");
+    assert_eq!(top_kids[1].name, "Sub B");
+
+    let sub_a_kids = top_kids[0]
+        .children
+        .as_ref()
+        .expect("Sub A should have children");
+    assert_eq!(sub_a_kids.len(), 1);
+    assert_eq!(sub_a_kids[0].name, "Deeper");
+
+    assert_eq!(syms[1].name, "Other");
+}
+
+#[test]
+fn heading_title_strips_inline_markers() {
+    let doc = parse("= Hello *world* and `code` =\n");
+    let syms = headings_to_symbols(&doc, "", true);
+    assert_eq!(syms.len(), 1);
+    assert_eq!(syms[0].name, "Hello world and code");
+}
+
+#[test]
+fn empty_heading_falls_back_to_placeholder_name() {
+    let doc = parse("=  =\n");
+    let syms = headings_to_symbols(&doc, "", true);
+    if let Some(s) = syms.first() {
+        assert!(!s.name.is_empty());
+    }
+}
+
+// ===== End-to-end through the registry =====
+
+#[test]
+fn outline_symbols_match_registry_dispatch() {
+    let src = "= A =\n== B ==\n";
+    let plugin = VimwikiSyntax::new();
+    let doc = plugin.parse(src);
+    let syms = headings_to_symbols(&doc, src, true);
+    assert_eq!(syms.len(), 1);
+    assert_eq!(syms[0].name, "A");
+    let kids = syms[0].children.as_ref().unwrap();
+    assert_eq!(kids[0].name, "B");
+    // Sanity: symbols can be wrapped in DocumentSymbolResponse::Nested.
+    let _resp = DocumentSymbolResponse::Nested(syms);
+}