@ai-matrx/content-ir 0.19.12 → 0.19.14

package/CHANGELOG.md

@@ -1,5 +1,17 @@
 # Changelog
 
+## 0.19.14
+
+Automatic changed-only republish (docs/metadata drift since the last tag — see
+`git diff npm/content-ir-core/v0.19.12..npm/content-ir-core/v0.19.14 -- apps/shared/content-ir-core`).
+No source changes intended and no consumer action required.
+
+## 0.19.13
+
+Automatic changed-only republish (docs/metadata drift since the last tag — see
+`git diff npm/content-ir-core/v0.19.12..npm/content-ir-core/v0.19.13 -- apps/shared/content-ir-core`).
+No source changes intended and no consumer action required.
+
 ## 0.19.12
 
 Automatic changed-only republish (docs/metadata drift since the last tag — see
@@ -21,6 +33,7 @@ No source changes intended and no consumer action required.
 ## 0.19.9
 ### Unreleased
 
+- `/source` **`isHtmlBlockTagName(name)`** (verify-RC-B4 round 9): the tag names CommonMark reads as an HTML BLOCK when they open a line (conditions 1 and 6: `div`, `details`, `section`, `script`, …; case-insensitive) — so a splitter can leave such a line to the markdown renderer, ending where GFM ends it, instead of opening a custom XML container that owns what follows.
 - `/source` **A table ends where GFM ends it (verify-RC-B4 round 9):** `continuesTable` / `findTableEnd` end a table at an HTML block start (all seven CommonMark 4.6 conditions — `<script`/`<pre`/`<style`/`<textarea`, `<!--`, `<?`, `<!X`, `<![CDATA[`, a block tag name, or any other complete tag alone on its line such as `</artifact>`) and at indented code: a line 4+ columns past the table's container. New `startsHtmlBlock`, `lineIndent`, `tableContainerIndent` (the content column of the list item a table sits in); `continuesTable(line, containerIndent = 0)`. A lone `|` is a row (one empty cell) — `rowCells("|")` is `[""]`, no longer `[]`. Since 0.19.4 an HTML line or an indented line after a table was read as a table row. Shared vectors gain `tableEnds` (30 cases, each judged against remark-gfm in matrx-frontend); the Python twin moved with them.
 - `/source` **A table never opens on a lazy continuation line (verify-RC-B4 round 9):** `opensTable` / `tableStartsAt` / `findTableStart` refuse a header that lazily continues a list item's or a block quote's paragraph (no blank line above, not indented to the item's content column, or missing the quote's `>`) — GFM reads it as paragraph text (`- item` then `| a | b |` is no table). Vectors gain `tableStarts` (15 cases, judged against remark-gfm).
 - `/source` **A table opens exactly where GFM opens one (verify-RC-B4 round 9):** `tableStartsAt` requires the delimiter row to match the header's width for pipe-led headers too (`| a | b |` over `|---|` is no table), refuses a "delimiter" that is a list item (`- | -`), and refuses a header indented 4+ columns past its container (indented code). `startsPipelessTable` accepts a header with one pipe (`| a`) or none (`Notes` over `|---|` is a one-column table), as GFM does. Vectors gain 8 opening cases.

package/dist/index.cjs

@@ -7665,6 +7665,51 @@ function mapRange(changes, start, end, unit = "utf16") {
   return { start: newStart, end: newEnd, touched, collapsed: newEnd === newStart && end > start };
 }
 
+// source/link-references.ts
+function normalizeLinkLabel(label) {
+  return label.trim().replace(/\s+/g, " ").toLowerCase();
+}
+var DEFINITION = /^ {0,3}\[((?:[^\]\\]|\\.){1,999})\]:[ \t]*(?:<([^<>\n]*)>|(\S+))(?:[ \t]+(?:"((?:[^"\\]|\\.)*)"|'((?:[^'\\]|\\.)*)'|\(((?:[^()\\]|\\.)*)\)))?[ \t]*$/;
+function collectLinkDefinitions(text) {
+  const defs = /* @__PURE__ */ new Map();
+  if (!text.includes("]:")) return defs;
+  const lines = text.split("\n");
+  const kinds = fenceLineKinds(text);
+  lines.forEach((line, i) => {
+    if (kinds[i] !== "prose") return;
+    const m = DEFINITION.exec(line);
+    if (!m) return;
+    const label = m[1] ?? "";
+    if (label.startsWith("^")) return;
+    const key = normalizeLinkLabel(label);
+    if (!key || defs.has(key)) return;
+    defs.set(key, { url: m[2] ?? m[3] ?? "", title: m[4] ?? m[5] ?? m[6] ?? null });
+  });
+  return defs;
+}
+var REFERENCE = /(!?)\[((?:[^\]\\]|\\.)*)\](?:\[((?:[^\]\\]|\\.)*)\])?(?![(:[])/g;
+var escapeTitle = (title) => title.replace(/["\\]/g, "\\$&");
+function resolveReferenceLinks(fragment, defs) {
+  if (defs.size === 0 || !fragment.includes("[")) return fragment;
+  const PLACEHOLDER = "\uE011";
+  const spans = [];
+  const masked = mapInlineCodeSpans(fragment, (_range, raw) => {
+    spans.push(raw);
+    return `${PLACEHOLDER}${spans.length - 1}${PLACEHOLDER}`;
+  });
+  const resolved = masked.replace(REFERENCE, (whole, bang, text, ref, at) => {
+    if (at > 0 && masked[at - 1] === "\\") return whole;
+    if (text.startsWith("^")) return whole;
+    const key = normalizeLinkLabel(ref === void 0 || ref === "" ? text : ref);
+    const def = defs.get(key);
+    if (!def) return whole;
+    const title = def.title === null ? "" : ` "${escapeTitle(def.title)}"`;
+    const url = /[\s()<>]/.test(def.url) ? `<${def.url}>` : def.url;
+    return `${bang}[${text}](${url}${title})`;
+  });
+  return resolved.replace(new RegExp(`${PLACEHOLDER}(\\d+)${PLACEHOLDER}`, "g"), (_m, i) => spans[Number(i)] ?? "");
+}
+
 // source/gfm-table.ts
 function splitRowSegments(line) {
   const segments = [];
@@ -7693,7 +7738,7 @@ function rowCells(line) {
 function unescapeCellPipes(cell) {
   return cell.replace(/\\\|/g, "|");
 }
-var BLOCK_START = /^(?:[-+*](?:\s|$)|\d{1,9}[.)](?:\s|$)|>|#{1,6}(?:\s|$)|(?:-\s*){3,}$|(?:\*\s*){3,}$|(?:_\s*){3,}$)/;
+var BLOCK_START = /^(?:[-+*](?:\s|$)|\d{1,9}[.)](?:\s|$)|>|#{1,6}(?:\s|$)|(?:-\s*){3,}$|(?:\*\s*){3,}$|(?:_\s*){3,}$|\[\^[^\]\s]+\]:)/;
 var startsBlock = (trimmed) => BLOCK_START.test(trimmed) || opensStrictFence(trimmed) || startsHtmlBlock(trimmed);
 var HTML_RAW = /^<(?:script|pre|style|textarea)(?:[\s>]|$)/i;
 var HTML_COMMENT_ETC = /^(?:<!--|<\?|<![A-Za-z]|<!\[CDATA\[)/;
@@ -7783,11 +7828,45 @@ function continuesTable(line, containerIndent = 0) {
 function opensTable(lines, index) {
   const line = lines[index] ?? "";
   if (!isPipeLedRow(line) && !startsPipelessTable(line, lines[index + 1])) return false;
-  return !isLazyContinuation(lines, index);
+  return !isLazyContinuation(lines, index) && !insideHtmlBlock(lines, index);
 }
 var QUOTE_LINE = /^ {0,3}>/;
 var THEMATIC_BREAK = /^(?:(?:-[ \t]*){3,}|(?:\*[ \t]*){3,}|(?:_[ \t]*){3,})$/;
 var breaksParagraph = (trimmed) => /^#{1,6}(?:\s|$)/.test(trimmed) || THEMATIC_BREAK.test(trimmed) || opensStrictFence(trimmed) || startsHtmlBlock(trimmed);
+function htmlBlockEnd(trimmed) {
+  if (HTML_RAW.test(trimmed)) return /<\/(?:script|pre|style|textarea)>/i;
+  if (trimmed.startsWith("<!--")) return /-->/;
+  if (trimmed.startsWith("<?")) return /\?>/;
+  if (trimmed.startsWith("<![CDATA[")) return /\]\]>/;
+  if (/^<![A-Za-z]/.test(trimmed)) return />/;
+  if (HTML_BLOCK_TAG.test(trimmed)) return null;
+  if (HTML_COMPLETE_TAG.test(trimmed)) return null;
+  return void 0;
+}
+function insideHtmlBlock(lines, index) {
+  let start = index;
+  while (start > 0 && (lines[start - 1] ?? "").trim()) start -= 1;
+  let open;
+  let paragraph = false;
+  for (let k = start; k < index; k += 1) {
+    const line = lines[k] ?? "";
+    if (open !== void 0) {
+      if (open && open.test(line)) open = void 0;
+      continue;
+    }
+    const trimmed = line.trim();
+    const end = lineIndent(line) < 4 ? htmlBlockEnd(trimmed) : void 0;
+    const type7 = end === null && !HTML_BLOCK_TAG.test(trimmed);
+    const interrupts = end !== void 0 && !(paragraph && type7);
+    if (!interrupts) {
+      paragraph = true;
+      continue;
+    }
+    paragraph = false;
+    open = end && end.test(trimmed.slice(2)) ? void 0 : end;
+  }
+  return open !== void 0;
+}
 function isLazyContinuation(lines, index) {
   const header = lines[index] ?? "";
   let first = -1;
@@ -7876,6 +7955,7 @@ exports.classifyInboundEnvelopeMetadata = classifyInboundEnvelopeMetadata;
 exports.classifyInnerFenceLine = classifyInnerFenceLine;
 exports.closeFence = closeFence;
 exports.codeSpanText = codeSpanText;
+exports.collectLinkDefinitions = collectLinkDefinitions;
 exports.collectReferencedKinds = collectReferencedKinds;
 exports.collectSchemaReferencedKinds = collectSchemaReferencedKinds;
 exports.compareWithExistingKindSchema = compareWithExistingKindSchema;
@@ -7955,6 +8035,7 @@ exports.mapRange = mapRange;
 exports.mergeResidueIntoValue = mergeResidueIntoValue;
 exports.normalizeAiSchemaInput = normalizeAiSchemaInput;
 exports.normalizeJsonRegion = normalizeJsonRegion;
+exports.normalizeLinkLabel = normalizeLinkLabel;
 exports.nounFamily = nounFamily;
 exports.nounLabel = nounLabel;
 exports.nounTitleColumn = nounTitleColumn;
@@ -7975,6 +8056,7 @@ exports.reconstructRegionValue = reconstructRegionValue;
 exports.rehydrateNodeOutcome = rehydrateNodeOutcome;
 exports.rehydrateRunResult = rehydrateRunResult;
 exports.resetLegacyShellUses = resetLegacyShellUses;
+exports.resolveReferenceLinks = resolveReferenceLinks;
 exports.resolvesInContent = resolvesInContent;
 exports.reuseEnvelopeIfCurrent = reuseEnvelopeIfCurrent;
 exports.rowCells = rowCells;