castdown-cleaners 0.1.0

package/dist/regex/html-tables-to-gfm.d.ts

@@ -0,0 +1,21 @@
+/**
+ * htmlTablesToGfm — convert HTML `<table>` blocks to GFM pipe tables.
+ *
+ * MarkItDown emits HTML tables for DOCX/XLSX/PPTX/PDF sources. GFM viewers
+ * and downstream Pandoc/Typst templates expect pipe-style tables, so HTML
+ * tables silently break the pipeline. Run BEFORE `fixTables` so the rebuilt
+ * pipe tables get normalized.
+ *
+ * Strategy (regex, no DOM):
+ *   - Find each `<table …>…</table>` block (case-insensitive, multiline).
+ *   - Extract `<tr>` rows; per row extract `<th>`/`<td>` cells.
+ *   - First row carrying `<th>` (else first row) → header.
+ *   - Inside cells: strip inline tags, collapse whitespace, escape pipes,
+ *     map `<br>` to a space.
+ *   - rowspan/colspan flattened (GFM has no equivalent). Cells padded to
+ *     max column count.
+ *
+ * Idempotent on already-pipe MD (no `<table>` ⇒ no-op).
+ */
+export declare function htmlTablesToGfm(md: string): string;
+//# sourceMappingURL=html-tables-to-gfm.d.ts.map

package/dist/regex/html-tables-to-gfm.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"html-tables-to-gfm.d.ts","sourceRoot":"","sources":["../../src/regex/html-tables-to-gfm.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;GAkBG;AACH,wBAAgB,eAAe,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAwBlD"}

package/dist/regex/html-tables-to-gfm.js

@@ -0,0 +1,76 @@
+/**
+ * htmlTablesToGfm — convert HTML `<table>` blocks to GFM pipe tables.
+ *
+ * MarkItDown emits HTML tables for DOCX/XLSX/PPTX/PDF sources. GFM viewers
+ * and downstream Pandoc/Typst templates expect pipe-style tables, so HTML
+ * tables silently break the pipeline. Run BEFORE `fixTables` so the rebuilt
+ * pipe tables get normalized.
+ *
+ * Strategy (regex, no DOM):
+ *   - Find each `<table …>…</table>` block (case-insensitive, multiline).
+ *   - Extract `<tr>` rows; per row extract `<th>`/`<td>` cells.
+ *   - First row carrying `<th>` (else first row) → header.
+ *   - Inside cells: strip inline tags, collapse whitespace, escape pipes,
+ *     map `<br>` to a space.
+ *   - rowspan/colspan flattened (GFM has no equivalent). Cells padded to
+ *     max column count.
+ *
+ * Idempotent on already-pipe MD (no `<table>` ⇒ no-op).
+ */
+export function htmlTablesToGfm(md) {
+    const TABLE_RE = /<table\b[^>]*>([\s\S]*?)<\/table\s*>/gi;
+    return md.replace(TABLE_RE, (_full, inner) => {
+        const rows = extractRows(inner);
+        if (rows.length === 0)
+            return _full;
+        const headerIdx = rows.findIndex((r) => r.isHeader);
+        const headerRowIdx = headerIdx >= 0 ? headerIdx : 0;
+        const header = rows[headerRowIdx].cells;
+        const body = rows.filter((_, i) => i !== headerRowIdx).map((r) => r.cells);
+        const maxCols = Math.max(header.length, ...body.map((r) => r.length), 1);
+        const pad = (r) => {
+            const c = [...r];
+            while (c.length < maxCols)
+                c.push("");
+            return c;
+        };
+        const lines = [];
+        lines.push(renderRow(pad(header)));
+        lines.push(renderRow(Array(maxCols).fill("---")));
+        for (const r of body)
+            lines.push(renderRow(pad(r)));
+        return "\n" + lines.join("\n") + "\n";
+    });
+}
+function extractRows(inner) {
+    const out = [];
+    const TR_RE = /<tr\b[^>]*>([\s\S]*?)<\/tr\s*>/gi;
+    let m;
+    while ((m = TR_RE.exec(inner)) !== null) {
+        const body = m[1] ?? "";
+        const cells = [];
+        let isHeader = false;
+        const CELL_RE = /<(th|td)\b[^>]*>([\s\S]*?)<\/\1\s*>/gi;
+        let c;
+        while ((c = CELL_RE.exec(body)) !== null) {
+            if ((c[1] ?? "").toLowerCase() === "th")
+                isHeader = true;
+            cells.push(cleanCell(c[2] ?? ""));
+        }
+        if (cells.length > 0)
+            out.push({ cells, isHeader });
+    }
+    return out;
+}
+function cleanCell(raw) {
+    return raw
+        .replace(/<br\s*\/?>/gi, " ")
+        .replace(/<\/?[^>]+>/g, "") // strip remaining inline tags
+        .replace(/\s+/g, " ")
+        .trim()
+        .replace(/\|/g, "\\|");
+}
+function renderRow(cells) {
+    return "| " + cells.join(" | ") + " |";
+}
+//# sourceMappingURL=html-tables-to-gfm.js.map

package/dist/regex/html-tables-to-gfm.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"html-tables-to-gfm.js","sourceRoot":"","sources":["../../src/regex/html-tables-to-gfm.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;GAkBG;AACH,MAAM,UAAU,eAAe,CAAC,EAAU;IACxC,MAAM,QAAQ,GAAG,wCAAwC,CAAC;IAC1D,OAAO,EAAE,CAAC,OAAO,CAAC,QAAQ,EAAE,CAAC,KAAK,EAAE,KAAa,EAAE,EAAE;QACnD,MAAM,IAAI,GAAG,WAAW,CAAC,KAAK,CAAC,CAAC;QAChC,IAAI,IAAI,CAAC,MAAM,KAAK,CAAC;YAAE,OAAO,KAAe,CAAC;QAE9C,MAAM,SAAS,GAAG,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC;QACpD,MAAM,YAAY,GAAG,SAAS,IAAI,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC;QACpD,MAAM,MAAM,GAAG,IAAI,CAAC,YAAY,CAAE,CAAC,KAAK,CAAC;QACzC,MAAM,IAAI,GAAG,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,KAAK,YAAY,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC;QAE3E,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC,CAAC;QACzE,MAAM,GAAG,GAAG,CAAC,CAAW,EAAE,EAAE;YAC1B,MAAM,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,CAAC;YACjB,OAAO,CAAC,CAAC,MAAM,GAAG,OAAO;gBAAE,CAAC,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;YACtC,OAAO,CAAC,CAAC;QACX,CAAC,CAAC;QAEF,MAAM,KAAK,GAAa,EAAE,CAAC;QAC3B,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;QACnC,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;QAClD,KAAK,MAAM,CAAC,IAAI,IAAI;YAAE,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;QACpD,OAAO,IAAI,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC;IACxC,CAAC,CAAC,CAAC;AACL,CAAC;AAOD,SAAS,WAAW,CAAC,KAAa;IAChC,MAAM,GAAG,GAAU,EAAE,CAAC;IACtB,MAAM,KAAK,GAAG,kCAAkC,CAAC;IACjD,IAAI,CAAyB,CAAC;IAC9B,OAAO,CAAC,CAAC,GAAG,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,KAAK,IAAI,EAAE,CAAC;QACxC,MAAM,IAAI,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QACxB,MAAM,KAAK,GAAa,EAAE,CAAC;QAC3B,IAAI,QAAQ,GAAG,KAAK,CAAC;QACrB,MAAM,OAAO,GAAG,uCAAuC,CAAC;QACxD,IAAI,CAAyB,CAAC;QAC9B,OAAO,CAAC,CAAC,GAAG,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,KAAK,IAAI,EAAE,CAAC;YACzC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,WAAW,EAAE,KAAK,IAAI;gBAAE,QAAQ,GAAG,IAAI,CAAC;YACzD,KAAK,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC;QACpC,CAAC;QACD,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC;YAAE,GAAG,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,QAAQ,EAAE,CAAC,CAAC;IACtD,CAAC;IACD,OAAO,GAAG,CAAC;AACb,CAAC;AAED,SAAS,SAAS,CAAC,GAAW;IAC5B,OAAO,GAAG;SACP,OAAO,CAAC,cAAc,EAAE,GAAG,CAAC;SAC5B,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC,CAAI,8BAA8B;SAC5D,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC;SACpB,IAAI,EAAE;SACN,OAAO,CAAC,KAAK,EAAE,KAAK,CAAC,CAAC;AAC3B,CAAC;AAED,SAAS,SAAS,CAAC,KAAe;IAChC,OAAO,IAAI,GAAG,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC;AACzC,CAAC"}

package/dist/regex/join-broken-lines.d.ts

@@ -0,0 +1,10 @@
+/**
+ * joinBrokenLines — join lines that were broken mid-sentence (PDF column wrap).
+ *
+ * Rule: if line ends with a lowercase letter or comma AND next line starts
+ * with a lowercase letter, treat as single sentence.
+ *
+ * Conservative: skips lines inside fenced code blocks, lists, tables, headings.
+ */
+export declare function joinBrokenLines(md: string): string;
+//# sourceMappingURL=join-broken-lines.d.ts.map

package/dist/regex/join-broken-lines.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"join-broken-lines.d.ts","sourceRoot":"","sources":["../../src/regex/join-broken-lines.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AACH,wBAAgB,eAAe,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAuBlD"}

package/dist/regex/join-broken-lines.js

@@ -0,0 +1,40 @@
+/**
+ * joinBrokenLines — join lines that were broken mid-sentence (PDF column wrap).
+ *
+ * Rule: if line ends with a lowercase letter or comma AND next line starts
+ * with a lowercase letter, treat as single sentence.
+ *
+ * Conservative: skips lines inside fenced code blocks, lists, tables, headings.
+ */
+export function joinBrokenLines(md) {
+    const lines = md.split("\n");
+    const out = [];
+    let inCode = false;
+    for (let i = 0; i < lines.length; i++) {
+        const line = lines[i] ?? "";
+        const next = lines[i + 1] ?? "";
+        if (/^```/.test(line.trim()))
+            inCode = !inCode;
+        if (!inCode &&
+            out.length > 0 &&
+            shouldJoin(out[out.length - 1] ?? "", line)) {
+            out[out.length - 1] = (out[out.length - 1] ?? "").replace(/\s+$/, "") + " " + line.trim();
+            continue;
+        }
+        out.push(line);
+        void next; // lookahead reserved for future heuristics
+    }
+    return out.join("\n");
+}
+function shouldJoin(prev, current) {
+    if (!prev.trim() || !current.trim())
+        return false;
+    if (/^[-*+]\s|^\d+\.\s|^#{1,6}\s|^>\s|^\|/.test(current))
+        return false; // list/heading/quote/table
+    if (/^[-*+]\s|^\d+\.\s|^#{1,6}\s|^>\s|^\|/.test(prev))
+        return false;
+    if (/[.!?:;]$/.test(prev.trim()))
+        return false;
+    return /[a-záéíóúñü,]\s*$/.test(prev) && /^[a-záéíóúñü]/.test(current.trim());
+}
+//# sourceMappingURL=join-broken-lines.js.map

package/dist/regex/join-broken-lines.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"join-broken-lines.js","sourceRoot":"","sources":["../../src/regex/join-broken-lines.ts"],"names":[],"mappings":"AAAA;;;;;;;GAOG;AACH,MAAM,UAAU,eAAe,CAAC,EAAU;IACxC,MAAM,KAAK,GAAG,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAC7B,MAAM,GAAG,GAAa,EAAE,CAAC;IACzB,IAAI,MAAM,GAAG,KAAK,CAAC;IAEnB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACtC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAC5B,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC;QAEhC,IAAI,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;YAAE,MAAM,GAAG,CAAC,MAAM,CAAC;QAE/C,IACE,CAAC,MAAM;YACP,GAAG,CAAC,MAAM,GAAG,CAAC;YACd,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,MAAM,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,IAAI,CAAC,EAC3C,CAAC;YACD,GAAG,CAAC,GAAG,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,GAAG,CAAC,MAAM,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC,GAAG,GAAG,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC;YAC1F,SAAS;QACX,CAAC;QACD,GAAG,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QACf,KAAK,IAAI,CAAC,CAAC,2CAA2C;IACxD,CAAC;IACD,OAAO,GAAG,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AACxB,CAAC;AAED,SAAS,UAAU,CAAC,IAAY,EAAE,OAAe;IAC/C,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,IAAI,CAAC,OAAO,CAAC,IAAI,EAAE;QAAE,OAAO,KAAK,CAAC;IAClD,IAAI,sCAAsC,CAAC,IAAI,CAAC,OAAO,CAAC;QAAE,OAAO,KAAK,CAAC,CAAC,2BAA2B;IACnG,IAAI,sCAAsC,CAAC,IAAI,CAAC,IAAI,CAAC;QAAE,OAAO,KAAK,CAAC;IACpE,IAAI,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;QAAE,OAAO,KAAK,CAAC;IAC/C,OAAO,mBAAmB,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,eAAe,CAAC,IAAI,CAAC,OAAO,CAAC,IAAI,EAAE,CAAC,CAAC;AAChF,CAAC"}

package/dist/regex/join-soft-hyphens.d.ts

@@ -0,0 +1,9 @@
+/**
+ * joinSoftHyphens — rejoin words split by line-end hyphens (common in PDF).
+ *   "exam-\nple"  →  "example"
+ *   "Conway-\nMaxwell"  → "Conway-Maxwell"  (preserves real compound hyphens)
+ *
+ * Heuristic: only join when next char is lowercase letter.
+ */
+export declare function joinSoftHyphens(md: string): string;
+//# sourceMappingURL=join-soft-hyphens.d.ts.map

package/dist/regex/join-soft-hyphens.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"join-soft-hyphens.d.ts","sourceRoot":"","sources":["../../src/regex/join-soft-hyphens.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,wBAAgB,eAAe,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAElD"}

package/dist/regex/join-soft-hyphens.js

@@ -0,0 +1,11 @@
+/**
+ * joinSoftHyphens — rejoin words split by line-end hyphens (common in PDF).
+ *   "exam-\nple"  →  "example"
+ *   "Conway-\nMaxwell"  → "Conway-Maxwell"  (preserves real compound hyphens)
+ *
+ * Heuristic: only join when next char is lowercase letter.
+ */
+export function joinSoftHyphens(md) {
+    return md.replace(/(\w)-\n([a-záéíóúñü])/g, "$1$2");
+}
+//# sourceMappingURL=join-soft-hyphens.js.map

package/dist/regex/join-soft-hyphens.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"join-soft-hyphens.js","sourceRoot":"","sources":["../../src/regex/join-soft-hyphens.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AACH,MAAM,UAAU,eAAe,CAAC,EAAU;IACxC,OAAO,EAAE,CAAC,OAAO,CAAC,wBAAwB,EAAE,MAAM,CAAC,CAAC;AACtD,CAAC"}

package/dist/regex/normalize-horizontal-rules.d.ts

@@ -0,0 +1,2 @@
+export declare function normalizeHorizontalRules(md: string): string;
+//# sourceMappingURL=normalize-horizontal-rules.d.ts.map

package/dist/regex/normalize-horizontal-rules.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-horizontal-rules.d.ts","sourceRoot":"","sources":["../../src/regex/normalize-horizontal-rules.ts"],"names":[],"mappings":"AAKA,wBAAgB,wBAAwB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAc3D"}

package/dist/regex/normalize-horizontal-rules.js

@@ -0,0 +1,20 @@
+import { withProtectedCode } from "../util/protect-code.js";
+const HR_SIMPLE_RE = /^[ \t]*([-=_*~—–])\1{2,}[ \t]*$/;
+const HR_SPACED_RE = /^[ \t]*([-*•—–])(?:[ \t]+\1){2,}[ \t]*$/;
+export function normalizeHorizontalRules(md) {
+    return withProtectedCode(md, (s) => {
+        const lines = s.split("\n");
+        return lines
+            .map((line, i) => {
+            const prev = lines[i - 1] ?? "";
+            const isSetextCandidate = /^[-=]+$/.test(line) && prev.trim().length > 0;
+            if (isSetextCandidate)
+                return line;
+            if (HR_SIMPLE_RE.test(line) || HR_SPACED_RE.test(line))
+                return "---";
+            return line;
+        })
+            .join("\n");
+    });
+}
+//# sourceMappingURL=normalize-horizontal-rules.js.map

package/dist/regex/normalize-horizontal-rules.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-horizontal-rules.js","sourceRoot":"","sources":["../../src/regex/normalize-horizontal-rules.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAE5D,MAAM,YAAY,GAAG,iCAAiC,CAAC;AACvD,MAAM,YAAY,GAAG,yCAAyC,CAAC;AAE/D,MAAM,UAAU,wBAAwB,CAAC,EAAU;IACjD,OAAO,iBAAiB,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,EAAE;QACjC,MAAM,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAC5B,OAAO,KAAK;aACT,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,EAAE,EAAE;YACf,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC;YAChC,MAAM,iBAAiB,GACrB,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC,MAAM,GAAG,CAAC,CAAC;YACjD,IAAI,iBAAiB;gBAAE,OAAO,IAAI,CAAC;YACnC,IAAI,YAAY,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,YAAY,CAAC,IAAI,CAAC,IAAI,CAAC;gBAAE,OAAO,KAAK,CAAC;YACrE,OAAO,IAAI,CAAC;QACd,CAAC,CAAC;aACD,IAAI,CAAC,IAAI,CAAC,CAAC;IAChB,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/regex/normalize-list-markers.d.ts

@@ -0,0 +1,2 @@
+export declare function normalizeListMarkers(md: string): string;
+//# sourceMappingURL=normalize-list-markers.d.ts.map

package/dist/regex/normalize-list-markers.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-list-markers.d.ts","sourceRoot":"","sources":["../../src/regex/normalize-list-markers.ts"],"names":[],"mappings":"AAqBA,wBAAgB,oBAAoB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAgBvD"}

package/dist/regex/normalize-list-markers.js

@@ -0,0 +1,35 @@
+import { withProtectedCode } from "../util/protect-code.js";
+const BULLET_LEVEL = {
+    "•": 0, "▪": 0, "▶": 0, "►": 0, "◆": 0, "❖": 0, "→": 0, "»": 0,
+    "◦": 1, "▫": 1, "▷": 1, "▸": 1, "◇": 1, "○": 1,
+    "‣": 2, "·": 2,
+};
+const CHECK_TRUE_CHARS = new Set(["✓", "✔", "☑"]);
+const CHECK_FALSE_CHARS = new Set(["✗", "✘", "☐"]);
+const ALL_BULLET_CHARS = [
+    ...Object.keys(BULLET_LEVEL),
+    ...CHECK_TRUE_CHARS,
+    ...CHECK_FALSE_CHARS,
+].join("");
+const BULLET_LINE_RE = new RegExp(`^([ \\t]*)([${ALL_BULLET_CHARS}])\\s+(.+)$`);
+export function normalizeListMarkers(md) {
+    return withProtectedCode(md, (s) => {
+        const lines = s.split("\n");
+        return lines
+            .map((line) => {
+            const m = line.match(BULLET_LINE_RE);
+            if (!m)
+                return line;
+            const [, indent, sym, content] = m;
+            if (CHECK_TRUE_CHARS.has(sym))
+                return `${indent}- [x] ${content}`;
+            if (CHECK_FALSE_CHARS.has(sym))
+                return `${indent}- [ ] ${content}`;
+            const level = BULLET_LEVEL[sym] ?? 0;
+            const extra = "  ".repeat(level);
+            return `${indent}${extra}- ${content}`;
+        })
+            .join("\n");
+    });
+}
+//# sourceMappingURL=normalize-list-markers.js.map

package/dist/regex/normalize-list-markers.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-list-markers.js","sourceRoot":"","sources":["../../src/regex/normalize-list-markers.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAE5D,MAAM,YAAY,GAA2B;IAC3C,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC;IAC9D,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC;IAC9C,GAAG,EAAE,CAAC,EAAE,GAAG,EAAE,CAAC;CACf,CAAC;AAEF,MAAM,gBAAgB,GAAG,IAAI,GAAG,CAAC,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC;AAClD,MAAM,iBAAiB,GAAG,IAAI,GAAG,CAAC,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC;AAEnD,MAAM,gBAAgB,GAAG;IACvB,GAAG,MAAM,CAAC,IAAI,CAAC,YAAY,CAAC;IAC5B,GAAG,gBAAgB;IACnB,GAAG,iBAAiB;CACrB,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;AAEX,MAAM,cAAc,GAAG,IAAI,MAAM,CAC/B,eAAe,gBAAgB,aAAa,CAC7C,CAAC;AAEF,MAAM,UAAU,oBAAoB,CAAC,EAAU;IAC7C,OAAO,iBAAiB,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,EAAE;QACjC,MAAM,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAC5B,OAAO,KAAK;aACT,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE;YACZ,MAAM,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,cAAc,CAAC,CAAC;YACrC,IAAI,CAAC,CAAC;gBAAE,OAAO,IAAI,CAAC;YACpB,MAAM,CAAC,EAAE,MAAM,EAAE,GAAG,EAAE,OAAO,CAAC,GAAG,CAAqC,CAAC;YACvE,IAAI,gBAAgB,CAAC,GAAG,CAAC,GAAG,CAAC;gBAAE,OAAO,GAAG,MAAM,SAAS,OAAO,EAAE,CAAC;YAClE,IAAI,iBAAiB,CAAC,GAAG,CAAC,GAAG,CAAC;gBAAE,OAAO,GAAG,MAAM,SAAS,OAAO,EAAE,CAAC;YACnE,MAAM,KAAK,GAAG,YAAY,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;YACrC,MAAM,KAAK,GAAG,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC;YACjC,OAAO,GAAG,MAAM,GAAG,KAAK,KAAK,OAAO,EAAE,CAAC;QACzC,CAAC,CAAC;aACD,IAAI,CAAC,IAAI,CAAC,CAAC;IAChB,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/regex/normalize-numbered-lists.d.ts

@@ -0,0 +1,2 @@
+export declare function normalizeNumberedLists(md: string): string;
+//# sourceMappingURL=normalize-numbered-lists.d.ts.map

package/dist/regex/normalize-numbered-lists.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-numbered-lists.d.ts","sourceRoot":"","sources":["../../src/regex/normalize-numbered-lists.ts"],"names":[],"mappings":"AAEA,wBAAgB,sBAAsB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAQzD"}

package/dist/regex/normalize-numbered-lists.js

@@ -0,0 +1,9 @@
+import { withProtectedCode } from "../util/protect-code.js";
+export function normalizeNumberedLists(md) {
+    return withProtectedCode(md, (s) => s
+        .replace(/^([ \t]*)(\d+)\)(?=[ \t])/gm, "$1$2.")
+        .replace(/^([ \t]*)\((\d+)\)(?=[ \t])/gm, "$1$2.")
+        .replace(/^([ \t]*)([a-z])\)(?=[ \t])/gm, "$1$2.")
+        .replace(/^([ \t]*)\(([a-z])\)(?=[ \t])/gm, "$1$2."));
+}
+//# sourceMappingURL=normalize-numbered-lists.js.map

package/dist/regex/normalize-numbered-lists.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-numbered-lists.js","sourceRoot":"","sources":["../../src/regex/normalize-numbered-lists.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAE5D,MAAM,UAAU,sBAAsB,CAAC,EAAU;IAC/C,OAAO,iBAAiB,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,EAAE,CACjC,CAAC;SACE,OAAO,CAAC,6BAA6B,EAAE,OAAO,CAAC;SAC/C,OAAO,CAAC,+BAA+B,EAAE,OAAO,CAAC;SACjD,OAAO,CAAC,+BAA+B,EAAE,OAAO,CAAC;SACjD,OAAO,CAAC,iCAAiC,EAAE,OAAO,CAAC,CACvD,CAAC;AACJ,CAAC"}

package/dist/regex/normalize-unicode.d.ts

@@ -0,0 +1,2 @@
+export declare function normalizeUnicode(md: string): string;
+//# sourceMappingURL=normalize-unicode.d.ts.map

package/dist/regex/normalize-unicode.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-unicode.d.ts","sourceRoot":"","sources":["../../src/regex/normalize-unicode.ts"],"names":[],"mappings":"AAmCA,wBAAgB,gBAAgB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAkBnD"}

package/dist/regex/normalize-unicode.js

@@ -0,0 +1,49 @@
+/**
+ * normalizeUnicode — fix misplaced PDF accents + NFC + smart-quote/dash mapping.
+ * Idempotent. Run first.
+ *
+ * PDF text extraction (pdfminer/pdfplumber) emits the accent BEFORE the vowel
+ * it belongs to, as either a combining mark ("est"+U+0301+"a") or, more often,
+ * a standalone non-ASCII spacing diacritic ("duraci´on", "electr´onicos").
+ * Neither composes under plain NFC. We move any such accent forward onto the
+ * next letter — but only when that letter has a precomposed accented form, so
+ * an accent is never forced onto a consonant — then NFC composes it.
+ *
+ * Only NON-ASCII spacing diacritics are handled; ASCII look-alikes (^ ~ ` ' ")
+ * are Markdown/code syntax and are deliberately left untouched.
+ */
+import { withProtectedCode } from "../util/protect-code.js";
+// Non-ASCII spacing diacritic → its combining counterpart.
+const SPACING_TO_COMBINING = {
+    "´": "́", // ´ acute            → á é í ó ú
+    "ˊ": "́", // ˊ modifier acute
+    "¨": "̈", // ¨ diaeresis        → ä ë ï ö ü
+    "˜": "̃", // ˜ small tilde      → ã ñ õ
+    "ˆ": "̂", // ˆ modifier circumflex → â ê î ô û
+    "¸": "̧", // ¸ cedilla          → ç
+    "ˋ": "̀", // ˋ modifier grave   → à è ì ò ù
+};
+const SPACING_CHARS = Object.keys(SPACING_TO_COMBINING).join("");
+// an accent = a combining mark OR one of the spacing diacritics above
+const ACCENT_BEFORE_LETTER = new RegExp(`([̀-ͯ${SPACING_CHARS}])[ \\u00a0]?(\\p{L})`, "gu");
+function composesToOne(base, mark) {
+    return (base + mark).normalize("NFC").length === 1;
+}
+export function normalizeUnicode(md) {
+    return withProtectedCode(md, (s) => {
+        let out = s.replace(ACCENT_BEFORE_LETTER, (full, acc, letter) => {
+            const mark = SPACING_TO_COMBINING[acc] ?? acc; // spacing → combining; combining stays
+            return composesToOne(letter, mark) ? letter + mark : full;
+        });
+        out = out.normalize("NFC");
+        out = out
+            .replace(/[‘’‚‛]/g, "'")
+            .replace(/[“”„‟]/g, '"')
+            .replace(/[–—]/g, "—")
+            .replace(/­/g, "") // soft hyphen
+            .replace(/ /g, " ") // nbsp → space
+            .replace(/​/g, ""); // zero-width space
+        return out;
+    });
+}
+//# sourceMappingURL=normalize-unicode.js.map

package/dist/regex/normalize-unicode.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-unicode.js","sourceRoot":"","sources":["../../src/regex/normalize-unicode.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;GAaG;AACH,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAE5D,2DAA2D;AAC3D,MAAM,oBAAoB,GAAqC;IAC7D,GAAG,EAAE,GAAG,EAAE,iCAAiC;IAC3C,GAAG,EAAE,GAAG,EAAE,mBAAmB;IAC7B,GAAG,EAAE,GAAG,EAAE,iCAAiC;IAC3C,GAAG,EAAE,GAAG,EAAE,6BAA6B;IACvC,GAAG,EAAE,GAAG,EAAE,oCAAoC;IAC9C,GAAG,EAAE,GAAG,EAAE,yBAAyB;IACnC,GAAG,EAAE,GAAG,EAAE,iCAAiC;CAC5C,CAAC;AAEF,MAAM,aAAa,GAAG,MAAM,CAAC,IAAI,CAAC,oBAAoB,CAAC,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;AACjE,sEAAsE;AACtE,MAAM,oBAAoB,GAAG,IAAI,MAAM,CAAC,QAAQ,aAAa,uBAAuB,EAAE,IAAI,CAAC,CAAC;AAE5F,SAAS,aAAa,CAAC,IAAY,EAAE,IAAY;IAC/C,OAAO,CAAC,IAAI,GAAG,IAAI,CAAC,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,MAAM,KAAK,CAAC,CAAC;AACrD,CAAC;AAED,MAAM,UAAU,gBAAgB,CAAC,EAAU;IACzC,OAAO,iBAAiB,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,EAAE;QACjC,IAAI,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC,oBAAoB,EAAE,CAAC,IAAI,EAAE,GAAW,EAAE,MAAc,EAAE,EAAE;YAC9E,MAAM,IAAI,GAAG,oBAAoB,CAAC,GAAG,CAAC,IAAI,GAAG,CAAC,CAAC,uCAAuC;YACtF,OAAO,aAAa,CAAC,MAAM,EAAE,IAAI,CAAC,CAAC,CAAC,CAAC,MAAM,GAAG,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC;QAC5D,CAAC,CAAC,CAAC;QAEH,GAAG,GAAG,GAAG,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC;QAE3B,GAAG,GAAG,GAAG;aACN,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC;aACvB,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC;aACvB,OAAO,CAAC,OAAO,EAAE,GAAG,CAAC;aACrB,OAAO,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,cAAc;aAChC,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC,eAAe;aAClC,OAAO,CAAC,IAAI,EAAE,EAAE,CAAC,CAAC,CAAC,mBAAmB;QACzC,OAAO,GAAG,CAAC;IACb,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/regex/normalize-whitespace-in-lines.d.ts

@@ -0,0 +1,2 @@
+export declare function normalizeWhitespaceInLines(md: string): string;
+//# sourceMappingURL=normalize-whitespace-in-lines.d.ts.map

package/dist/regex/normalize-whitespace-in-lines.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-whitespace-in-lines.d.ts","sourceRoot":"","sources":["../../src/regex/normalize-whitespace-in-lines.ts"],"names":[],"mappings":"AAEA,wBAAgB,0BAA0B,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAmB7D"}

package/dist/regex/normalize-whitespace-in-lines.js

@@ -0,0 +1,24 @@
+import { withProtectedCode } from "../util/protect-code.js";
+export function normalizeWhitespaceInLines(md) {
+    return withProtectedCode(md, (s) => {
+        const lines = s.split("\n");
+        return lines
+            .map((line) => {
+            // Whitespace-only line → empty
+            if (/^\s+$/.test(line))
+                return "";
+            // Preserve double-space hard line break at end (exactly 2 trailing spaces)
+            if (/[^ \t]  $/.test(line))
+                return line;
+            // Strip trailing whitespace
+            line = line.trimEnd();
+            // Collapse multiple internal spaces (skip code-indented lines)
+            if (!/^(?:    |\t)/.test(line)) {
+                line = line.replace(/([^\s])[ \t]{2,}([^\s])/g, "$1 $2");
+            }
+            return line;
+        })
+            .join("\n");
+    });
+}
+//# sourceMappingURL=normalize-whitespace-in-lines.js.map

package/dist/regex/normalize-whitespace-in-lines.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalize-whitespace-in-lines.js","sourceRoot":"","sources":["../../src/regex/normalize-whitespace-in-lines.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAE5D,MAAM,UAAU,0BAA0B,CAAC,EAAU;IACnD,OAAO,iBAAiB,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,EAAE;QACjC,MAAM,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAC5B,OAAO,KAAK;aACT,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE;YACZ,+BAA+B;YAC/B,IAAI,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC;gBAAE,OAAO,EAAE,CAAC;YAClC,2EAA2E;YAC3E,IAAI,WAAW,CAAC,IAAI,CAAC,IAAI,CAAC;gBAAE,OAAO,IAAI,CAAC;YACxC,4BAA4B;YAC5B,IAAI,GAAG,IAAI,CAAC,OAAO,EAAE,CAAC;YACtB,+DAA+D;YAC/D,IAAI,CAAC,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;gBAC/B,IAAI,GAAG,IAAI,CAAC,OAAO,CAAC,0BAA0B,EAAE,OAAO,CAAC,CAAC;YAC3D,CAAC;YACD,OAAO,IAAI,CAAC;QACd,CAAC,CAAC;aACD,IAAI,CAAC,IAAI,CAAC,CAAC;IAChB,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/regex/strip-boilerplate.d.ts

@@ -0,0 +1,3 @@
+import type { CleanOptions } from "../index.js";
+export declare function stripBoilerplate(md: string, opts?: CleanOptions): string;
+//# sourceMappingURL=strip-boilerplate.d.ts.map

package/dist/regex/strip-boilerplate.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-boilerplate.d.ts","sourceRoot":"","sources":["../../src/regex/strip-boilerplate.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,aAAa,CAAC;AAQhD,wBAAgB,gBAAgB,CAAC,EAAE,EAAE,MAAM,EAAE,IAAI,CAAC,EAAE,YAAY,GAAG,MAAM,CAQxE"}

package/dist/regex/strip-boilerplate.js

@@ -0,0 +1,16 @@
+const BOILERPLATE_PATTERNS = [
+    /^[^\n]*(?:©|©|\bCopyright\b)[^\n]*\d{4}[^\n]*$/gim,
+    /^\s*All rights reserved\.?\s*$/gim,
+    /^\s*(?:CONFIDENTIAL|PROPRIETARY|FOR INTERNAL USE ONLY|DRAFT)\s*\.?\s*$/gim,
+];
+export function stripBoilerplate(md, opts) {
+    if (opts?.keepBoilerplate)
+        return md;
+    let out = md;
+    for (const re of BOILERPLATE_PATTERNS) {
+        re.lastIndex = 0;
+        out = out.replace(re, "");
+    }
+    return out;
+}
+//# sourceMappingURL=strip-boilerplate.js.map

package/dist/regex/strip-boilerplate.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-boilerplate.js","sourceRoot":"","sources":["../../src/regex/strip-boilerplate.ts"],"names":[],"mappings":"AAEA,MAAM,oBAAoB,GAAa;IACrC,wDAAwD;IACxD,mCAAmC;IACnC,2EAA2E;CAC5E,CAAC;AAEF,MAAM,UAAU,gBAAgB,CAAC,EAAU,EAAE,IAAmB;IAC9D,IAAI,IAAI,EAAE,eAAe;QAAE,OAAO,EAAE,CAAC;IACrC,IAAI,GAAG,GAAG,EAAE,CAAC;IACb,KAAK,MAAM,EAAE,IAAI,oBAAoB,EAAE,CAAC;QACtC,EAAE,CAAC,SAAS,GAAG,CAAC,CAAC;QACjB,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,EAAE,EAAE,EAAE,CAAC,CAAC;IAC5B,CAAC;IACD,OAAO,GAAG,CAAC;AACb,CAAC"}

package/dist/regex/strip-docx-artifacts.d.ts

@@ -0,0 +1,19 @@
+/**
+ * stripDocxArtifacts — remove pandoc DOCX-conversion leftovers.
+ *
+ * When pandoc converts .docx → markdown it emits span attributes for
+ * formatting that has no GFM equivalent: `[text]{.underline}`,
+ * `[text]{.smallcaps}`, `[text]{.mark}`, `[text]{.highlight}`.
+ * These pass through remark unchanged and end up in final output as
+ * literal punctuation noise.
+ *
+ * Also handles:
+ * - `[text]{.strikethrough}` → `~~text~~` (GFM has this)
+ * - ` `  → regular space
+ * - `\\ ` (pandoc hard line-break) → single space at end-of-line
+ * - `<!-- {.XXX} -->` comment-style span leftovers
+ *
+ * Safe on non-docx input — zero matches on plain markdown.
+ */
+export declare function stripDocxArtifacts(md: string): string;
+//# sourceMappingURL=strip-docx-artifacts.d.ts.map

package/dist/regex/strip-docx-artifacts.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-docx-artifacts.d.ts","sourceRoot":"","sources":["../../src/regex/strip-docx-artifacts.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AACH,wBAAgB,kBAAkB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAqBrD"}

package/dist/regex/strip-docx-artifacts.js

@@ -0,0 +1,34 @@
+/**
+ * stripDocxArtifacts — remove pandoc DOCX-conversion leftovers.
+ *
+ * When pandoc converts .docx → markdown it emits span attributes for
+ * formatting that has no GFM equivalent: `[text]{.underline}`,
+ * `[text]{.smallcaps}`, `[text]{.mark}`, `[text]{.highlight}`.
+ * These pass through remark unchanged and end up in final output as
+ * literal punctuation noise.
+ *
+ * Also handles:
+ * - `[text]{.strikethrough}` → `~~text~~` (GFM has this)
+ * - ` `  → regular space
+ * - `\\ ` (pandoc hard line-break) → single space at end-of-line
+ * - `<!-- {.XXX} -->` comment-style span leftovers
+ *
+ * Safe on non-docx input — zero matches on plain markdown.
+ */
+export function stripDocxArtifacts(md) {
+    let out = md;
+    // [text]{.strikethrough} → ~~text~~ (semantic preservation)
+    out = out.replace(/\[([^\]]+)\]\{\.strikethrough\}/g, "~~$1~~");
+    // [text]{.class} spans with no GFM equivalent → bare text
+    out = out.replace(/\[([^\]]+)\]\{\.(?:underline|smallcaps|mark|highlight|subscript|superscript)\}/g, "$1");
+    // Generic span with only ignored attribute(s): [text]{.anything}
+    // More conservative: only strip if the entire attribute block has no
+    // semantic meaning we want to keep (no # id, no key=val other than class).
+    out = out.replace(/\[([^\]]+)\]\{(?:\.[a-z][a-z0-9-]*\s*)+\}/g, "$1");
+    // Pandoc hard line-break: trailing "\ " before newline
+    out = out.replace(/\\ $/gm, "");
+    // <!-- {.someclass} --> inline comments from div-fence conversion
+    out = out.replace(/<!--\s*\{[^}]+\}\s*-->/g, "");
+    return out;
+}
+//# sourceMappingURL=strip-docx-artifacts.js.map

package/dist/regex/strip-docx-artifacts.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-docx-artifacts.js","sourceRoot":"","sources":["../../src/regex/strip-docx-artifacts.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AACH,MAAM,UAAU,kBAAkB,CAAC,EAAU;IAC3C,IAAI,GAAG,GAAG,EAAE,CAAC;IAEb,4DAA4D;IAC5D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,kCAAkC,EAAE,QAAQ,CAAC,CAAC;IAEhE,0DAA0D;IAC1D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,iFAAiF,EAAE,IAAI,CAAC,CAAC;IAE3G,iEAAiE;IACjE,qEAAqE;IACrE,2EAA2E;IAC3E,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,4CAA4C,EAAE,IAAI,CAAC,CAAC;IAEtE,uDAAuD;IACvD,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC;IAEhC,kEAAkE;IAClE,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,yBAAyB,EAAE,EAAE,CAAC,CAAC;IAEjD,OAAO,GAAG,CAAC;AACb,CAAC"}

package/dist/regex/strip-empty-headings.d.ts

@@ -0,0 +1,2 @@
+export declare function stripEmptyHeadings(md: string): string;
+//# sourceMappingURL=strip-empty-headings.d.ts.map

package/dist/regex/strip-empty-headings.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-empty-headings.d.ts","sourceRoot":"","sources":["../../src/regex/strip-empty-headings.ts"],"names":[],"mappings":"AAIA,wBAAgB,kBAAkB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAErD"}

package/dist/regex/strip-empty-headings.js

@@ -0,0 +1,6 @@
+const EMPTY_RE = /^#{1,6}[ \t]*$/gm;
+const PUNCT_ONLY_RE = /^(#{1,6})[ \t]+(\d+\.[\d.]*|[ivxlcdmIVXLCDM]+\.|[-–—.:;])[ \t]*$/gm;
+export function stripEmptyHeadings(md) {
+    return md.replace(EMPTY_RE, "").replace(PUNCT_ONLY_RE, "");
+}
+//# sourceMappingURL=strip-empty-headings.js.map

package/dist/regex/strip-empty-headings.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-empty-headings.js","sourceRoot":"","sources":["../../src/regex/strip-empty-headings.ts"],"names":[],"mappings":"AAAA,MAAM,QAAQ,GAAG,kBAAkB,CAAC;AACpC,MAAM,aAAa,GACjB,oEAAoE,CAAC;AAEvE,MAAM,UAAU,kBAAkB,CAAC,EAAU;IAC3C,OAAO,EAAE,CAAC,OAAO,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC,CAAC;AAC7D,CAAC"}

package/dist/regex/strip-html-artifacts.d.ts

@@ -0,0 +1,2 @@
+export declare function stripHtmlArtifacts(md: string): string;
+//# sourceMappingURL=strip-html-artifacts.d.ts.map

package/dist/regex/strip-html-artifacts.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-html-artifacts.d.ts","sourceRoot":"","sources":["../../src/regex/strip-html-artifacts.ts"],"names":[],"mappings":"AAKA,wBAAgB,kBAAkB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAuBrD"}

package/dist/regex/strip-html-artifacts.js

@@ -0,0 +1,24 @@
+import { withProtectedCode } from "../util/protect-code.js";
+const SAFE_TAGS = /^\/?(p|section|article|header|footer|main|figure|figcaption|caption|mark|abbr|cite|q|ins|u)$/i;
+export function stripHtmlArtifacts(md) {
+    return withProtectedCode(md, (s) => {
+        let out = s;
+        out = out.replace(/^[ \t]*<br\s*\/?>[ \t]*$/gim, "");
+        out = out.replace(/<br\s*\/?>/gi, " ");
+        out = out.replace(/^[ \t]*<hr\s*\/?>[ \t]*$/gim, "\n---\n");
+        out = out.replace(/<strong\b[^>]*>([\s\S]*?)<\/strong>/gi, "**$1**");
+        out = out.replace(/<b\b[^>]*>([\s\S]*?)<\/b>/gi, "**$1**");
+        out = out.replace(/<em\b[^>]*>([\s\S]*?)<\/em>/gi, "_$1_");
+        out = out.replace(/<i\b[^>]*>([\s\S]*?)<\/i>/gi, "_$1_");
+        out = out.replace(/<s\b[^>]*>([\s\S]*?)<\/s>/gi, "~~$1~~");
+        out = out.replace(/<del\b[^>]*>([\s\S]*?)<\/del>/gi, "~~$1~~");
+        out = out.replace(/<code\b[^>]*>([\s\S]*?)<\/code>/gi, "`$1`");
+        out = out.replace(/<sup\b[^>]*>([\s\S]*?)<\/sup>/gi, "$1");
+        out = out.replace(/<sub\b[^>]*>([\s\S]*?)<\/sub>/gi, "$1");
+        out = out.replace(/<div\b[^>]*>([\s\S]*?)<\/div>/gi, "$1");
+        out = out.replace(/<span\b[^>]*>([\s\S]*?)<\/span>/gi, "$1");
+        out = out.replace(/<(\/?)([a-zA-Z][a-zA-Z0-9]*)\b[^>]*>/g, (full, slash, tag) => (SAFE_TAGS.test(`${slash}${tag}`) ? "" : full));
+        return out;
+    });
+}
+//# sourceMappingURL=strip-html-artifacts.js.map

package/dist/regex/strip-html-artifacts.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-html-artifacts.js","sourceRoot":"","sources":["../../src/regex/strip-html-artifacts.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,iBAAiB,EAAE,MAAM,yBAAyB,CAAC;AAE5D,MAAM,SAAS,GACb,+FAA+F,CAAC;AAElG,MAAM,UAAU,kBAAkB,CAAC,EAAU;IAC3C,OAAO,iBAAiB,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,EAAE;QACjC,IAAI,GAAG,GAAG,CAAC,CAAC;QACZ,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,6BAA6B,EAAE,EAAE,CAAC,CAAC;QACrD,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,cAAc,EAAE,GAAG,CAAC,CAAC;QACvC,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,6BAA6B,EAAE,SAAS,CAAC,CAAC;QAC5D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,uCAAuC,EAAE,QAAQ,CAAC,CAAC;QACrE,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,6BAA6B,EAAE,QAAQ,CAAC,CAAC;QAC3D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,+BAA+B,EAAE,MAAM,CAAC,CAAC;QAC3D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,6BAA6B,EAAE,MAAM,CAAC,CAAC;QACzD,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,6BAA6B,EAAE,QAAQ,CAAC,CAAC;QAC3D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,iCAAiC,EAAE,QAAQ,CAAC,CAAC;QAC/D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,mCAAmC,EAAE,MAAM,CAAC,CAAC;QAC/D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,iCAAiC,EAAE,IAAI,CAAC,CAAC;QAC3D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,iCAAiC,EAAE,IAAI,CAAC,CAAC;QAC3D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,iCAAiC,EAAE,IAAI,CAAC,CAAC;QAC3D,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,mCAAmC,EAAE,IAAI,CAAC,CAAC;QAC7D,GAAG,GAAG,GAAG,CAAC,OAAO,CACf,uCAAuC,EACvC,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,EAAE,EAAE,CAAC,CAAC,SAAS,CAAC,IAAI,CAAC,GAAG,KAAK,GAAG,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CACrE,CAAC;QACF,OAAO,GAAG,CAAC;IACb,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/regex/strip-page-numbers.d.ts

@@ -0,0 +1,2 @@
+export declare function stripPageNumbers(md: string): string;
+//# sourceMappingURL=strip-page-numbers.d.ts.map

package/dist/regex/strip-page-numbers.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-page-numbers.d.ts","sourceRoot":"","sources":["../../src/regex/strip-page-numbers.ts"],"names":[],"mappings":"AAiBA,wBAAgB,gBAAgB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAInD"}

package/dist/regex/strip-page-numbers.js

@@ -0,0 +1,23 @@
+/**
+ * stripPageNumbers — remove lines that are *only* a page marker.
+ *
+ * Patterns covered:
+ *   "12"
+ *   "Page 12"
+ *   "Page 12 of 340"
+ *   "- 12 -"
+ *   "[12]"
+ */
+const PATTERNS = [
+    /^\s*\d{1,4}\s*$/gm,
+    /^\s*Page\s+\d+(\s+of\s+\d+)?\s*$/gim,
+    /^\s*-\s*\d{1,4}\s*-\s*$/gm,
+    /^\s*\[\s*\d{1,4}\s*\]\s*$/gm,
+];
+export function stripPageNumbers(md) {
+    let out = md;
+    for (const p of PATTERNS)
+        out = out.replace(p, "");
+    return out;
+}
+//# sourceMappingURL=strip-page-numbers.js.map

package/dist/regex/strip-page-numbers.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-page-numbers.js","sourceRoot":"","sources":["../../src/regex/strip-page-numbers.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AACH,MAAM,QAAQ,GAAG;IACf,mBAAmB;IACnB,qCAAqC;IACrC,2BAA2B;IAC3B,6BAA6B;CAC9B,CAAC;AAEF,MAAM,UAAU,gBAAgB,CAAC,EAAU;IACzC,IAAI,GAAG,GAAG,EAAE,CAAC;IACb,KAAK,MAAM,CAAC,IAAI,QAAQ;QAAE,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;IACnD,OAAO,GAAG,CAAC;AACb,CAAC"}