castdown-cleaners 0.1.0

package/dist/regex/strip-pptx-notes.d.ts

@@ -0,0 +1,22 @@
+/**
+ * stripPptxNotes — remove speaker-notes sections from PPTX → MD output.
+ *
+ * MarkItDown (via python-pptx) appends speaker notes after each slide's
+ * content. The typical patterns:
+ *
+ *   ## Notes:         (heading variant)
+ *   <notes text>
+ *
+ *   **Notes:**        (bold-paragraph variant)
+ *   <notes text>
+ *
+ *   <!-- notes -->    (comment marker MarkItDown occasionally emits)
+ *   <notes text>
+ *
+ * The section is terminated by the next heading or end-of-string.
+ * Stripping prevents notes from appearing in rendered output.
+ *
+ * Idempotent and safe on non-pptx markdown — no match ⇒ no-op.
+ */
+export declare function stripPptxNotes(md: string): string;
+//# sourceMappingURL=strip-pptx-notes.d.ts.map

package/dist/regex/strip-pptx-notes.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-pptx-notes.d.ts","sourceRoot":"","sources":["../../src/regex/strip-pptx-notes.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AACH,wBAAgB,cAAc,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAcjD"}

package/dist/regex/strip-pptx-notes.js

@@ -0,0 +1,32 @@
+/**
+ * stripPptxNotes — remove speaker-notes sections from PPTX → MD output.
+ *
+ * MarkItDown (via python-pptx) appends speaker notes after each slide's
+ * content. The typical patterns:
+ *
+ *   ## Notes:         (heading variant)
+ *   <notes text>
+ *
+ *   **Notes:**        (bold-paragraph variant)
+ *   <notes text>
+ *
+ *   <!-- notes -->    (comment marker MarkItDown occasionally emits)
+ *   <notes text>
+ *
+ * The section is terminated by the next heading or end-of-string.
+ * Stripping prevents notes from appearing in rendered output.
+ *
+ * Idempotent and safe on non-pptx markdown — no match ⇒ no-op.
+ */
+export function stripPptxNotes(md) {
+    let out = md;
+    // "## Notes:" or "### Notes:" heading + all following non-heading lines
+    // Pattern: match the Notes: heading line, then consume lines until next heading or EOF.
+    out = out.replace(/^#{1,4}\s+Notes:?\s*$(?:\n(?!#{1,6}\s)[^\n]*|\n)*\n?/gim, "");
+    // "**Notes:**" or "**Notes**:" bold-paragraph variant + following paragraph
+    out = out.replace(/^\*\*Notes:?\*\*:?\s*$(?:\n(?!\*\*|#{1,6}\s)[^\n]*|\n)*\n?/gim, "");
+    // <!-- notes --> comment marker + following paragraph
+    out = out.replace(/^<!--\s*notes?\s*-->$(?:\n(?!<!--|#{1,6}\s)[^\n]*|\n)*\n?/gim, "");
+    return out;
+}
+//# sourceMappingURL=strip-pptx-notes.js.map

package/dist/regex/strip-pptx-notes.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-pptx-notes.js","sourceRoot":"","sources":["../../src/regex/strip-pptx-notes.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;GAmBG;AACH,MAAM,UAAU,cAAc,CAAC,EAAU;IACvC,IAAI,GAAG,GAAG,EAAE,CAAC;IAEb,wEAAwE;IACxE,wFAAwF;IACxF,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,yDAAyD,EAAE,EAAE,CAAC,CAAC;IAEjF,4EAA4E;IAC5E,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,+DAA+D,EAAE,EAAE,CAAC,CAAC;IAEvF,sDAAsD;IACtD,GAAG,GAAG,GAAG,CAAC,OAAO,CAAC,8DAA8D,EAAE,EAAE,CAAC,CAAC;IAEtF,OAAO,GAAG,CAAC;AACb,CAAC"}

package/dist/regex/strip-repeated-headers.d.ts

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

package/dist/regex/strip-repeated-headers.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-repeated-headers.d.ts","sourceRoot":"","sources":["../../src/regex/strip-repeated-headers.ts"],"names":[],"mappings":"AAaA,wBAAgB,oBAAoB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAmBvD"}

package/dist/regex/strip-repeated-headers.js

@@ -0,0 +1,37 @@
+/**
+ * stripRepeatedHeaders — remove lines repeated across many pages.
+ *
+ * Heuristic: any line ≥ 8 chars that appears ≥ 4 times across the doc
+ * is likely a running header/footer (e.g., "Chapter 3 — Foundations").
+ *
+ * Avoids stripping real content by requiring multiple occurrences AND
+ * that the line is short enough to be a header (≤ 120 chars).
+ */
+const MIN_OCCURRENCES = 4;
+const MIN_LEN = 8;
+const MAX_LEN = 120;
+export function stripRepeatedHeaders(md) {
+    const lines = md.split("\n");
+    const counts = new Map();
+    for (const raw of lines) {
+        const line = raw.trim();
+        if (line.length < MIN_LEN || line.length > MAX_LEN)
+            continue;
+        if (/^#{1,6}\s/.test(line))
+            continue; // real headings
+        if (/^[-*+]\s/.test(line))
+            continue; // lists
+        if (/^>\s/.test(line))
+            continue; // quotes
+        counts.set(line, (counts.get(line) ?? 0) + 1);
+    }
+    const toStrip = new Set();
+    for (const [line, n] of counts) {
+        if (n >= MIN_OCCURRENCES)
+            toStrip.add(line);
+    }
+    if (toStrip.size === 0)
+        return md;
+    return lines.filter((raw) => !toStrip.has(raw.trim())).join("\n");
+}
+//# sourceMappingURL=strip-repeated-headers.js.map

package/dist/regex/strip-repeated-headers.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-repeated-headers.js","sourceRoot":"","sources":["../../src/regex/strip-repeated-headers.ts"],"names":[],"mappings":"AAAA;;;;;;;;GAQG;AACH,MAAM,eAAe,GAAG,CAAC,CAAC;AAC1B,MAAM,OAAO,GAAG,CAAC,CAAC;AAClB,MAAM,OAAO,GAAG,GAAG,CAAC;AAEpB,MAAM,UAAU,oBAAoB,CAAC,EAAU;IAC7C,MAAM,KAAK,GAAG,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAC7B,MAAM,MAAM,GAAG,IAAI,GAAG,EAAkB,CAAC;IACzC,KAAK,MAAM,GAAG,IAAI,KAAK,EAAE,CAAC;QACxB,MAAM,IAAI,GAAG,GAAG,CAAC,IAAI,EAAE,CAAC;QACxB,IAAI,IAAI,CAAC,MAAM,GAAG,OAAO,IAAI,IAAI,CAAC,MAAM,GAAG,OAAO;YAAE,SAAS;QAC7D,IAAI,WAAW,CAAC,IAAI,CAAC,IAAI,CAAC;YAAE,SAAS,CAAC,gBAAgB;QACtD,IAAI,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC;YAAE,SAAS,CAAC,QAAQ;QAC7C,IAAI,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC;YAAE,SAAS,CAAC,SAAS;QAC1C,MAAM,CAAC,GAAG,CAAC,IAAI,EAAE,CAAC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;IAChD,CAAC;IAED,MAAM,OAAO,GAAG,IAAI,GAAG,EAAU,CAAC;IAClC,KAAK,MAAM,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,MAAM,EAAE,CAAC;QAC/B,IAAI,CAAC,IAAI,eAAe;YAAE,OAAO,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;IAC9C,CAAC;IACD,IAAI,OAAO,CAAC,IAAI,KAAK,CAAC;QAAE,OAAO,EAAE,CAAC;IAElC,OAAO,KAAK,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,GAAG,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AACpE,CAAC"}

package/dist/regex/strip-url-tracking-params.d.ts

@@ -0,0 +1,2 @@
+export declare function stripUrlTrackingParams(md: string): string;
+//# sourceMappingURL=strip-url-tracking-params.d.ts.map

package/dist/regex/strip-url-tracking-params.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-url-tracking-params.d.ts","sourceRoot":"","sources":["../../src/regex/strip-url-tracking-params.ts"],"names":[],"mappings":"AAqBA,wBAAgB,sBAAsB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAKzD"}

package/dist/regex/strip-url-tracking-params.js

@@ -0,0 +1,26 @@
+const TRACKING_PARAMS = new Set([
+    "utm_source", "utm_medium", "utm_campaign", "utm_term", "utm_content",
+    "utm_id", "utm_source_platform", "utm_creative_format", "utm_marketing_tactic",
+    "fbclid", "gclid", "msclkid", "twclid", "ttclid", "li_fat_id",
+    "mc_cid", "mc_eid", "_ga", "_gl",
+]);
+function cleanUrl(url) {
+    try {
+        const u = new URL(url);
+        const toDelete = [...u.searchParams.keys()].filter((k) => TRACKING_PARAMS.has(k));
+        if (toDelete.length === 0)
+            return url;
+        toDelete.forEach((k) => u.searchParams.delete(k));
+        return u.toString();
+    }
+    catch {
+        return url;
+    }
+}
+export function stripUrlTrackingParams(md) {
+    return md.replace(/\[([^\]]*)\]\(([^)]+)\)/g, (full, text, url) => {
+        const clean = cleanUrl(url);
+        return clean === url ? full : `[${text}](${clean})`;
+    });
+}
+//# sourceMappingURL=strip-url-tracking-params.js.map

package/dist/regex/strip-url-tracking-params.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"strip-url-tracking-params.js","sourceRoot":"","sources":["../../src/regex/strip-url-tracking-params.ts"],"names":[],"mappings":"AAAA,MAAM,eAAe,GAAG,IAAI,GAAG,CAAC;IAC9B,YAAY,EAAE,YAAY,EAAE,cAAc,EAAE,UAAU,EAAE,aAAa;IACrE,QAAQ,EAAE,qBAAqB,EAAE,qBAAqB,EAAE,sBAAsB;IAC9E,QAAQ,EAAE,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,QAAQ,EAAE,WAAW;IAC7D,QAAQ,EAAE,QAAQ,EAAE,KAAK,EAAE,KAAK;CACjC,CAAC,CAAC;AAEH,SAAS,QAAQ,CAAC,GAAW;IAC3B,IAAI,CAAC;QACH,MAAM,CAAC,GAAG,IAAI,GAAG,CAAC,GAAG,CAAC,CAAC;QACvB,MAAM,QAAQ,GAAG,CAAC,GAAG,CAAC,CAAC,YAAY,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CACvD,eAAe,CAAC,GAAG,CAAC,CAAC,CAAC,CACvB,CAAC;QACF,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC;YAAE,OAAO,GAAG,CAAC;QACtC,QAAQ,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,YAAY,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC;QAClD,OAAO,CAAC,CAAC,QAAQ,EAAE,CAAC;IACtB,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,GAAG,CAAC;IACb,CAAC;AACH,CAAC;AAED,MAAM,UAAU,sBAAsB,CAAC,EAAU;IAC/C,OAAO,EAAE,CAAC,OAAO,CAAC,0BAA0B,EAAE,CAAC,IAAI,EAAE,IAAI,EAAE,GAAG,EAAE,EAAE;QAChE,MAAM,KAAK,GAAG,QAAQ,CAAC,GAAa,CAAC,CAAC;QACtC,OAAO,KAAK,KAAK,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,IAAc,KAAK,KAAK,GAAG,CAAC;IAChE,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/regex/wrap-long-cell-text.d.ts

@@ -0,0 +1,28 @@
+/**
+ * wrapLongCellText — insert zero-width spaces (U+200B) at natural break
+ * points inside pipe-table cells whose content is a single very long token
+ * (no whitespace gaps, ≥ MIN_CHARS characters).
+ *
+ * Problem: URLs, file paths, and long identifiers in table cells cannot wrap
+ * because they contain no space characters. Typst, HTML, and Word renderers
+ * have limited support for `word-break: break-all` equivalents. Inserting
+ * ZWS at common delimiters gives the renderer explicit break opportunities
+ * without changing the visual content.
+ *
+ * Break points injected after: / . - _ = & ? # , : @ +
+ *
+ * Conservative by design:
+ *   - Only triggers for tokens ≥ MIN_CHARS (default 40).
+ *   - Skips cells that are already short or contain natural spaces.
+ *   - Skips code-fenced lines.
+ *   - Skips separator rows (`| --- |`).
+ *   - Idempotent: ZWS are already invisible so re-running is safe.
+ *
+ * Downstream compatibility:
+ *   - Typst: ZWS is a Unicode line-break opportunity (UAX#14 BK rule).
+ *   - HTML/EPUB: browsers honour ZWS as a word-break point.
+ *   - DOCX: Word respects ZWS.
+ *   - LaTeX: pandoc translates ZWS to `\hspace{0pt}` or ignores it — safe.
+ */
+export declare function wrapLongCellText(md: string): string;
+//# sourceMappingURL=wrap-long-cell-text.d.ts.map

package/dist/regex/wrap-long-cell-text.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"wrap-long-cell-text.d.ts","sourceRoot":"","sources":["../../src/regex/wrap-long-cell-text.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;GAyBG;AAWH,wBAAgB,gBAAgB,CAAC,EAAE,EAAE,MAAM,GAAG,MAAM,CAqBnD"}

package/dist/regex/wrap-long-cell-text.js

@@ -0,0 +1,66 @@
+/**
+ * wrapLongCellText — insert zero-width spaces (U+200B) at natural break
+ * points inside pipe-table cells whose content is a single very long token
+ * (no whitespace gaps, ≥ MIN_CHARS characters).
+ *
+ * Problem: URLs, file paths, and long identifiers in table cells cannot wrap
+ * because they contain no space characters. Typst, HTML, and Word renderers
+ * have limited support for `word-break: break-all` equivalents. Inserting
+ * ZWS at common delimiters gives the renderer explicit break opportunities
+ * without changing the visual content.
+ *
+ * Break points injected after: / . - _ = & ? # , : @ +
+ *
+ * Conservative by design:
+ *   - Only triggers for tokens ≥ MIN_CHARS (default 40).
+ *   - Skips cells that are already short or contain natural spaces.
+ *   - Skips code-fenced lines.
+ *   - Skips separator rows (`| --- |`).
+ *   - Idempotent: ZWS are already invisible so re-running is safe.
+ *
+ * Downstream compatibility:
+ *   - Typst: ZWS is a Unicode line-break opportunity (UAX#14 BK rule).
+ *   - HTML/EPUB: browsers honour ZWS as a word-break point.
+ *   - DOCX: Word respects ZWS.
+ *   - LaTeX: pandoc translates ZWS to `\hspace{0pt}` or ignores it — safe.
+ */
+const ZWS = "​";
+const MIN_CHARS = 40;
+// Characters that mark a good break point. We insert ZWS AFTER each.
+const BREAK_RE = /([/._\-=&?#,:@+])/g;
+const isSeparatorRow = (line) => /^\s*\|[\s|:-]+\|\s*$/.test(line);
+const isPipeRow = (line) => /^\s*\|/.test(line) && !isSeparatorRow(line);
+export function wrapLongCellText(md) {
+    const lines = md.split("\n");
+    let inCode = false;
+    return lines.map((line) => {
+        if (/^\s*```/.test(line)) {
+            inCode = !inCode;
+            return line;
+        }
+        if (inCode || !isPipeRow(line))
+            return line;
+        // Split on pipe-delimiters, wrap long tokens, re-join.
+        return line
+            .split(/(?<!\\)\|/)
+            .map((cell, idx) => {
+            // Keep leading / trailing pipes as-is (empty strings at boundaries).
+            if (!cell.trim())
+                return cell;
+            return wrapCellTokens(cell);
+        })
+            .join("|");
+    }).join("\n");
+}
+function wrapCellTokens(cell) {
+    return cell.replace(/\S+/g, (token) => {
+        // ZWS (​) is non-whitespace so \S+ captures already-wrapped tokens.
+        // Skip them to stay idempotent.
+        if (token.includes(ZWS))
+            return token;
+        if (token.length < MIN_CHARS)
+            return token;
+        return token.replace(BREAK_RE, `$1${ZWS}`);
+    });
+}
+//# sourceMappingURL=wrap-long-cell-text.js.map

package/dist/regex/wrap-long-cell-text.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"wrap-long-cell-text.js","sourceRoot":"","sources":["../../src/regex/wrap-long-cell-text.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;GAyBG;AAEH,MAAM,GAAG,GAAG,GAAG,CAAC;AAChB,MAAM,SAAS,GAAG,EAAE,CAAC;AAErB,qEAAqE;AACrE,MAAM,QAAQ,GAAG,oBAAoB,CAAC;AAEtC,MAAM,cAAc,GAAG,CAAC,IAAY,EAAE,EAAE,CAAC,sBAAsB,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AAC3E,MAAM,SAAS,GAAO,CAAC,IAAY,EAAE,EAAE,CAAC,QAAQ,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,cAAc,CAAC,IAAI,CAAC,CAAC;AAErF,MAAM,UAAU,gBAAgB,CAAC,EAAU;IACzC,MAAM,KAAK,GAAG,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAC7B,IAAI,MAAM,GAAG,KAAK,CAAC;IAEnB,OAAO,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE;QACxB,IAAI,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;YACzB,MAAM,GAAG,CAAC,MAAM,CAAC;YACjB,OAAO,IAAI,CAAC;QACd,CAAC;QACD,IAAI,MAAM,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC;YAAE,OAAO,IAAI,CAAC;QAE5C,uDAAuD;QACvD,OAAO,IAAI;aACR,KAAK,CAAC,WAAW,CAAC;aAClB,GAAG,CAAC,CAAC,IAAI,EAAE,GAAG,EAAE,EAAE;YACjB,qEAAqE;YACrE,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;gBAAE,OAAO,IAAI,CAAC;YAC9B,OAAO,cAAc,CAAC,IAAI,CAAC,CAAC;QAC9B,CAAC,CAAC;aACD,IAAI,CAAC,GAAG,CAAC,CAAC;IACf,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AAChB,CAAC;AAED,SAAS,cAAc,CAAC,IAAY;IAClC,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,CAAC,KAAK,EAAE,EAAE;QACpC,oEAAoE;QACpE,gCAAgC;QAChC,IAAI,KAAK,CAAC,QAAQ,CAAC,GAAG,CAAC;YAAE,OAAO,KAAK,CAAC;QACtC,IAAI,KAAK,CAAC,MAAM,GAAG,SAAS;YAAE,OAAO,KAAK,CAAC;QAC3C,OAAO,KAAK,CAAC,OAAO,CAAC,QAAQ,EAAE,KAAK,GAAG,EAAE,CAAC,CAAC;IAC7C,CAAC,CAAC,CAAC;AACL,CAAC"}

package/dist/util/protect-code.d.ts

@@ -0,0 +1,6 @@
+/**
+ * Masks fenced and indented code blocks with placeholders before applying a
+ * transformation, then restores them. Prevents cleaners from corrupting code.
+ */
+export declare function withProtectedCode(md: string, fn: (masked: string) => string): string;
+//# sourceMappingURL=protect-code.d.ts.map

package/dist/util/protect-code.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"protect-code.d.ts","sourceRoot":"","sources":["../../src/util/protect-code.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,wBAAgB,iBAAiB,CAAC,EAAE,EAAE,MAAM,EAAE,EAAE,EAAE,CAAC,MAAM,EAAE,MAAM,KAAK,MAAM,GAAG,MAAM,CAqBpF"}

package/dist/util/protect-code.js

@@ -0,0 +1,20 @@
+/**
+ * Masks fenced and indented code blocks with placeholders before applying a
+ * transformation, then restores them. Prevents cleaners from corrupting code.
+ */
+export function withProtectedCode(md, fn) {
+    const blocks = [];
+    const FENCED_RE = /^(`{3,}|~{3,})[^\n]*\n[\s\S]*?\n\1[ \t]*$/gm;
+    let masked = md.replace(FENCED_RE, (match) => {
+        const i = blocks.push(match) - 1;
+        return `\x00CODE_${i}\x00`;
+    });
+    // Only protect indented blocks that contain at least one non-whitespace char
+    masked = masked.replace(/(^|\n)((?:[ ]{4}|\t)[^\n]*\S[^\n]*(?:\n(?:[ ]{4}|\t)[^\n]*\S[^\n]*)*)/g, (_, sep, block) => {
+        const i = blocks.push(block) - 1;
+        return `${sep}\x00CODE_${i}\x00`;
+    });
+    const result = fn(masked);
+    return result.replace(/\x00CODE_(\d+)\x00/g, (_, i) => blocks[+i] ?? "");
+}
+//# sourceMappingURL=protect-code.js.map

package/dist/util/protect-code.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"protect-code.js","sourceRoot":"","sources":["../../src/util/protect-code.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,MAAM,UAAU,iBAAiB,CAAC,EAAU,EAAE,EAA8B;IAC1E,MAAM,MAAM,GAAa,EAAE,CAAC;IAE5B,MAAM,SAAS,GAAG,6CAA6C,CAAC;IAChE,IAAI,MAAM,GAAG,EAAE,CAAC,OAAO,CAAC,SAAS,EAAE,CAAC,KAAK,EAAE,EAAE;QAC3C,MAAM,CAAC,GAAG,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;QACjC,OAAO,YAAY,CAAC,MAAM,CAAC;IAC7B,CAAC,CAAC,CAAC;IAEH,6EAA6E;IAC7E,MAAM,GAAG,MAAM,CAAC,OAAO,CACrB,wEAAwE,EACxE,CAAC,CAAC,EAAE,GAAW,EAAE,KAAa,EAAE,EAAE;QAChC,MAAM,CAAC,GAAG,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;QACjC,OAAO,GAAG,GAAG,YAAY,CAAC,MAAM,CAAC;IACnC,CAAC,CACF,CAAC;IAEF,MAAM,MAAM,GAAG,EAAE,CAAC,MAAM,CAAC,CAAC;IAE1B,OAAO,MAAM,CAAC,OAAO,CAAC,qBAAqB,EAAE,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC;AAC3E,CAAC"}

package/package.json

@@ -0,0 +1,63 @@
+{
+  "name": "castdown-cleaners",
+  "version": "0.1.0",
+  "description": "Composable Markdown post-processing pipeline for MarkItDown, Docling, Pandoc, and LlamaParse output.",
+  "type": "module",
+  "main": "./src/index.ts",
+  "types": "./src/index.ts",
+  "exports": {
+    ".": "./src/index.ts"
+  },
+  "publishConfig": {
+    "main": "./dist/index.js",
+    "types": "./dist/index.d.ts",
+    "exports": {
+      ".": {
+        "import": "./dist/index.js",
+        "types": "./dist/index.d.ts"
+      }
+    }
+  },
+  "files": [
+    "dist",
+    "README.md",
+    "LICENSE"
+  ],
+  "scripts": {
+    "build": "tsc -p tsconfig.build.json",
+    "test": "vitest",
+    "typecheck": "tsc -p tsconfig.json --noEmit",
+    "prepublishOnly": "pnpm build && pnpm exec vitest run"
+  },
+  "keywords": [
+    "markdown",
+    "cleaner",
+    "pipeline",
+    "pdf",
+    "docx",
+    "rag",
+    "markitdown",
+    "docling",
+    "pandoc"
+  ],
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/castdown/castdown",
+    "directory": "packages/cleaners"
+  },
+  "dependencies": {
+    "remark": "^15.0.1",
+    "remark-gfm": "^4.0.0",
+    "remark-stringify": "^11.0.0",
+    "unified": "^11.0.5"
+  },
+  "devDependencies": {
+    "@types/node": "^22.7.0",
+    "typescript": "^5.6.0",
+    "vitest": "^2.1.0"
+  },
+  "engines": {
+    "node": ">=20"
+  }
+}