@cesarandreslopez/occ 0.1.1 → 0.2.0

package/README.md

@@ -19,6 +19,7 @@ OCC scans directories for office documents (DOCX, XLSX, PPTX, PDF, ODT, ODS, ODP
 
 - **Office document metrics** — words, pages, paragraphs, slides, sheets, rows, cells
 - **Seven formats supported** — DOCX, XLSX, PPTX, PDF, ODT, ODS, ODP
+- **Document structure extraction** — `--structure` parses heading hierarchy into a navigable tree with dotted section codes (1, 1.1, 1.2, ...)
 - **Code metrics via scc** — auto-detects code files and integrates scc output
 - **Multiple output modes** — grouped by type, per-file breakdown, or JSON
 - **CI-friendly** — ASCII-only, no-color mode for pipelines
@@ -46,6 +47,7 @@ npx @cesarandreslopez/occ docs/ reports/
 ```bash
 git clone https://github.com/cesarandreslopez/occ.git && cd occ
 npm install
+npm run build
 npm start
 ```
 
@@ -64,6 +66,12 @@ occ --by-file docs/
 # JSON output
 occ --format json docs/
 
+# Extract document structure (heading hierarchy)
+occ --structure docs/
+
+# Structure as JSON
+occ --structure --format json docs/
+
 # Only specific formats
 occ --include-ext pdf,docx docs/
 
@@ -77,37 +85,55 @@ occ --ci docs/
 ## Example Output
 
 ```
--- Documents -----------------------------------------------------------
-Format        Files    Words    Pages    Extra          Size
---------------------------------------------------------------------
-Word             12   34,210      137    1,203 paras    1.2 MB
-PDF               8   22,540       64                   4.5 MB
-Excel             3                      12 sheets      890 KB
---------------------------------------------------------------------
-Total            23   56,750      201    1,203 paras    6.5 MB
---------------------------------------------------------------------
-
--- Code (via scc) ------------------------------------------------------
-Language         Files    Lines   Blanks  Comments     Code
---------------------------------------------------------------------
-JavaScript          15     2340      180       320     1840
-Python               8     1200       90       150      960
---------------------------------------------------------------------
-Total               23     3540      270       470     2800
---------------------------------------------------------------------
+-- Documents ---------------------------------------------------------------
+  Format    Files    Words    Pages                  Details      Size
+----------------------------------------------------------------------------
+  Word         12   34,210      137              1,203 paras    1.2 MB
+  PDF           8   22,540       64                             4.5 MB
+  Excel         3                                12 sheets      890 KB
+----------------------------------------------------------------------------
+  Total        23   56,750      201              1,203 paras    6.5 MB
+
+-- Code (via scc) ----------------------------------------------------------
+  Language    Files    Lines   Blanks  Comments     Code
+----------------------------------------------------------------------------
+  JavaScript     15     2340      180       320     1840
+  Python          8     1200       90       150      960
+----------------------------------------------------------------------------
+  Total          23     3540      270       470     2800
+
+Scanned 23 documents (56,750 words, 201 pages) in 120ms
+```
+
+### Structure Output (`--structure`)
+
+```
+-- Structure: report.docx --------------------------------------------------
+1   Executive Summary
+  1.1   Background ......................................... p.1
+  1.2   Key Findings ....................................... p.1-2
+2   Methodology
+  2.1   Data Collection .................................... p.3
+  2.2   Analysis Framework ................................. p.4
+    2.2.1   Quantitative Methods ........................... p.4
+    2.2.2   Qualitative Methods ............................ p.5
+3   Results ................................................ p.6-8
+4   Conclusions ............................................ p.9
+
+4 sections, 10 nodes, max depth 3
 ```
 
 ## Supported Formats
 
-| Format | Extension | Metrics |
-|--------|-----------|---------|
-| Word | `.docx` | words, pages*, paragraphs |
-| PDF | `.pdf` | words, pages |
-| Excel | `.xlsx` | sheets, rows, cells |
-| PowerPoint | `.pptx` | words, slides |
-| ODT | `.odt` | words, pages*, paragraphs |
-| ODS | `.ods` | sheets, rows, cells |
-| ODP | `.odp` | words, slides |
+| Format | Extension | Metrics | Structure |
+|--------|-----------|---------|-----------|
+| Word | `.docx` | words, pages*, paragraphs | Yes |
+| PDF | `.pdf` | words, pages | Yes (with page mapping) |
+| Excel | `.xlsx` | sheets, rows, cells | — |
+| PowerPoint | `.pptx` | words, slides | Yes (slide headers) |
+| ODT | `.odt` | words, pages*, paragraphs | Yes (best-effort) |
+| ODS | `.ods` | sheets, rows, cells | — |
+| ODP | `.odp` | words, slides | Yes (slide headers) |
 
 \* Pages for Word/ODT are estimated at 250 words/page.
 
@@ -117,6 +143,7 @@ Total               23     3540      270       470     2800
 |------|-------------|---------|
 | `--by-file` / `-f` | Row per file | grouped by type |
 | `--format <type>` | `tabular` or `json` | `tabular` |
+| `--structure` | Extract and display document heading hierarchy | off |
 | `--include-ext <exts>` | Comma-separated extensions | all supported |
 | `--exclude-ext <exts>` | Comma-separated to skip | none |
 | `--exclude-dir <dirs>` | Directories to skip | `node_modules,.git` |
@@ -151,12 +178,15 @@ Tools like `scc`, `cloc`, and `tokei` give you instant visibility into codebases
 
 - **Context budgeting** — LLMs have finite context windows. OCC's word and page counts let agents estimate how much of a document set they can ingest before hitting token limits
 - **Prioritization** — an agent deciding which documents to read can use OCC's JSON output to rank files by size, word count, or type, focusing on the most relevant content first
+- **RAG chunk mapping** — `--structure --format json` outputs heading trees with character offsets, enabling chunk-to-section mapping, scoped retrieval, and citation paths in RAG pipelines
 - **Repository mapping** — agents exploring an unfamiliar codebase can run `occ --format json` to build a structured inventory of all non-code content alongside `scc` code metrics
 - **Pipeline integration** — JSON output pipes directly into agent toolchains for automated document analysis, summarization, or compliance checking
 
 ## How It Works
 
-OCC uses [fast-glob](https://github.com/mrmlnc/fast-glob) for file discovery, dispatches to format-specific parsers (mammoth for DOCX, pdf-parse for PDF, SheetJS for XLSX, JSZip + officeparser for PPTX/ODF), aggregates metrics, and renders output via cli-table3. For code metrics, it shells out to a vendored [scc](https://github.com/boyter/scc) binary (auto-downloaded during `npm install`, with PATH fallback).
+OCC is written in TypeScript and uses [fast-glob](https://github.com/mrmlnc/fast-glob) for file discovery, dispatches to format-specific parsers (mammoth for DOCX, pdf-parse for PDF, SheetJS for XLSX, JSZip + officeparser for PPTX/ODF), aggregates metrics, and renders output via cli-table3. For code metrics, it shells out to a vendored [scc](https://github.com/boyter/scc) binary (auto-downloaded during `npm install`, with PATH fallback).
+
+For structure extraction (`--structure`), documents are first converted to markdown (mammoth + [turndown](https://github.com/mixmark-io/turndown) for DOCX, pdf-parse with page markers for PDF), then headers are extracted and assembled into a tree with dotted section codes.
 
 ## Contributing
 

package/dist/bin/occ.d.ts

@@ -0,0 +1,2 @@
+#!/usr/bin/env node
+export {};

package/{bin → dist/bin}/occ.js

@@ -1,3 +1,4 @@
 #!/usr/bin/env node
 import { run } from '../src/cli.js';
 run(process.argv);
+//# sourceMappingURL=occ.js.map

package/dist/bin/occ.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"occ.js","sourceRoot":"","sources":["../../bin/occ.ts"],"names":[],"mappings":";AACA,OAAO,EAAE,GAAG,EAAE,MAAM,eAAe,CAAC;AACpC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC"}

package/dist/src/cli.d.ts

@@ -0,0 +1 @@
+export declare function run(argv: string[]): Promise<void>;

package/dist/src/cli.js

@@ -0,0 +1,184 @@
+import { Command, Option } from 'commander';
+import { readFile, writeFile } from 'node:fs/promises';
+import { fileURLToPath } from 'node:url';
+import path from 'node:path';
+import { findFiles } from './walker.js';
+import { parseFiles } from './parsers/index.js';
+import { aggregate } from './stats.js';
+import { formatDocumentTable, formatSccTable, formatSummaryLine } from './output/tabular.js';
+import { formatJson } from './output/json.js';
+import { checkScc, runScc } from './scc.js';
+import { createProgress } from './progress.js';
+import { documentToMarkdown } from './markdown/convert.js';
+import { extractFromMarkdown } from './structure/index.js';
+import { formatStructureTree } from './output/tree.js';
+import { getExtension } from './utils.js';
+// Find package.json — works from both src/ (dev) and dist/src/ (built)
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+async function loadPkg() {
+    for (const rel of ['..', '../..']) {
+        try {
+            return JSON.parse(await readFile(path.resolve(__dirname, rel, 'package.json'), 'utf8'));
+        }
+        catch { /* try next */ }
+    }
+    return { version: '0.0.0' };
+}
+const pkg = await loadPkg();
+export async function run(argv) {
+    const program = new Command();
+    program
+        .name('occ')
+        .description('Office Cloc and Count — scc-style summary tables for office documents')
+        .version(pkg.version)
+        .argument('[directories...]', 'directories to scan', [])
+        .option('-f, --by-file', 'show a row per file instead of grouped by type')
+        .option('--format <type>', 'output format: tabular or json', 'tabular')
+        .option('--include-ext <exts>', 'comma-separated extensions to include')
+        .option('--exclude-ext <exts>', 'comma-separated extensions to exclude')
+        .option('--exclude-dir <dirs>', 'directories to skip (comma-separated)', 'node_modules,.git')
+        .option('--no-gitignore', 'disable .gitignore respect')
+        .addOption(new Option('--sort <col>', 'sort by: files, name, words, size').choices(['files', 'name', 'words', 'size']).default('files'))
+        .option('-o, --output <file>', 'write output to file')
+        .option('--ci', 'ASCII-only output, no colors')
+        .option('--large-file-limit <mb>', 'skip files over this size in MB', '50')
+        .option('--no-code', 'skip scc code analysis')
+        .option('--structure', 'extract and display document structure')
+        .action(async (directories, opts) => {
+        try {
+            await execute(directories, opts);
+        }
+        catch (err) {
+            const error = err;
+            process.stderr.write(`Error: ${error.message}\n`);
+            process.exit(1);
+        }
+    });
+    await program.parseAsync(argv);
+}
+function validateLargeFileLimit(value) {
+    const n = parseFloat(value);
+    if (Number.isNaN(n) || n <= 0) {
+        throw new Error(`Invalid --large-file-limit value: "${value}" (must be a positive number)`);
+    }
+    return n;
+}
+const STRUCTURABLE_EXTS = new Set(['docx', 'pdf', 'pptx', 'odt', 'odp']);
+async function extractStructures(files, concurrency, onProgress) {
+    const results = [];
+    for (let i = 0; i < files.length; i += concurrency) {
+        const batch = files.slice(i, i + concurrency);
+        const batchResults = await Promise.allSettled(batch.map(async (f) => {
+            const markdown = await documentToMarkdown(f.path);
+            if (markdown == null)
+                return null;
+            const structure = extractFromMarkdown(markdown);
+            return { file: f.path, structure, markdown };
+        }));
+        for (let j = 0; j < batchResults.length; j++) {
+            const r = batchResults[j];
+            if (r.status === 'fulfilled' && r.value) {
+                results.push(r.value);
+            }
+            if (onProgress)
+                onProgress(1, batch[j]?.path);
+        }
+    }
+    return results;
+}
+async function execute(directories, opts) {
+    const startTime = Date.now();
+    const excludeDirs = opts.excludeDir
+        ? opts.excludeDir.split(',').map(d => d.trim())
+        : ['node_modules', '.git'];
+    const includeCode = opts.code !== false;
+    let sccBinary = null;
+    if (includeCode) {
+        sccBinary = await checkScc();
+    }
+    // Find and parse office documents
+    const { files, skipped } = await findFiles(directories, {
+        includeExt: opts.includeExt,
+        excludeExt: opts.excludeExt,
+        excludeDir: excludeDirs,
+        noGitignore: !opts.gitignore,
+        largeFileLimit: validateLargeFileLimit(opts.largeFileLimit),
+    });
+    const showProgress = opts.format !== 'json' && process.stderr.isTTY;
+    let results = [];
+    if (files.length > 0) {
+        const progress = createProgress({ total: files.length, label: 'Parsing', enabled: showProgress });
+        results = await parseFiles(files, 10, (inc, detail) => progress.update(inc, detail));
+        progress.done();
+    }
+    const stats = aggregate(results, {
+        byFile: opts.byFile,
+        sort: opts.sort,
+    });
+    let sccData = null;
+    if (includeCode) {
+        if (showProgress)
+            process.stderr.write('\rAnalyzing code with scc...');
+        sccData = await runScc(sccBinary, directories, {
+            byFile: opts.byFile,
+            excludeDir: excludeDirs,
+            sort: opts.sort,
+            ci: opts.ci,
+            noGitignore: !opts.gitignore,
+        });
+        if (showProgress) {
+            const cols = process.stderr.columns || 80;
+            process.stderr.write('\r' + ' '.repeat(cols) + '\r');
+        }
+    }
+    // Structure extraction
+    let structureResults = [];
+    if (opts.structure) {
+        const structurableFiles = files.filter(f => STRUCTURABLE_EXTS.has(getExtension(f.path)));
+        if (structurableFiles.length > 0) {
+            const progress = createProgress({ total: structurableFiles.length, label: 'Extracting structure', enabled: showProgress });
+            structureResults = await extractStructures(structurableFiles, 10, (inc, detail) => progress.update(inc, detail));
+            progress.done();
+        }
+    }
+    // Format output
+    let output;
+    if (opts.format === 'json') {
+        output = formatJson(stats, sccData, opts.structure ? structureResults : undefined);
+    }
+    else {
+        const parts = [];
+        if (files.length === 0 && (!sccData || sccData.length === 0)) {
+            parts.push('No files found.');
+        }
+        else {
+            if (files.length > 0) {
+                parts.push(formatDocumentTable(stats, { ci: opts.ci }));
+            }
+            if (sccData && sccData.length > 0) {
+                parts.push(formatSccTable(sccData, { ci: opts.ci, byFile: opts.byFile }));
+            }
+            // Structure trees
+            if (structureResults.length > 0) {
+                for (const sr of structureResults) {
+                    parts.push(formatStructureTree(sr, { ci: opts.ci }));
+                }
+            }
+            const elapsed = Date.now() - startTime;
+            const summary = formatSummaryLine(stats, sccData, elapsed, { ci: opts.ci });
+            if (summary)
+                parts.push(summary);
+        }
+        if (skipped.length > 0) {
+            parts.push(`\n${skipped.length} file(s) skipped (use --large-file-limit to adjust)`);
+        }
+        output = parts.join('\n') + '\n';
+    }
+    if (opts.output) {
+        await writeFile(opts.output, output);
+    }
+    else {
+        process.stdout.write(output);
+    }
+}
+//# sourceMappingURL=cli.js.map

package/dist/src/cli.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"cli.js","sourceRoot":"","sources":["../../src/cli.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,MAAM,EAAE,MAAM,WAAW,CAAC;AAC5C,OAAO,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAC;AACvD,OAAO,EAAE,aAAa,EAAE,MAAM,UAAU,CAAC;AACzC,OAAO,IAAI,MAAM,WAAW,CAAC;AAC7B,OAAO,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AACxC,OAAO,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AAChD,OAAO,EAAE,SAAS,EAAE,MAAM,YAAY,CAAC;AACvC,OAAO,EAAE,mBAAmB,EAAE,cAAc,EAAE,iBAAiB,EAAE,MAAM,qBAAqB,CAAC;AAC7F,OAAO,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAC;AAC9C,OAAO,EAAE,QAAQ,EAAE,MAAM,EAAE,MAAM,UAAU,CAAC;AAC5C,OAAO,EAAE,cAAc,EAAE,MAAM,eAAe,CAAC;AAC/C,OAAO,EAAE,kBAAkB,EAAE,MAAM,uBAAuB,CAAC;AAC3D,OAAO,EAAE,mBAAmB,EAAE,MAAM,sBAAsB,CAAC;AAC3D,OAAO,EAAE,mBAAmB,EAAuB,MAAM,kBAAkB,CAAC;AAK5E,OAAO,EAAE,YAAY,EAAE,MAAM,YAAY,CAAC;AAiB1C,uEAAuE;AACvE,MAAM,SAAS,GAAG,IAAI,CAAC,OAAO,CAAC,aAAa,CAAC,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC;AAC/D,KAAK,UAAU,OAAO;IACpB,KAAK,MAAM,GAAG,IAAI,CAAC,IAAI,EAAE,OAAO,CAAC,EAAE,CAAC;QAClC,IAAI,CAAC;YAAC,OAAO,IAAI,CAAC,KAAK,CAAC,MAAM,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,SAAS,EAAE,GAAG,EAAE,cAAc,CAAC,EAAE,MAAM,CAAC,CAAC,CAAC;QAAC,CAAC;QAChG,MAAM,CAAC,CAAC,cAAc,CAAC,CAAC;IAC1B,CAAC;IACD,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,CAAC;AAC9B,CAAC;AACD,MAAM,GAAG,GAAG,MAAM,OAAO,EAAE,CAAC;AAE5B,MAAM,CAAC,KAAK,UAAU,GAAG,CAAC,IAAc;IACtC,MAAM,OAAO,GAAG,IAAI,OAAO,EAAE,CAAC;IAE9B,OAAO;SACJ,IAAI,CAAC,KAAK,CAAC;SACX,WAAW,CAAC,uEAAuE,CAAC;SACpF,OAAO,CAAC,GAAG,CAAC,OAAO,CAAC;SACpB,QAAQ,CAAC,kBAAkB,EAAE,qBAAqB,EAAE,EAAE,CAAC;SACvD,MAAM,CAAC,eAAe,EAAE,gDAAgD,CAAC;SACzE,MAAM,CAAC,iBAAiB,EAAE,gCAAgC,EAAE,SAAS,CAAC;SACtE,MAAM,CAAC,sBAAsB,EAAE,uCAAuC,CAAC;SACvE,MAAM,CAAC,sBAAsB,EAAE,uCAAuC,CAAC;SACvE,MAAM,CAAC,sBAAsB,EAAE,uCAAuC,EAAE,mBAAmB,CAAC;SAC5F,MAAM,CAAC,gBAAgB,EAAE,4BAA4B,CAAC;SACtD,SAAS,CAAC,IAAI,MAAM,CAAC,cAAc,EAAE,mCAAmC,CAAC,CAAC,OAAO,CAAC,CAAC,OAAO,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,CAAC,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC;SACvI,MAAM,CAAC,qBAAqB,EAAE,sBAAsB,CAAC;SACrD,MAAM,CAAC,MAAM,EAAE,8BAA8B,CAAC;SAC9C,MAAM,CAAC,yBAAyB,EAAE,iCAAiC,EAAE,IAAI,CAAC;SAC1E,MAAM,CAAC,WAAW,EAAE,wBAAwB,CAAC;SAC7C,MAAM,CAAC,aAAa,EAAE,wCAAwC,CAAC;SAC/D,MAAM,CAAC,KAAK,EAAE,WAAqB,EAAE,IAAgB,EAAE,EAAE;QACxD,IAAI,CAAC;YACH,MAAM,OAAO,CAAC,WAAW,EAAE,IAAI,CAAC,CAAC;QACnC,CAAC;QAAC,OAAO,GAAY,EAAE,CAAC;YACtB,MAAM,KAAK,GAAG,GAAY,CAAC;YAC3B,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,UAAU,KAAK,CAAC,OAAO,IAAI,CAAC,CAAC;YAClD,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;QAClB,CAAC;IACH,CAAC,CAAC,CAAC;IAEL,MAAM,OAAO,CAAC,UAAU,CAAC,IAAI,CAAC,CAAC;AACjC,CAAC;AAED,SAAS,sBAAsB,CAAC,KAAa;IAC3C,MAAM,CAAC,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC;IAC5B,IAAI,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;QAC9B,MAAM,IAAI,KAAK,CAAC,sCAAsC,KAAK,+BAA+B,CAAC,CAAC;IAC9F,CAAC;IACD,OAAO,CAAC,CAAC;AACX,CAAC;AAED,MAAM,iBAAiB,GAAG,IAAI,GAAG,CAAC,CAAC,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,CAAC,CAAC,CAAC;AAEzE,KAAK,UAAU,iBAAiB,CAC9B,KAAkB,EAClB,WAAmB,EACnB,UAAmD;IAEnD,MAAM,OAAO,GAAsB,EAAE,CAAC;IAEtC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,IAAI,WAAW,EAAE,CAAC;QACnD,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,GAAG,WAAW,CAAC,CAAC;QAC9C,MAAM,YAAY,GAAG,MAAM,OAAO,CAAC,UAAU,CAC3C,KAAK,CAAC,GAAG,CAAC,KAAK,EAAE,CAAC,EAAE,EAAE;YACpB,MAAM,QAAQ,GAAG,MAAM,kBAAkB,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;YAClD,IAAI,QAAQ,IAAI,IAAI;gBAAE,OAAO,IAAI,CAAC;YAClC,MAAM,SAAS,GAAG,mBAAmB,CAAC,QAAQ,CAAC,CAAC;YAChD,OAAO,EAAE,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,SAAS,EAAE,QAAQ,EAAqB,CAAC;QAClE,CAAC,CAAC,CACH,CAAC;QACF,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,YAAY,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;YAC7C,MAAM,CAAC,GAAG,YAAY,CAAC,CAAC,CAAC,CAAC;YAC1B,IAAI,CAAC,CAAC,MAAM,KAAK,WAAW,IAAI,CAAC,CAAC,KAAK,EAAE,CAAC;gBACxC,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC;YACxB,CAAC;YACD,IAAI,UAAU;gBAAE,UAAU,CAAC,CAAC,EAAE,KAAK,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,CAAC;QAChD,CAAC;IACH,CAAC;IAED,OAAO,OAAO,CAAC;AACjB,CAAC;AAED,KAAK,UAAU,OAAO,CAAC,WAAqB,EAAE,IAAgB;IAC5D,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;IAC7B,MAAM,WAAW,GAAG,IAAI,CAAC,UAAU;QACjC,CAAC,CAAC,IAAI,CAAC,UAAU,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAC/C,CAAC,CAAC,CAAC,cAAc,EAAE,MAAM,CAAC,CAAC;IAE7B,MAAM,WAAW,GAAG,IAAI,CAAC,IAAI,KAAK,KAAK,CAAC;IAExC,IAAI,SAAS,GAAkB,IAAI,CAAC;IACpC,IAAI,WAAW,EAAE,CAAC;QAChB,SAAS,GAAG,MAAM,QAAQ,EAAE,CAAC;IAC/B,CAAC;IAED,kCAAkC;IAClC,MAAM,EAAE,KAAK,EAAE,OAAO,EAAE,GAAG,MAAM,SAAS,CAAC,WAAW,EAAE;QACtD,UAAU,EAAE,IAAI,CAAC,UAAU;QAC3B,UAAU,EAAE,IAAI,CAAC,UAAU;QAC3B,UAAU,EAAE,WAAW;QACvB,WAAW,EAAE,CAAC,IAAI,CAAC,SAAS;QAC5B,cAAc,EAAE,sBAAsB,CAAC,IAAI,CAAC,cAAc,CAAC;KAC5D,CAAC,CAAC;IAEH,MAAM,YAAY,GAAG,IAAI,CAAC,MAAM,KAAK,MAAM,IAAI,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC;IACpE,IAAI,OAAO,GAAkB,EAAE,CAAC;IAChC,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACrB,MAAM,QAAQ,GAAG,cAAc,CAAC,EAAE,KAAK,EAAE,KAAK,CAAC,MAAM,EAAE,KAAK,EAAE,SAAS,EAAE,OAAO,EAAE,YAAY,EAAE,CAAC,CAAC;QAClG,OAAO,GAAG,MAAM,UAAU,CAAC,KAAK,EAAE,EAAE,EAAE,CAAC,GAAG,EAAE,MAAM,EAAE,EAAE,CAAC,QAAQ,CAAC,MAAM,CAAC,GAAG,EAAE,MAAM,CAAC,CAAC,CAAC;QACrF,QAAQ,CAAC,IAAI,EAAE,CAAC;IAClB,CAAC;IAED,MAAM,KAAK,GAAG,SAAS,CAAC,OAAO,EAAE;QAC/B,MAAM,EAAE,IAAI,CAAC,MAAM;QACnB,IAAI,EAAE,IAAI,CAAC,IAAI;KAChB,CAAC,CAAC;IAEH,IAAI,OAAO,GAAyB,IAAI,CAAC;IACzC,IAAI,WAAW,EAAE,CAAC;QAChB,IAAI,YAAY;YAAE,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,8BAA8B,CAAC,CAAC;QACvE,OAAO,GAAG,MAAM,MAAM,CAAC,SAAS,EAAE,WAAW,EAAE;YAC7C,MAAM,EAAE,IAAI,CAAC,MAAM;YACnB,UAAU,EAAE,WAAW;YACvB,IAAI,EAAE,IAAI,CAAC,IAAI;YACf,EAAE,EAAE,IAAI,CAAC,EAAE;YACX,WAAW,EAAE,CAAC,IAAI,CAAC,SAAS;SAC7B,CAAC,CAAC;QACH,IAAI,YAAY,EAAE,CAAC;YACjB,MAAM,IAAI,GAAG,OAAO,CAAC,MAAM,CAAC,OAAO,IAAI,EAAE,CAAC;YAC1C,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,IAAI,GAAG,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC,CAAC;QACvD,CAAC;IACH,CAAC;IAED,uBAAuB;IACvB,IAAI,gBAAgB,GAAsB,EAAE,CAAC;IAC7C,IAAI,IAAI,CAAC,SAAS,EAAE,CAAC;QACnB,MAAM,iBAAiB,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,iBAAiB,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;QACzF,IAAI,iBAAiB,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACjC,MAAM,QAAQ,GAAG,cAAc,CAAC,EAAE,KAAK,EAAE,iBAAiB,CAAC,MAAM,EAAE,KAAK,EAAE,sBAAsB,EAAE,OAAO,EAAE,YAAY,EAAE,CAAC,CAAC;YAC3H,gBAAgB,GAAG,MAAM,iBAAiB,CAAC,iBAAiB,EAAE,EAAE,EAAE,CAAC,GAAG,EAAE,MAAM,EAAE,EAAE,CAAC,QAAQ,CAAC,MAAM,CAAC,GAAG,EAAE,MAAM,CAAC,CAAC,CAAC;YACjH,QAAQ,CAAC,IAAI,EAAE,CAAC;QAClB,CAAC;IACH,CAAC;IAED,gBAAgB;IAChB,IAAI,MAAc,CAAC;IACnB,IAAI,IAAI,CAAC,MAAM,KAAK,MAAM,EAAE,CAAC;QAC3B,MAAM,GAAG,UAAU,CAAC,KAAK,EAAE,OAAO,EAAE,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,gBAAgB,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC;IACrF,CAAC;SAAM,CAAC;QACN,MAAM,KAAK,GAAa,EAAE,CAAC;QAE3B,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,IAAI,CAAC,CAAC,OAAO,IAAI,OAAO,CAAC,MAAM,KAAK,CAAC,CAAC,EAAE,CAAC;YAC7D,KAAK,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAC;QAChC,CAAC;aAAM,CAAC;YACN,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACrB,KAAK,CAAC,IAAI,CAAC,mBAAmB,CAAC,KAAK,EAAE,EAAE,EAAE,EAAE,IAAI,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC;YAC1D,CAAC;YAED,IAAI,OAAO,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBAClC,KAAK,CAAC,IAAI,CAAC,cAAc,CAAC,OAAO,EAAE,EAAE,EAAE,EAAE,IAAI,CAAC,EAAE,EAAE,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC;YAC5E,CAAC;YAED,kBAAkB;YAClB,IAAI,gBAAgB,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBAChC,KAAK,MAAM,EAAE,IAAI,gBAAgB,EAAE,CAAC;oBAClC,KAAK,CAAC,IAAI,CAAC,mBAAmB,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE,IAAI,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC;gBACvD,CAAC;YACH,CAAC;YAED,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,EAAE,GAAG,SAAS,CAAC;YACvC,MAAM,OAAO,GAAG,iBAAiB,CAAC,KAAK,EAAE,OAAO,EAAE,OAAO,EAAE,EAAE,EAAE,EAAE,IAAI,CAAC,EAAE,EAAE,CAAC,CAAC;YAC5E,IAAI,OAAO;gBAAE,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;QACnC,CAAC;QAED,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACvB,KAAK,CAAC,IAAI,CAAC,KAAK,OAAO,CAAC,MAAM,qDAAqD,CAAC,CAAC;QACvF,CAAC;QAED,MAAM,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC;IACnC,CAAC;IAED,IAAI,IAAI,CAAC,MAAM,EAAE,CAAC;QAChB,MAAM,SAAS,CAAC,IAAI,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACvC,CAAC;SAAM,CAAC;QACN,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IAC/B,CAAC;AACH,CAAC"}

package/dist/src/markdown/convert.d.ts

@@ -0,0 +1,2 @@
+/** Convert a document to markdown. Returns null for unsupported formats (xlsx, ods). */
+export declare function documentToMarkdown(filePath: string): Promise<string | null>;

package/dist/src/markdown/convert.js

@@ -0,0 +1,117 @@
+import { readFile } from 'node:fs/promises';
+import mammoth from 'mammoth';
+import pdf from 'pdf-parse';
+import JSZip from 'jszip';
+import officeparser from 'officeparser';
+import TurndownService from 'turndown';
+import { getExtension } from '../utils.js';
+const turndown = new TurndownService({ headingStyle: 'atx' });
+/** Convert a DOCX file to markdown via mammoth → HTML → turndown */
+async function docxToMarkdown(filePath) {
+    const result = await mammoth.convertToHtml({ path: filePath });
+    const html = result.value || '';
+    if (!html.trim())
+        return '';
+    return turndown.turndown(html);
+}
+/** Convert a PDF to markdown with [Page N] markers */
+async function pdfToMarkdown(filePath) {
+    const buffer = await readFile(filePath);
+    // Suppress pdf.js warnings
+    const originalLog = console.log;
+    console.log = (...args) => {
+        if (typeof args[0] === 'string' && (args[0].startsWith('Warning: ') || args[0].startsWith('Info: ') || args[0].startsWith('Deprecated API usage: '))) {
+            return;
+        }
+        originalLog.apply(console, args);
+    };
+    let data;
+    try {
+        data = await pdf(buffer, {
+            pagerender: async (pageData) => {
+                const textContent = await pageData.getTextContent();
+                const strings = textContent.items.map(item => item.str);
+                return `[Page ${pageData.pageIndex + 1}]\n${strings.join(' ')}`;
+            },
+        });
+    }
+    finally {
+        console.log = originalLog;
+    }
+    return data.text;
+}
+/** Convert a PPTX to markdown with slide headers */
+async function pptxToMarkdown(filePath) {
+    const buffer = await readFile(filePath);
+    const zip = await JSZip.loadAsync(buffer);
+    // Get slide filenames sorted by number
+    const slideFiles = Object.keys(zip.files)
+        .filter(name => /^ppt\/slides\/slide\d+\.xml$/.test(name))
+        .sort((a, b) => {
+        const numA = parseInt(a.match(/slide(\d+)/)?.[1] || '0', 10);
+        const numB = parseInt(b.match(/slide(\d+)/)?.[1] || '0', 10);
+        return numA - numB;
+    });
+    // Get full text via officeparser
+    const fullText = await officeparser.parseOffice(buffer);
+    const slideCount = slideFiles.length;
+    if (slideCount <= 1) {
+        return `# Slide 1\n\n${fullText}`;
+    }
+    // Split text roughly between slides
+    const lines = fullText.split('\n');
+    const linesPerSlide = Math.max(1, Math.ceil(lines.length / slideCount));
+    const parts = [];
+    for (let i = 0; i < slideCount; i++) {
+        const start = i * linesPerSlide;
+        const end = Math.min((i + 1) * linesPerSlide, lines.length);
+        const slideText = lines.slice(start, end).join('\n').trim();
+        parts.push(`# Slide ${i + 1}\n\n${slideText}`);
+    }
+    return parts.join('\n\n');
+}
+/** Convert an ODT file to markdown (best-effort heading detection) */
+async function odtToMarkdown(filePath) {
+    const text = await officeparser.parseOffice(filePath);
+    return text;
+}
+/** Convert an ODP file to markdown with slide headers */
+async function odpToMarkdown(filePath) {
+    const buffer = await readFile(filePath);
+    const zip = await JSZip.loadAsync(buffer);
+    const contentXml = await zip.file('content.xml')?.async('text');
+    if (!contentXml)
+        return '';
+    const slides = (contentXml.match(/<draw:page /g) || []).length;
+    const text = await officeparser.parseOffice(buffer);
+    if (slides <= 1) {
+        return `# Slide 1\n\n${text}`;
+    }
+    const lines = text.split('\n');
+    const linesPerSlide = Math.max(1, Math.ceil(lines.length / slides));
+    const parts = [];
+    for (let i = 0; i < slides; i++) {
+        const start = i * linesPerSlide;
+        const end = Math.min((i + 1) * linesPerSlide, lines.length);
+        const slideText = lines.slice(start, end).join('\n').trim();
+        parts.push(`# Slide ${i + 1}\n\n${slideText}`);
+    }
+    return parts.join('\n\n');
+}
+/** Convert a document to markdown. Returns null for unsupported formats (xlsx, ods). */
+export async function documentToMarkdown(filePath) {
+    const ext = getExtension(filePath);
+    switch (ext) {
+        case 'docx': return docxToMarkdown(filePath);
+        case 'pdf': return pdfToMarkdown(filePath);
+        case 'pptx': return pptxToMarkdown(filePath);
+        case 'odt': return odtToMarkdown(filePath);
+        case 'odp': return odpToMarkdown(filePath);
+        case 'xlsx':
+        case 'ods':
+            return null;
+        default:
+            return null;
+    }
+}
+//# sourceMappingURL=convert.js.map

package/dist/src/markdown/convert.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"convert.js","sourceRoot":"","sources":["../../../src/markdown/convert.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,QAAQ,EAAE,MAAM,kBAAkB,CAAC;AAC5C,OAAO,OAAO,MAAM,SAAS,CAAC;AAC9B,OAAO,GAAG,MAAM,WAAW,CAAC;AAC5B,OAAO,KAAK,MAAM,OAAO,CAAC;AAC1B,OAAO,YAAY,MAAM,cAAc,CAAC;AACxC,OAAO,eAAe,MAAM,UAAU,CAAC;AACvC,OAAO,EAAE,YAAY,EAAE,MAAM,aAAa,CAAC;AAE3C,MAAM,QAAQ,GAAG,IAAI,eAAe,CAAC,EAAE,YAAY,EAAE,KAAK,EAAE,CAAC,CAAC;AAE9D,oEAAoE;AACpE,KAAK,UAAU,cAAc,CAAC,QAAgB;IAC5C,MAAM,MAAM,GAAG,MAAM,OAAO,CAAC,aAAa,CAAC,EAAE,IAAI,EAAE,QAAQ,EAAE,CAAC,CAAC;IAC/D,MAAM,IAAI,GAAG,MAAM,CAAC,KAAK,IAAI,EAAE,CAAC;IAChC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;QAAE,OAAO,EAAE,CAAC;IAC5B,OAAO,QAAQ,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC;AACjC,CAAC;AAED,sDAAsD;AACtD,KAAK,UAAU,aAAa,CAAC,QAAgB;IAC3C,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC;IAExC,2BAA2B;IAC3B,MAAM,WAAW,GAAG,OAAO,CAAC,GAAG,CAAC;IAChC,OAAO,CAAC,GAAG,GAAG,CAAC,GAAG,IAAe,EAAE,EAAE;QACnC,IAAI,OAAO,IAAI,CAAC,CAAC,CAAC,KAAK,QAAQ,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,WAAW,CAAC,IAAI,IAAI,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,QAAQ,CAAC,IAAI,IAAI,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,wBAAwB,CAAC,CAAC,EAAE,CAAC;YACrJ,OAAO;QACT,CAAC;QACD,WAAW,CAAC,KAAK,CAAC,OAAO,EAAE,IAAI,CAAC,CAAC;IACnC,CAAC,CAAC;IAEF,IAAI,IAAwC,CAAC;IAC7C,IAAI,CAAC;QACH,IAAI,GAAG,MAAM,GAAG,CAAC,MAAM,EAAE;YACvB,UAAU,EAAE,KAAK,EAAE,QAAuE,EAAE,EAAE;gBAC5F,MAAM,WAAW,GAAG,MAAM,QAAQ,CAAC,cAAc,EAAuC,CAAC;gBACzF,MAAM,OAAO,GAAG,WAAW,CAAC,KAAK,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;gBACxD,OAAO,SAAS,QAAQ,CAAC,SAAS,GAAG,CAAC,MAAM,OAAO,CAAC,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC;YAClE,CAAC;SACF,CAAC,CAAC;IACL,CAAC;YAAS,CAAC;QACT,OAAO,CAAC,GAAG,GAAG,WAAW,CAAC;IAC5B,CAAC;IAED,OAAO,IAAI,CAAC,IAAI,CAAC;AACnB,CAAC;AAED,oDAAoD;AACpD,KAAK,UAAU,cAAc,CAAC,QAAgB;IAC5C,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC;IACxC,MAAM,GAAG,GAAG,MAAM,KAAK,CAAC,SAAS,CAAC,MAAM,CAAC,CAAC;IAE1C,uCAAuC;IACvC,MAAM,UAAU,GAAG,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC;SACtC,MAAM,CAAC,IAAI,CAAC,EAAE,CAAC,8BAA8B,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;SACzD,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE;QACb,MAAM,IAAI,GAAG,QAAQ,CAAC,CAAC,CAAC,KAAK,CAAC,YAAY,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,GAAG,EAAE,EAAE,CAAC,CAAC;QAC7D,MAAM,IAAI,GAAG,QAAQ,CAAC,CAAC,CAAC,KAAK,CAAC,YAAY,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,GAAG,EAAE,EAAE,CAAC,CAAC;QAC7D,OAAO,IAAI,GAAG,IAAI,CAAC;IACrB,CAAC,CAAC,CAAC;IAEL,iCAAiC;IACjC,MAAM,QAAQ,GAAG,MAAM,YAAY,CAAC,WAAW,CAAC,MAAM,CAAsB,CAAC;IAC7E,MAAM,UAAU,GAAG,UAAU,CAAC,MAAM,CAAC;IAErC,IAAI,UAAU,IAAI,CAAC,EAAE,CAAC;QACpB,OAAO,gBAAgB,QAAQ,EAAE,CAAC;IACpC,CAAC;IAED,oCAAoC;IACpC,MAAM,KAAK,GAAG,QAAQ,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IACnC,MAAM,aAAa,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,GAAG,UAAU,CAAC,CAAC,CAAC;IACxE,MAAM,KAAK,GAAa,EAAE,CAAC;IAE3B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,EAAE,CAAC,EAAE,EAAE,CAAC;QACpC,MAAM,KAAK,GAAG,CAAC,GAAG,aAAa,CAAC;QAChC,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,aAAa,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QAC5D,MAAM,SAAS,GAAG,KAAK,CAAC,KAAK,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,IAAI,EAAE,CAAC;QAC5D,KAAK,CAAC,IAAI,CAAC,WAAW,CAAC,GAAG,CAAC,OAAO,SAAS,EAAE,CAAC,CAAC;IACjD,CAAC;IAED,OAAO,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;AAC5B,CAAC;AAED,sEAAsE;AACtE,KAAK,UAAU,aAAa,CAAC,QAAgB;IAC3C,MAAM,IAAI,GAAG,MAAM,YAAY,CAAC,WAAW,CAAC,QAAQ,CAAsB,CAAC;IAC3E,OAAO,IAAI,CAAC;AACd,CAAC;AAED,yDAAyD;AACzD,KAAK,UAAU,aAAa,CAAC,QAAgB;IAC3C,MAAM,MAAM,GAAG,MAAM,QAAQ,CAAC,QAAQ,CAAC,CAAC;IACxC,MAAM,GAAG,GAAG,MAAM,KAAK,CAAC,SAAS,CAAC,MAAM,CAAC,CAAC;IAC1C,MAAM,UAAU,GAAG,MAAM,GAAG,CAAC,IAAI,CAAC,aAAa,CAAC,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;IAChE,IAAI,CAAC,UAAU;QAAE,OAAO,EAAE,CAAC;IAE3B,MAAM,MAAM,GAAG,CAAC,UAAU,CAAC,KAAK,CAAC,cAAc,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC;IAC/D,MAAM,IAAI,GAAG,MAAM,YAAY,CAAC,WAAW,CAAC,MAAM,CAAsB,CAAC;IAEzE,IAAI,MAAM,IAAI,CAAC,EAAE,CAAC;QAChB,OAAO,gBAAgB,IAAI,EAAE,CAAC;IAChC,CAAC;IAED,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAC/B,MAAM,aAAa,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,GAAG,MAAM,CAAC,CAAC,CAAC;IACpE,MAAM,KAAK,GAAa,EAAE,CAAC;IAE3B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QAChC,MAAM,KAAK,GAAG,CAAC,GAAG,aAAa,CAAC;QAChC,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,aAAa,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QAC5D,MAAM,SAAS,GAAG,KAAK,CAAC,KAAK,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,IAAI,EAAE,CAAC;QAC5D,KAAK,CAAC,IAAI,CAAC,WAAW,CAAC,GAAG,CAAC,OAAO,SAAS,EAAE,CAAC,CAAC;IACjD,CAAC;IAED,OAAO,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;AAC5B,CAAC;AAED,wFAAwF;AACxF,MAAM,CAAC,KAAK,UAAU,kBAAkB,CAAC,QAAgB;IACvD,MAAM,GAAG,GAAG,YAAY,CAAC,QAAQ,CAAC,CAAC;IAEnC,QAAQ,GAAG,EAAE,CAAC;QACZ,KAAK,MAAM,CAAC,CAAC,OAAO,cAAc,CAAC,QAAQ,CAAC,CAAC;QAC7C,KAAK,KAAK,CAAC,CAAC,OAAO,aAAa,CAAC,QAAQ,CAAC,CAAC;QAC3C,KAAK,MAAM,CAAC,CAAC,OAAO,cAAc,CAAC,QAAQ,CAAC,CAAC;QAC7C,KAAK,KAAK,CAAC,CAAC,OAAO,aAAa,CAAC,QAAQ,CAAC,CAAC;QAC3C,KAAK,KAAK,CAAC,CAAC,OAAO,aAAa,CAAC,QAAQ,CAAC,CAAC;QAC3C,KAAK,MAAM,CAAC;QACZ,KAAK,KAAK;YACR,OAAO,IAAI,CAAC;QACd;YACE,OAAO,IAAI,CAAC;IAChB,CAAC;AACH,CAAC"}

package/dist/src/output/json.d.ts

@@ -0,0 +1,4 @@
+import type { AggregateResult } from '../stats.js';
+import type { SccLanguage } from '../scc.js';
+import type { StructureResult } from './tree.js';
+export declare function formatJson(stats: AggregateResult, sccData?: SccLanguage[] | null, structureResults?: StructureResult[]): string;

package/dist/src/output/json.js

@@ -0,0 +1,42 @@
+import { METRIC_FIELDS, hasKey } from '../utils.js';
+import { formatStructureJson } from './tree.js';
+export function formatJson(stats, sccData = null, structureResults) {
+    const { columns } = stats;
+    const mapRow = (r) => {
+        const entry = {
+            type: r.fileType,
+            ...(r.fileName ? { name: r.fileName } : {}),
+            ...(r.filePath ? { path: r.filePath } : {}),
+            count: r.files,
+        };
+        for (const f of METRIC_FIELDS) {
+            if (r[hasKey(f)])
+                entry[f] = r[f] || 0;
+        }
+        entry.size = r.size;
+        return entry;
+    };
+    const mapTotals = (t) => {
+        const entry = { files: t.files };
+        for (const f of METRIC_FIELDS) {
+            if (columns[hasKey(f)])
+                entry[f] = t[f];
+        }
+        entry.size = t.size;
+        return entry;
+    };
+    const output = {
+        documents: {
+            files: stats.rows.map(mapRow),
+            totals: mapTotals(stats.totals),
+        },
+    };
+    if (sccData && sccData.length > 0) {
+        output.code = sccData;
+    }
+    if (structureResults && structureResults.length > 0) {
+        output.structures = formatStructureJson(structureResults);
+    }
+    return JSON.stringify(output, null, 2);
+}
+//# sourceMappingURL=json.js.map

package/dist/src/output/json.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"json.js","sourceRoot":"","sources":["../../../src/output/json.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,aAAa,EAAE,MAAM,EAAE,MAAM,aAAa,CAAC;AACpD,OAAO,EAAE,mBAAmB,EAAE,MAAM,WAAW,CAAC;AAKhD,MAAM,UAAU,UAAU,CACxB,KAAsB,EACtB,UAAgC,IAAI,EACpC,gBAAoC;IAEpC,MAAM,EAAE,OAAO,EAAE,GAAG,KAAK,CAAC;IAE1B,MAAM,MAAM,GAAG,CAAC,CAAW,EAAE,EAAE;QAC7B,MAAM,KAAK,GAA4B;YACrC,IAAI,EAAE,CAAC,CAAC,QAAQ;YAChB,GAAG,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;YAC3C,GAAG,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;YAC3C,KAAK,EAAE,CAAC,CAAC,KAAK;SACf,CAAC;QACF,KAAK,MAAM,CAAC,IAAI,aAAa,EAAE,CAAC;YAC9B,IAAI,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;gBAAE,KAAK,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;QACzC,CAAC;QACD,KAAK,CAAC,IAAI,GAAG,CAAC,CAAC,IAAI,CAAC;QACpB,OAAO,KAAK,CAAC;IACf,CAAC,CAAC;IAEF,MAAM,SAAS,GAAG,CAAC,CAAW,EAAE,EAAE;QAChC,MAAM,KAAK,GAA4B,EAAE,KAAK,EAAE,CAAC,CAAC,KAAK,EAAE,CAAC;QAC1D,KAAK,MAAM,CAAC,IAAI,aAAa,EAAE,CAAC;YAC9B,IAAI,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;gBAAE,KAAK,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QAC1C,CAAC;QACD,KAAK,CAAC,IAAI,GAAG,CAAC,CAAC,IAAI,CAAC;QACpB,OAAO,KAAK,CAAC;IACf,CAAC,CAAC;IAEF,MAAM,MAAM,GAA4B;QACtC,SAAS,EAAE;YACT,KAAK,EAAE,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC;YAC7B,MAAM,EAAE,SAAS,CAAC,KAAK,CAAC,MAAM,CAAC;SAChC;KACF,CAAC;IAEF,IAAI,OAAO,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QAClC,MAAM,CAAC,IAAI,GAAG,OAAO,CAAC;IACxB,CAAC;IAED,IAAI,gBAAgB,IAAI,gBAAgB,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACpD,MAAM,CAAC,UAAU,GAAG,mBAAmB,CAAC,gBAAgB,CAAC,CAAC;IAC5D,CAAC;IAED,OAAO,IAAI,CAAC,SAAS,CAAC,MAAM,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC;AACzC,CAAC"}

package/dist/src/output/tabular.d.ts

@@ -0,0 +1,12 @@
+import type { AggregateResult } from '../stats.js';
+import type { SccLanguage } from '../scc.js';
+export interface TableOptions {
+    ci?: boolean;
+    byFile?: boolean;
+}
+export declare function formatDocumentTable(stats: AggregateResult, options?: TableOptions): string;
+export declare function formatSccTable(sccData: SccLanguage[], options?: TableOptions): string;
+export declare function formatSummaryLine(stats: AggregateResult, sccData: SccLanguage[] | null, elapsed: number, options?: TableOptions): string;
+export declare function stripAnsi(str: string): string;
+export declare function sectionHeader(title: string, width: number, ci?: boolean): string;
+export declare function tableChars(ci: boolean): Record<string, string>;