feat(ofd): convert fixed-layout documents to Markdown

Cargo.toml

@@ -9,7 +9,7 @@ version = "0.2.4"
 edition = "2024"
 # Edition 2024 needs 1.85; zip raises it to 1.88.
 rust-version = "1.88"
-description = "Convert documents (doc, docx, odt, rtf, epub, pdf, presentations, spreadsheets, csv) to GitHub-Flavored Markdown"
+description = "Convert documents (doc, docx, odt, rtf, epub, pdf, ofd, presentations, spreadsheets, csv) to GitHub-Flavored Markdown"
 license = "MIT"
 repository = "https://github.com/firecrawl/anydoc"
 readme = "README.md"

README.md

@@ -6,7 +6,7 @@
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 [![skills.sh](https://skills.sh/b/firecrawl/anydoc)](https://skills.sh/firecrawl/anydoc)
 
-Fast Rust library that converts documents (Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF) into clean GitHub-Flavored Markdown. Includes bindings for [Node.js](node/README.md), [Python](python/README.md), and the [browser](wasm/README.md) (WebAssembly).
+Fast Rust library that converts documents (Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, PDF, and OFD) into clean GitHub-Flavored Markdown. Includes bindings for [Node.js](node/README.md), [Python](python/README.md), and the [browser](wasm/README.md) (WebAssembly).
 
 Built by [Firecrawl](https://firecrawl.dev) to turn any office document into LLM-ready Markdown in single-digit milliseconds, with one consistent output no matter which format goes in. It powers [Firecrawl Parse](https://firecrawl.dev/parse), so if you'd rather not run it yourself, the hosted API gives you the same conversion plus our OCR models for the scanned pages anydoc can't read on its own.
 
@@ -134,7 +134,7 @@ let document = anydoc::to_document(&bytes, None)?;
 
 ## OCR
 
-anydoc reads text-based PDFs locally but does no OCR, so a PDF with scanned or image-only pages fails with `NeedsOcr`. Opt in and those documents go to [Firecrawl Parse](https://firecrawl.dev/parse), which OCRs them and returns the same Markdown. No signup needed; set `FIRECRAWL_API_KEY` for higher limits.
+anydoc reads text-based PDFs and OFDs locally but does no OCR, so scanned or image-only pages fail with `NeedsOcr`. For PDFs, opt in and those documents go to [Firecrawl Parse](https://firecrawl.dev/parse), which OCRs them and returns the same Markdown. The hosted PDF parser does not accept OFD; scanned OFD pages still return `NeedsOcr`. No signup needed; set `FIRECRAWL_API_KEY` for higher limits.
 
 |        | Opt in                                         | Key, else `FIRECRAWL_API_KEY` |
 | ------ | ---------------------------------------------- | ----------------------------- |
@@ -168,6 +168,9 @@ Only documents that need OCR leave the machine, and the whole document goes, sin
 | EPUB             | `.epub`                                                    |
 | CSV              | `.csv`                                                     |
 | PDF              | `.pdf`                                                     |
+| OFD              | `.ofd` (text objects)                                    |
+
+OFD conversion preserves outline headings, large-font headings, bold and italic text, URI links, printed lists, and complete bordered table grids. Tables and large-font headings are inferred from fixed-page geometry; pages without those cues remain paragraphs. Image-only pages return `NeedsOcr`; images, vector drawings, and non-link annotations are not included in Markdown.
 
 ## Benchmark
 
@@ -242,7 +245,7 @@ match anydoc::to_markdown(path) {
 | Variant         | Meaning                                                             |
 | --------------- | ------------------------------------------------------------------- |
 | `Unsupported`   | Unknown format, or one that cannot be converted                     |
-| `NeedsOcr`      | Scanned or image-only pages of a PDF, listed in `pages`             |
+| `NeedsOcr`      | Scanned or image-only PDF/OFD pages, listed in `pages`             |
 | `Malformed`     | Structurally unusable: no meaningful content could be extracted     |
 | `Encrypted`     | Encrypted or password-protected                                     |
 | `ResourceLimit` | Crossed a fixed safety limit (decompression, nesting, node count)   |
@@ -251,7 +254,7 @@ match anydoc::to_markdown(path) {
 
 Node and wasm publish the variant name on `error.code`; Python raises one `anydoc.ConvertError` subclass per variant, or `OSError` when the file cannot be read.
 
-anydoc converts locally and does not do OCR, so a PDF with scanned pages fails with `NeedsOcr`. Opt in with `ocr: 'hosted'` in Node, `ocr="hosted"` in Python or `--ocr hosted` on the CLI to send that document to [Firecrawl Parse](https://firecrawl.dev/parse). No signup needed. Set `FIRECRAWL_API_KEY` for higher limits.
+anydoc converts locally and does not do OCR, so a PDF or OFD with scanned pages fails with `NeedsOcr`. For PDFs, opt in with `ocr: 'hosted'` in Node, `ocr="hosted"` in Python or `--ocr hosted` on the CLI to send that document to [Firecrawl Parse](https://firecrawl.dev/parse). OFD pages continue to return `NeedsOcr` because this hosted mode uses a PDF parser. No signup needed. Set `FIRECRAWL_API_KEY` for higher limits.
 
 ## How it works
 

node/README.md

@@ -3,7 +3,7 @@
 [![npm](https://img.shields.io/npm/v/@firecrawl/anydoc.svg)](https://www.npmjs.com/package/@firecrawl/anydoc)
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](https://github.com/firecrawl/anydoc/blob/main/LICENSE)
 
-Convert Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF files into clean GitHub-Flavored Markdown. Node.js bindings for the [anydoc](https://github.com/firecrawl/anydoc) Rust crate, built by [Firecrawl](https://firecrawl.dev). Also available as a hosted API through [Firecrawl Parse](https://firecrawl.dev/parse), which adds our OCR models for the scanned pages anydoc can't read on its own.
+Convert Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, PDF, and OFD files into clean GitHub-Flavored Markdown. Node.js bindings for the [anydoc](https://github.com/firecrawl/anydoc) Rust crate, built by [Firecrawl](https://firecrawl.dev). Also available as a hosted API through [Firecrawl Parse](https://firecrawl.dev/parse), which adds our OCR models for the scanned pages anydoc can't read on its own.
 
 Every format parses into one shared document model and renders through a single Markdown serializer, so headings, tables, lists, and footnotes come out the same no matter which format goes in. Conversion runs on the libuv thread pool and never blocks the event loop. TypeScript types ship with the package.
 
@@ -23,6 +23,9 @@ npm install @firecrawl/anydoc
 | EPUB             | `.epub`                                                    |
 | CSV              | `.csv`                                                     |
 | PDF              | `.pdf`                                                     |
+| OFD              | `.ofd` (text objects)                                    |
+
+OFD conversion preserves outline headings, large-font headings, bold and italic text, URI links, printed lists, and complete bordered table grids. Tables and large-font headings are inferred from fixed-page geometry; pages without those cues remain paragraphs. Image-only pages return `needsOcr`; images, vector drawings, and non-link annotations are not included in Markdown.
 
 ## CLI
 
@@ -57,7 +60,7 @@ const document = await toDocument(bytes);
 
 ## Scanned pages
 
-anydoc converts locally and does not do OCR, so a PDF with scanned or image-only pages rejects with `needsOcr`. Opt in with `ocr: 'hosted'` to send that document to [Firecrawl Parse](https://firecrawl.dev/parse). No signup needed. Set `apiKey` or `FIRECRAWL_API_KEY` for higher limits.
+anydoc converts locally and does not do OCR, so a PDF or OFD with scanned or image-only pages rejects with `needsOcr`. For PDFs, opt in with `ocr: 'hosted'` to send that document to [Firecrawl Parse](https://firecrawl.dev/parse). This hosted PDF parser does not accept OFD; OFD pages still reject with `needsOcr` even when hosted OCR is requested. No signup needed. Set `apiKey` or `FIRECRAWL_API_KEY` for higher limits.
 
 ```js
 const markdown = await toMarkdown('scan.pdf', { ocr: 'hosted' });
@@ -91,7 +94,7 @@ try {
 | `code`          | Meaning                                                             |
 | --------------- | ------------------------------------------------------------------- |
 | `unsupported`   | Unknown format, or one that cannot be converted                     |
-| `needsOcr`      | Scanned or image-only pages of a PDF, listed in `pages`             |
+| `needsOcr`      | Scanned or image-only PDF/OFD pages, listed in `pages`             |
 | `malformed`     | Structurally unusable: no meaningful content could be extracted     |
 | `encrypted`     | Encrypted or password-protected                                     |
 | `resourceLimit` | Crossed a fixed safety limit (decompression, nesting, node count)   |

node/anydoc.js

@@ -22,7 +22,9 @@ async function toMarkdown(path, options) {
     return await native.toMarkdown(path)
   } catch (error) {
     if (!sendsToHosted(error, options)) throw error
-    return parseHosted(await readFile(path), basename(path), options)
+    const bytes = await readFile(path)
+    if ((native.formatFromBytes(bytes) ?? native.formatFromPath(path)) !== 'pdf') throw error
+    return parseHosted(bytes, basename(path), options)
   }
 }
 
@@ -33,6 +35,7 @@ async function toMarkdownBytes(bytes, format, options) {
     return await native.toMarkdownBytes(bytes, format)
   } catch (error) {
     if (!sendsToHosted(error, options)) throw error
+    if ((format ?? native.formatFromBytes(bytes)) !== 'pdf') throw error
     return parseHosted(bytes, 'document.pdf', options)
   }
 }

node/cli.js

@@ -3,7 +3,7 @@
 
 const { readFile, writeFile } = require('node:fs/promises')
 
-const FORMATS = 'doc, docx, odt, pdf, ppt, pptx, rtf, epub, xlsx, ods, odp, csv'
+const FORMATS = 'doc, docx, odt, pdf, ofd, ppt, pptx, rtf, epub, xlsx, ods, odp, csv'
 
 const HELP = `anydoc: convert documents to GitHub-Flavored Markdown
 
@@ -37,14 +37,14 @@ Options:
 The format is detected from the file content; the file extension is the
 fallback for signature-less formats (CSV). stdin has no extension, so CSV
 input from stdin needs --format csv. Scanned or image-only pages need OCR,
-which anydoc does not do: the document exits 3, or goes to Firecrawl Parse
-with --ocr hosted.
+which anydoc does not do: the document exits 3. PDF can go to Firecrawl
+Parse with --ocr hosted; OFD still exits 3.
 
 Exit codes:
   0  success
   1  the document could not be read or converted
   2  usage error: unknown option, missing input, or invalid --format
-  3  pages of a PDF need OCR
+  3  PDF or OFD pages need OCR
 
 Examples:
   anydoc report.docx

node/dts-header.d.ts

@@ -9,7 +9,7 @@ export type ConvertErrorCode =
   /** Unknown format, or one that cannot be converted. */
   | 'unsupported'
   /**
-   * Pages of a PDF are scanned or image-only and need OCR, which anydoc does
+   * PDF or OFD pages are scanned or image-only and need OCR, which anydoc does
    * not do. The error is a `NeedsOcrError` naming them.
    */
   | 'needsOcr'

node/index.d.ts

@@ -9,7 +9,7 @@ export type ConvertErrorCode =
   /** Unknown format, or one that cannot be converted. */
   | 'unsupported'
   /**
-   * Pages of a PDF are scanned or image-only and need OCR, which anydoc does
+   * PDF or OFD pages are scanned or image-only and need OCR, which anydoc does
    * not do. The error is a `NeedsOcrError` naming them.
    */
   | 'needsOcr'
@@ -124,6 +124,7 @@ export declare const enum Format {
    * `needsOcr` naming them.
    */
   pdf = 'pdf',
+  ofd = 'ofd',
   ppt = 'ppt',
   pptx = 'pptx',
   rtf = 'rtf',

node/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@firecrawl/anydoc",
   "version": "0.2.4",
-  "description": "Convert documents (doc, docx, odt, rtf, epub, pdf, presentations, spreadsheets, csv) to GitHub-Flavored Markdown",
+  "description": "Convert documents (doc, docx, odt, rtf, epub, pdf, ofd, presentations, spreadsheets, csv) to GitHub-Flavored Markdown",
   "license": "MIT",
   "homepage": "https://github.com/firecrawl/anydoc#readme",
   "repository": {

node/src/lib.rs

@@ -22,6 +22,7 @@ pub enum Format {
     /// need OCR, which anydoc does not do: the document rejects with
     /// `needsOcr` naming them.
     pdf,
+    ofd,
     ppt,
     pptx,
     rtf,
@@ -39,6 +40,7 @@ impl From<Format> for anydoc::Format {
             Format::docx => anydoc::Format::Docx,
             Format::odt => anydoc::Format::Odt,
             Format::pdf => anydoc::Format::Pdf,
+            Format::ofd => anydoc::Format::Ofd,
             Format::ppt => anydoc::Format::Ppt,
             Format::pptx => anydoc::Format::Pptx,
             Format::rtf => anydoc::Format::Rtf,
@@ -58,6 +60,7 @@ impl From<anydoc::Format> for Format {
             anydoc::Format::Docx => Format::docx,
             anydoc::Format::Odt => Format::odt,
             anydoc::Format::Pdf => Format::pdf,
+            anydoc::Format::Ofd => Format::ofd,
             anydoc::Format::Ppt => Format::ppt,
             anydoc::Format::Pptx => Format::pptx,
             anydoc::Format::Rtf => Format::rtf,

node/test.mjs

@@ -1,7 +1,7 @@
 // Smoke test: the bindings load and every entry point round-trips a fixture.
 import assert from 'node:assert/strict'
 import { execFile } from 'node:child_process'
-import { mkdtemp, readFile, rm } from 'node:fs/promises'
+import { mkdtemp, readFile, rm, writeFile } from 'node:fs/promises'
 import { createServer } from 'node:http'
 import { tmpdir } from 'node:os'
 import { join } from 'node:path'
@@ -25,6 +25,7 @@ const RICH = fixture('docx/handmade-rich.docx')
 const CSV = fixture('csv/sheet.csv')
 const ENCRYPTED = fixture('malformed/encrypted--errors.odt')
 const MIXED = fixture('pdf/handmade-mixed.pdf')
+const SCANNED_OFD = fixture('ofd/image-only.ofd')
 
 test('toMarkdown detects the format from the file content', async () => {
   const markdown = await toMarkdown(OUTLINE)
@@ -67,6 +68,7 @@ test('format detection reads content, extension, and path', async () => {
   assert.equal(formatFromBytes(await readFile(CSV)), null)
   assert.equal(formatFromExtension('.pptm'), 'pptx')
   assert.equal(formatFromExtension('xls'), 'xlsx')
+  assert.equal(formatFromExtension('ofd'), 'ofd')
   assert.equal(formatFromPath('/tmp/report.odt'), 'odt')
   assert.equal(formatFromPath('/tmp/report.unknown'), null)
 })
@@ -149,6 +151,39 @@ test("ocr: 'hosted' sends a pdf with scanned pages to Firecrawl Parse, and nothi
   })
 })
 
+test("ocr: 'hosted' leaves image-only OFD as needsOcr", async () => {
+  await withHostedStub(HOSTED, async (hits) => {
+    for (const convert of [
+      () => toMarkdown(SCANNED_OFD, { ocr: 'hosted' }),
+      async () => toMarkdownBytes(await readFile(SCANNED_OFD), 'ofd', { ocr: 'hosted' }),
+      async () => toMarkdownBytes(await readFile(SCANNED_OFD), undefined, { ocr: 'hosted' }),
+    ]) {
+      await assert.rejects(convert(), (error) => error.code === 'needsOcr')
+    }
+    assert.deepEqual(hits, [])
+  })
+})
+
+test('hosted fallback uses file content before a misleading extension', async () => {
+  const dir = await mkdtemp(join(tmpdir(), 'anydoc-hosted-'))
+  try {
+    const ofdNamedPdf = join(dir, 'scan.pdf')
+    const pdfNamedOfd = join(dir, 'scan.ofd')
+    await Promise.all([
+      writeFile(ofdNamedPdf, await readFile(SCANNED_OFD)),
+      writeFile(pdfNamedOfd, await readFile(MIXED)),
+    ])
+    await withHostedStub(HOSTED, async (hits) => {
+      await assert.rejects(toMarkdown(ofdNamedPdf, { ocr: 'hosted' }), (error) => error.code === 'needsOcr')
+      assert.deepEqual(hits, [])
+      assert.equal(await toMarkdown(pdfNamedOfd, { ocr: 'hosted' }), HOSTED.body.data.markdown)
+      assert.deepEqual(hits, [['/v2/parse', true]])
+    })
+  } finally {
+    await rm(dir, { recursive: true, force: true })
+  }
+})
+
 test("parseMode picks the Parse pdf mode, 'auto' by default", async () => {
   await withHostedStub(HOSTED, async (hits, url, sent) => {
     await toMarkdown(MIXED, { ocr: 'hosted' })
@@ -241,6 +276,15 @@ test('cli --ocr hosted converts a pdf with scanned pages through Firecrawl Parse
   })
 })
 
+test('cli --ocr hosted keeps image-only OFD at needs OCR exit code', async () => {
+  await withHostedStub(HOSTED, async (hits, url) => {
+    const { code, stderr } = await runCli([SCANNED_OFD, '--ocr', 'hosted', '--api-url', url])
+    assert.equal(code, 3)
+    assert.match(stderr, /needs OCR/)
+    assert.deepEqual(hits, [])
+  })
+})
+
 test('cli --parse-mode picks the Parse pdf mode', async () => {
   await withHostedStub(HOSTED, async (hits, url, sent) => {
     const { code } = await runCli([MIXED, '--ocr', 'hosted', '--parse-mode', 'ocr', '--api-url', url])