feat: accept passwords for encrypted OOXML files

Cargo.toml

@@ -28,6 +28,7 @@ csv = "1.4.0"
 flate2 = "1"
 encoding_rs = "0.8.35"
 log = "0.4"
+office-crypto = "0.3.0"
 pdf-inspector = "1.14.2"
 quick-xml = "0.41.0"
 zip = { version = "8.6.0", default-features = false, features = ["deflate"] }

README.md

@@ -76,6 +76,9 @@ markdown = anydoc.to_markdown_bytes(data)
 # Or name it, which signature-less formats (CSV) need:
 markdown = anydoc.to_markdown_bytes(data, "csv")
 
+# Password-protected OOXML
+markdown = anydoc.to_markdown_bytes(data, password="secret")
+
 # Or stop at the document model, which also carries embedded assets:
 document = anydoc.to_document(data)
 ```
@@ -121,6 +124,9 @@ let markdown = anydoc::to_markdown_bytes(&bytes, None)?;
 // Or name it, which signature-less formats (CSV) need:
 let markdown = anydoc::to_markdown_bytes(&bytes, anydoc::Format::Csv)?;
 
+// Password-protected OOXML
+let markdown = anydoc::to_markdown_bytes_with_password(&bytes, None, "secret")?;
+
 // Or stop at the document model, which also carries embedded assets:
 let document = anydoc::to_document(&bytes, None)?;
 ```

node/README.md

@@ -32,8 +32,12 @@ The package ships an `anydoc` command, so `npx` converts a document with no inst
 npx @firecrawl/anydoc report.docx               # Markdown to stdout
 npx @firecrawl/anydoc slides.pptx -o slides.md  # or to a file
 npx @firecrawl/anydoc - --format csv < data.csv # read stdin
+npx @firecrawl/anydoc report.docx --password secret
 ```
 
+Set `ANYDOC_PASSWORD` instead of using `--password` when the password should
+not appear in shell history or the process list.
+
 Markdown goes to stdout, errors to stderr, and `anydoc --help` covers the rest.
 
 ## Usage

node/cli.js

@@ -13,14 +13,15 @@ Usage:
 
 Converts one document per invocation and writes the Markdown to stdout.
 Pass - as the input to read the document from stdin. Never prompts; all
-diagnostics go to stderr.
+diagnostics go to stderr. ANYDOC_PASSWORD is used when --password is omitted.
 
 Options:
   -o, --output <path>    Write the Markdown to <path> instead of stdout
   -f, --format <format>  Name the input format instead of detecting it:
                          ${FORMATS}
                          (extension aliases like xls, docm, ppsx resolve
                          to these)
+  -p, --password <value> Decrypt a password-protected OOXML document
   -h, --help             Print this help and exit
   -V, --version          Print the version and exit
 
@@ -50,7 +51,7 @@ function fail(code, message) {
 }
 
 function parseArgs(argv) {
-  const args = { input: null, output: null, format: null }
+  const args = { input: null, output: null, format: null, password: null }
   let positionalOnly = false
   for (let i = 0; i < argv.length; i++) {
     let arg = argv[i]
@@ -95,6 +96,10 @@ function parseArgs(argv) {
       case '--format':
         args.format = value()
         break
+      case '-p':
+      case '--password':
+        args.password = value()
+        break
       default:
         fail(USAGE_ERROR, `unknown option '${arg}' (see anydoc --help)`)
     }
@@ -133,12 +138,13 @@ async function main() {
 
   let markdown
   try {
+    const password = args.password ?? process.env.ANYDOC_PASSWORD
     if (args.input === '-') {
-      markdown = await toMarkdownBytes(await readStdin(), format)
+      markdown = await toMarkdownBytes(await readStdin(), format, password)
     } else if (format !== undefined) {
-      markdown = await toMarkdownBytes(await readFile(args.input), format)
+      markdown = await toMarkdownBytes(await readFile(args.input), format, password)
     } else {
-      markdown = await toMarkdown(args.input)
+      markdown = await toMarkdown(args.input, password)
     }
   } catch (error) {
     fail(CONVERSION_ERROR, error.message)

node/index.d.ts

@@ -271,7 +271,7 @@ export declare const enum TableKind {
  *
  * Rejects with an `Error` carrying a `ConvertErrorCode` on `code`.
  */
-export declare function toDocument(bytes: Uint8Array, format?: Format | undefined | null): Promise<Document>
+export declare function toDocument(bytes: Uint8Array, format?: Format | undefined | null, password?: string | undefined | null): Promise<Document>
 
 /**
  * Convert a document file to Markdown. The format is detected from the file
@@ -281,7 +281,7 @@ export declare function toDocument(bytes: Uint8Array, format?: Format | undefine
  * Rejects with an `Error` carrying a `ConvertErrorCode` on `code`; a file
  * that cannot be read is `'io'`.
  */
-export declare function toMarkdown(path: string): Promise<string>
+export declare function toMarkdown(path: string, password?: string | undefined | null): Promise<string>
 
 /**
  * Convert an in-memory document to Markdown. Without a format, it is
@@ -290,4 +290,4 @@ export declare function toMarkdown(path: string): Promise<string>
  *
  * Rejects with an `Error` carrying a `ConvertErrorCode` on `code`.
  */
-export declare function toMarkdownBytes(bytes: Uint8Array, format?: Format | undefined | null): Promise<string>
+export declare function toMarkdownBytes(bytes: Uint8Array, format?: Format | undefined | null, password?: string | undefined | null): Promise<string>