Batch PDF Processor (Workflows for Many PDFs)

About the Batch PDF Processor (Workflows for Many PDFs)

When a whole folder of PDFs needs the same treatment — scans that must become searchable, invoices that need a watermark and page numbers, statements that must be unlocked and then compressed for a portal — doing them one at a time is slow. Here you build a recipe once from the steps of MySmartCoPilot’s own PDF tools and run it on up to 100 files: unlock with a password you know, rotate pages, flatten forms and comments, OCR scans into searchable PDFs (English, Hindi, Bengali, Gujarati, Kannada, Malayalam, Marathi, Tamil, Telugu, Urdu and more), add a watermark and page numbers, compress to a size limit or quality level, and protect every file with one password.

Each step is the same engine as its single-file tool, so the results are the same too. Several files are processed at once in background workers, with a report for every file — pages, size before and after, what each step did, and any problem. Name the results by a pattern (original name, counter, date, or text found on page 1 such as an invoice number), then download one ZIP or save them into a folder on your computer. Recipes are saved in this browser; passwords never are. Nothing is uploaded.

Common uses

• Making a backlog of scanned documents searchable in Indian languages or English, then compressing them for archiving.
• Stamping “CONFIDENTIAL”, “DRAFT” or “COPY” and page numbers on every document of a case file or tender.
• Unlocking a month of bank or card statements with the password you know before importing or printing them.
• Protecting every report sent to a client with one password.
• Renaming invoices, policies or certificates by a number printed on their first page.

The steps, and why they run in this order

1. Unlock removes the password with qpdf, the engine of Unlock PDF. Files that open without a password but have restrictions are freed only with their owner password; nothing is guessed. Files that aren’t protected pass through.
2. Rotate turns all pages, or only landscape or only portrait ones, by 90° either way or by 180° — as a reader’s rotate command does, so nothing on the page is redrawn.
3. Flatten draws form fields, comments and visible layers into the pages, with Flatten PDF’s engine.
4. OCR reads scanned pages with Tesseract, as OCR PDF does, and adds an invisible text layer so the PDF can be searched and copied; pages that already have text are skipped.
5. Watermark and 6. Page numbers use the engines and fonts of Add Watermark and Add Page Numbers: choose Noto Sans for ₹, accented letters, Greek or Cyrillic. With Skip the first pages, a file that has no more pages than that is passed on without numbers, and its report says so.
6. Compress uses the PDF Compressor’s engine — with Compress PDF to Size’s search when you set a size limit.
7. Protect encrypts every file with AES-256 and one open password, with the engine of Password Protect PDF, and checks the result opens with it. Encryption adds a little to every file (usually a few kilobytes), so with a size limit Compress aims lower when the password would take a file over it.

The order is fixed so steps never undo each other: a password has to come off before anything else can change a file, OCR must read the pages before a watermark or numbers are drawn on them (it would read those too), compressing comes after everything that adds to the pages, and a new password goes on last.

File names

The name pattern can mix your own text with these parts: - {name} — the original file name, without .pdf - {n} — a counter (1, 2, 3 … padded to the number of files, so the names sort in order) - {date} — the day of the run, as YYYY-MM-DD - {text} — the first line of text on page 1 (after OCR, so scans have one too) - {after:words} — what follows those words on page 1, up to the next gap or the end of the line, for example {after:Invoice No} or {after:Policy number} (the words must match whole: “Invoice No” isn’t found in “Invoice Notes”)

Characters that file systems don’t allow are replaced with a hyphen, and if two results get the same name the second becomes “name (2).pdf”. When {text} or {after:…} finds nothing on page 1, that file keeps its original name and the report says so. Files from subfolders keep their subfolder in the ZIP. You can change the pattern after a run; the names update without running the recipe again.

Saving the results

• Download ZIP puts every finished file in one ZIP, stored as is (PDFs are compressed already).
• Save to a folder (with a Pro or Business pass, in Chrome, Edge and other Chromium browsers on computers) writes the files into a folder you pick — the one you processed, a subfolder or another. Existing files are never overwritten: a name that is taken gets “(2)”.
• Download report gives a CSV with each file’s name, pages, sizes, result and what every step did.

How many files at once

Files are processed side by side in background workers: up to three at a time on a computer with enough processor cores, two when the recipe runs OCR (it is the heaviest step), and fewer on phones and devices with little memory. Files of 60 MB or more run on their own. Cancel stops at once: files in progress are left as they were, and files that were already done stay done.

Limitations

• Up to 100 PDFs per run, each up to 200 MB and 2 GB together; on phones, smaller batches are faster.
• The watermark step adds text watermarks (with {filename}, {date}, {page} and {total}); for an image watermark, use Add Watermark on each file. Scripts that need joined letters or vowel signs, such as Devanagari, Tamil or Arabic, can’t be drawn as watermark or page-number text — an image of the text in Add Watermark can.
• Unlock only works with a password you know; restrictions are removed only with the owner password.
• A PDF with a password or editing restrictions goes through the other steps only after the Unlock step has removed them; without it, the file is listed as failed with the reason.
• Any change to a digitally signed PDF invalidates its signature; the report notes it.
• OCR is the slowest step, especially on phones, and the first run downloads the OCR engine and the language data once (about 7 MB for English).
• Choosing a folder works in Chromium browsers on computers, and so does saving into one, with a Pro or Business pass; elsewhere, drop a folder or choose the files, and download the ZIP.

Privacy

Every step runs in your browser. The PDFs, the passwords you type and the results stay on your device: nothing is uploaded. The recipe you are working on and the recipes you save are kept in this browser’s local storage, without any password. The PDF engines, and for OCR the recognition engine and its language data, are downloaded from this site the first time a recipe needs them.