# Repair Corrupted PDF Files Before You Convert Them

- Author: FastBankConvert (https://fastbankconvert.com)
- Published: 2026-10-17
- Category: Difficult and locked PDFs
- Canonical: https://fastbankconvert.com/blog/repair-corrupted-pdf

> How to repair a corrupted PDF before conversion: re-download, open in a second viewer, print to a new PDF, and when the data is gone for good.
To repair a corrupted PDF before converting it, work from cheapest to hardest: download a fresh copy, open it in a second viewer, print it to a new PDF, then rebuild it with a repair tool such as qpdf. A file cut short in transfer may never recover.

## What breaks inside a corrupted PDF file?

A corrupted PDF is a file whose internal structure no longer matches what a reader expects. A PDF is a set of numbered objects, such as pages, fonts and compressed page contents, plus an index at the end that says where each object starts. Damage to any of these shows up differently.

Three kinds of damage cover most cases:

- **A broken index.** The cross-reference table, the index near the end of the file, points to the wrong places. This often happens when a file is edited and saved by a tool that miscounts bytes. The objects themselves are fine, so this is the most repairable kind.
- **A truncated file.** The download or email transfer stopped early, so the end of the file is missing. The index lives at the end, so readers cannot find anything. Pages near the start may still be recoverable, but anything past the cut is simply not there.
- **Damaged contents.** A compressed stream, the block that holds a page's drawing instructions, has bad bytes inside it. The page may open blank, show half its text or stop the reader with an error.

The symptoms vary by reader, but some are typical. An error such as "the file is damaged and could not be repaired" points to a broken index or a short file. A file that opens with blank pages, or with the first page fine and the rest empty, points to damaged contents or a cut. A file of zero bytes, or one much smaller than other statements from the same bank, was never fully saved.

Text that opens fine but copies as boxes and symbols is a different problem. The file is intact; its fonts just lack a map from shapes to letters. [Why PDF copy and paste turns text into symbols](/blog/pdf-copy-paste-gibberish) covers that case, and no repair tool will fix it.

## Re-downloading a corrupted PDF statement first

A fresh download is the cheapest repair, because statements are often damaged in transit rather than at the bank. The file on the bank's server is usually fine. Your copy stopped early, or a mail filter or sync folder altered it on the way.

Before downloading again, look at what you have. Check the file size against another statement from the same account. A statement that is normally around 200 KB but arrived as 40 KB was cut short. Then get a new copy in a way that avoids the first failure:

1. **Download from the bank's site again,** not from an email attachment or a shared folder copy.
2. **Use a different browser** if the first one opened the file in its built-in viewer and saved that view rather than the original.
3. **Wait for the download to finish** before opening it. A file opened while it is still arriving can be locked half-written.
4. **Save to a local folder** rather than straight into a cloud-synced one, then move it once it opens.
5. **Compare the new size** with the old. If it is larger and opens, the old copy was truncated.

If the bank sends statements by email only, ask them to resend it, or download it from the online banking archive if they keep one.

## Does a second viewer open the corrupted PDF?

A second viewer often opens a file the first refuses, because readers differ in how hard they try. Some stop at the first structural error. Others rebuild the broken index in memory and show the pages anyway. If any viewer opens the file, you have a route to a clean copy.

Try viewers that use different engines. A web browser, a desktop PDF reader and the preview app built into your operating system are usually three separate pieces of code. Open the file in each. Note which one shows every page, and whether the text can be selected there.

A viewer that opens the file silently repairs it in memory. Saving from that viewer may write the repaired version back to disk. Use Save As with a new name rather than Save, so the original stays untouched. Then close everything and open the new copy in the viewer that failed first. If it opens there too, the repair held.

One warning: some viewers show a damaged page as blank without any error. Count the pages and scroll to the end before you trust a viewer's result. A file that seems to open but stops at page 3 of 6 has a cut that no viewer can fill.

## Printing a corrupted PDF to a new PDF file

Printing to PDF builds a completely new file from what the viewer draws on screen. If a viewer can show every page, print to PDF from it. The new file has a clean structure, because the print driver writes it from scratch rather than copying the damaged parts.

Printing has costs, and they matter for conversion:

- **Text may become a picture.** Most built-in PDF printers keep real text, but some drivers turn each page into an image. A converter then has to read it by OCR, which is slower and less exact than reading text.
- **Links, bookmarks and form fields are lost.** For a statement this rarely matters.
- **Page size can change.** A print set to fit the paper may shrink or crop the page. Check that the right edge, where amounts and balances sit, is still there.

After printing, open the new file and drag across a line of transactions. If the highlight picks out individual words, the text survived. If it draws one box over the whole page, the page became an image. Both can be converted, but the text version is better. Scale and margin settings matter here as much as for any statement page, and [printing a bank statement to PDF so the text survives](/blog/print-to-pdf-bank-statement) covers them.

When the statement came as several damaged files, repair each one before joining them. [Combining PDF files into one before you convert](/blog/combine-pdf-files-into-one) explains how to merge them in order without duplicating pages.

## Rebuilding a corrupted PDF with qpdf

qpdf is a free command-line tool that checks a PDF's structure and writes a rebuilt copy. According to the [qpdf command-line documentation](https://qpdf.readthedocs.io/en/stable/cli.html), it tries to reconstruct a damaged cross-reference table by default, and its check option reports whether a file is valid.

A short session looks like this:

```
qpdf --check statement.pdf
qpdf statement.pdf statement-fixed.pdf
```

The first command inspects the file. The documentation says an exit status of 0 means the file is syntactically valid, 2 means errors were found, and 3 means only warnings. The second command reads the file, recovering what it can, and writes a new copy with a fresh index.

Read the result rather than trusting it. The same documentation notes that recovery relies on heuristics that do not work on every file, and that qpdf is not always successful when it issues warnings. A rebuilt file can open with every page present, or it can open with a page missing that the old index used to point to. Compare the page count with what the statement should have. Then compare the last transaction and closing balance with any copy you trust, such as the online banking screen.

The qpdf route suits someone comfortable with a terminal. If that is not you, opening the file in a second viewer or printing it to a new PDF reaches the same result for most files.

## Corrupted PDF pages whose data is gone for good

Some corruption cannot be repaired because the data is no longer in the file. A truncated download is missing its end, and no tool can rebuild bytes that never arrived. When every route shows the same blank pages or the same cut-off point, stop repairing and get the statement another way.

Signs that the data is gone:

- every viewer and qpdf stop at the same page
- the file is a fraction of the normal size for that account
- the file is zero bytes or opens as plain text with a few garbled lines
- the rebuilt copy shows pages, but they are blank where transactions should be

At that point the sources are, in order of preference: a fresh download from the bank, a copy the bank emails again, a printed statement you scan, or the transaction list on the online banking screen. A scan or a screen copy converts too, with more checking afterwards.

### A repair ladder, step by step

Use this as a checklist. Stop at the first step that gives you a file where every page opens and the text can be selected.

| Step | What to do | Fixes | Example sign it worked |
| --- | --- | --- | --- |
| 1 | Download a fresh copy from the bank | Truncated or altered transfers | New file is larger and opens |
| 2 | Open in a second viewer and Save As | Broken index | Every page shows, text selects |
| 3 | Print to a new PDF from that viewer | Broken index, some damaged contents | New file opens everywhere |
| 4 | Rebuild with qpdf | Broken index | Check reports no errors |
| 5 | Get the statement another way | Data missing from the file | Not applicable: the file is replaced |

Whichever step works, check the repaired file before you trust any numbers from it. The opening balance plus credits minus debits should equal the closing balance, and [the balance check for a PDF bank statement in Excel](/blog/pdf-bank-statement-to-excel) shows how to run it in a sheet.

FastBankConvert is built to warn you before download when a statement's balances do not add up after conversion, which is how a page lost to corruption tends to show itself. It is getting ready to open; [join through Get started](/get-started) to hear when trial conversions are available.