PDF is page-addressable. In theory you can pull page 738 out of a 1,000-page document with precision. In practice the toolchain trips over real questions: does page numbering start from 0 or 1? How do you write “1 to 3 plus 7 to 9”? After extraction, is the original file still there?

This article skips theory and goes straight to the method plus the 5 page-range notations you’ll actually use. All operations run locally in your browser — files are never uploaded.

30-Second Overview

PDF page extraction has two paths:

  1. Range mode — input 1-3,5,7-9, output a single PDF containing only those pages. Use when you’re “pulling a few pages together”.
  2. Split every page — output a ZIP with N single-page PDFs. Use when “each page goes to a separate archive entry”.

The technical implementation is identical; only the packaging differs. Which one to pick depends on how you’ll use the output — sending via WeChat is Range mode (one file), archiving to cloud drive is Every-page mode (one file per page makes indexing easier).

5 Real Page-Range Notations

1. Simple range — 1-3

Extract pages 1 through 3 (3 pages total). Output: report-pages-1-3.pdf.

Pitfall: Some PDF editors interpret 1-3 as “1, 2, 3” (both ends inclusive). Others interpret it as “1, 2” (left-closed, right-open). The standard is both ends inclusive — our tool follows that.

2. Multiple ranges — 1-3,5,7-9

Extract pages 1-3, plus page 5, plus pages 7-9 (7 pages total).

Pitfall: The comma must be a half-width English comma. If you write 1-3,5,7-9 with a full-width comma, you’ll get an “invalid syntax” error.

3. Reverse order — 5,1-3

Extract page 5 first, then pages 1, 2, 3 in order. Useful when you want “summary first, then acknowledgements last”.

Pitfall: Some PDF tools auto-sort 5,1-3 into ascending 1-5, losing the user’s reverse intent. Our tool preserves the order you typed.

4. Reverse range — 3-1

Tool auto-normalizes to 1-3. Even if you typed it backwards, you won’t get an error.

5. Duplicate range — 1-3,2

Tool auto-dedupes to 1-3. Duplicates won’t double the output.

3 Real Scenarios

Scenario 1: Pull out the signature page from a 30-page contract, send to legal

Typical input: 30-page contract, signature on page 28.

How to do it: Use PDF Tools — Split mode, input 28, output contract-page-28.pdf.

Pitfall: A contract’s “page number” and “physical page index” often disagree — the cover and TOC may use Roman numerals (i, ii, iii) and the body uses Arabic (1, 2, 3). The tool operates by physical page index, ignoring PDF’s displayed page label. So the page you want as “page 28” may be the tool’s understanding of “physical page 30”. Verify with your PDF reader: flip to the signature page and read the bottom status bar “3 / 30”.

Scenario 2: Split a 200-page paper into chapter-zips for co-authors

Typical input: 200-page paper, each co-author handles 30-50 pages.

How to do it: Use “Split every page” mode, output a single ZIP with paper-pages-001.pdf … paper-pages-200.pdf. Hand the ZIP to co-authors and they pick their section.

Pitfall: Single-page PDF filenames are zero-padded by digit count — under 100 pages output 001.pdf, over 100 pages output 001.pdf through 200.pdf (3 digits). This avoids the dictionary-order trap where 10.pdf sorts before 2.pdf.

Scenario 3: Take page 1 from 5 different PDFs and combine them into a “table of contents” PDF

Typical input: 5 PDFs, each contributing its first page → a 5-page table-of-contents PDF.

How to do it: This is merge-after-extract. With our tool, first extract 1 from each (each output is 1 page), then merge the 5 into one. Or skip our tool and write a script using pdf-lib, but the manual approach works — just two more steps.

Pitfall: If the original PDFs are “scan-only” (each page is an image), extraction succeeds but search-through-text on the extracted page is empty. For that case, OCR is the right tool; extracting a PDF page doesn’t solve it.

5 Common Mistakes and Fixes

Mistake 1: Assuming page-number = “the number printed at the top of the PDF”

Fix: Use your PDF reader’s bottom status bar — “3 / 30” — that’s the physical page index.

Mistake 2: Entering huge ranges (e.g. 10,000 pages)

Fix: Check the PDF’s actual page count first. Out-of-range will error out — you won’t silently get an empty file.

Mistake 3: Trying to split encrypted PDFs

Fix: Encrypted PDFs need a decryption password; our tool doesn’t support that. If it’s your PDF, remove encryption in the software that generated it. If it’s someone else’s, ask them for an unprotected copy.

Mistake 4: Splitting a very large PDF (over 100 MB)

Fix: Our 100 MB cap exists to prevent browser tab crashes. If your PDF really is that big, first use a desktop tool (like pdftk) to chunk it into 50-page pieces, then use ours for fine-grained splitting.

Mistake 5: Forgetting to check how many files “Split every page” will produce

Splitting a 200-page PDF with “Every page” outputs 200 files + 1 ZIP. Browsers can download 200 small files fine, but if your cloud drive has bandwidth limits, plan accordingly.

  1. Check page count first — most PDF readers show total pages at a glance. Plan output size before splitting a 50+ page document.
  2. Test 1 page first — when first splitting a new PDF, extract page 1 to verify the page-index numbering and naming convention, then batch-split.
  3. Keep the originals — extraction produces new PDFs; the original file stays untouched.
  4. Naming convention — extracted files should follow originalname-pages-1-3.pdf format. One glance tells you what it is.

Try our PDF Tools — Split mode. The range input accepts 1-3,5,7-9 directly.