How does bookmark-based PDF splitting read document outlines?
Bookmark-based splitting parses the PDF catalog’s /Outlines tree to locate the exact page index of each chapter heading, then cuts the document at those boundaries. Reading structural metadata directly avoids guessing page numbers from printed footers or Roman numerals.
Resolving /Dest arrays, named destinations, and /GoTo actions
Under the ISO 32000-1 PDF 1.7 Specification, an outline item points to a page either through a direct /Dest entry or an /A action dictionary with subtype /GoTo. Named destinations are resolved through both legacy /Dests dictionaries and /Names /Dests name trees up to 32 levels deep.
The parser reads up to 5000 outline items across levels 1 and 2, skipping malformed entries that lack a valid page target. If your file has no outline metadata at all, you can divide pages by custom ranges instead.
Grouping Front matter and deduplicating shared start pages
When the first bookmark starts on page 4 or later, pages 1 through 3 (covers, title sheets, and preface tables) are collected into a Front matter group. Multiple bookmarks that jump to the exact same page index are deduplicated so every output PDF contains at least 1 page.
If two subsection headings appear halfway down the same page, keeping only the first bookmark at that page index prevents duplicate copies of the same sheet from appearing in your ZIP archive. You can inspect every resulting group and page span in the preview list before clicking Split by bookmarks.
Choosing top-level chapters vs. second-level sections
Technical manuals, legal briefs, and ebooks often nest subsections underneath main chapter headings. When level-2 entries exist in the outline tree, the Split at selector lets you control how granular the output files should be.
| Split setting | Outline levels read | Typical output on a 120-page manual |
Number of files | Best suited for |
|---|---|---|---|---|
| Top-level bookmarks (chapters) | Level 1 only |
Front matter, Chapter 1, Chapter 2 |
Fewer, larger PDFs | Sharing whole chapters or modules |
| Second level too (sections) | Levels 1 and 2 |
1.1 Setup, 1.2 Config, 2.1 API |
Many focused PDFs | Extracting individual SOPs or lessons |
| Single-page bookmark | Any level | Shown as p. 12 in preview |
1-page PDF |
Isolated appendix or reference charts |
| Multi-page bookmark | Any level | Shown as pp. 14-29 in preview |
16-page PDF |
Full-length sections and essays |
| Fewer than 2 groups | Any level | Split by bookmarks stays disabled | 0 files |
Redirects to range or text splitters |
Per the Library of Congress PDF 2.0 Format Description, document outlines provide structural navigation independently of printed page labels. If you only need one specific chapter rather than the whole book, try a single-chapter page extractor to pull out that range alone.
Why do some PDFs with a table of contents have no bookmarks?
Many PDFs print a visual “Table of Contents” on page 2 or 3 without embedding an /Outlines dictionary in the PDF catalog. Because those printed lines are ordinary text glyphs rather than sidebar metadata, an outline parser finds zero bookmarks to cut on.
Printed TOC text vs. sidebar /Outlines metadata
When you open a PDF in Chrome, Preview, or Acrobat, look at the left navigation sidebar. If the bookmark icon is empty, the exporting software flattened the layout to print graphics without generating /Outlines nodes.
Tip: Always check the live file preview list before exporting. It shows the exact number of output files and each chapter’s
p.orpp.range immediately after you drop the file.
Splitting unbookmarked manuals by heading text or size
When a PDF lacks sidebar bookmarks but uses consistent chapter titles on every section’s opening page, Split PDF by text can scan page text for that heading pattern and cut there automatically.
Alternatively, if a scanned textbook was exported as flat images without either embedded outlines or selectable text, splitting by manual page ranges gives you full control over where each chapter begins and ends.
Numbered chapter exports and browser-local privacy
Large training manuals, compliance binders, and course packs often contain confidential internal procedures. Splitting chapters inside your browser keeps proprietary documentation off external servers while organizing every section cleanly.
How output filenames and ZIP bundles are structured
Each split chapter is numbered in order and named after its sanitized bookmark title so files sort naturally in any folder. Because bookmark splitting requires at least 2 groups, outputs bundle into a single archive following the PKWARE ZIP File Format Specification.
All outline scanning and page copying run inside a Web Worker with 0 bytes sent to any server, protecting unpublished manuscripts, legal binders, and internal training manuals. Even large 500-page reference guides with hundreds of nested bookmarks split locally without account limits, sign-in gates, or file-size restrictions.
Splitting by file size when chapters exceed upload caps
If a single chapter still exceeds an email or LMS upload limit after splitting by outline, you can run that chapter through a size-based splitter to stay under email attachment caps without manually counting pages.
Each exported chapter preserves original fonts, vector diagrams, and internal hyperlinks within that chapter’s page range, ready for immediate distribution, printing, or long-term archival. Even when splitting a 25 MB textbook into 12 separate chapters, the entire job finishes in local browser memory without server queues, upload delays, or watermarks.