PDF File Format Explained: A Complete Guide

A PDF file has four main components: a header, a body, a cross-reference table, and a trailer. The header identifies the file as a PDF and specifies the version (e.g., %PDF-1.7 or %PDF-2.0). Modern PDFs use version 1.7 or 2.0, which support the most current features including AES-256 encryption and advanced color management. The body contains all the objects that make up the document: pages, text streams, images, fonts, annotations, bookmarks, and form fields. Each object has a unique identifier. Objects reference each other by ID, building the complex structure of a complete document. The cross-reference table (xref) provides byte offsets for every object in the file. When a PDF reader opens a document, it reads the xref table first to know where to find each object without reading the entire file sequentially. This is why large PDFs can jump to any page instantly. The trailer points to the xref table location and contains key document information. When you modify a PDF, a new xref table is appended at the end rather than rewriting the entire file — a design that keeps incremental updates efficient.

1. 1The PDF header identifies version — %PDF-1.7 is the most common, %PDF-2.0 is the current standard.
2. 2The body contains all document objects: pages, images, fonts, text, and annotations.
3. 3The cross-reference table allows random access to any object without reading the whole file.
4. 4Incremental updates append changes to the end of the file rather than rewriting it entirely.

Text in a PDF is not stored the way it is in a word processor. Rather than storing 'this is a paragraph in Helvetica at 12pt,' a PDF stores precise instructions: 'at position (100, 700), set font to Helvetica-Bold at 14pt, and draw the characters H, e, l, l, o.' Text is stored in content streams — sequences of PDF operators and their arguments. The font resource is referenced by name. The actual font data — the shapes of the characters — may be embedded in the PDF or referenced externally. Font embedding is crucial for consistent rendering. When a font is embedded, the PDF contains all the information needed to draw every character, regardless of whether that font is installed on the reader's system. When a font is not embedded, the reader substitutes a similar font — which may change text metrics, reflow content, or produce incorrect character rendering for special characters. Since PDF 1.2, Unicode mapping has allowed the actual Unicode text value of displayed characters to be stored alongside the rendering instructions. This is what makes text selectable and searchable. Older PDFs, and PDFs created by some software, may not include proper Unicode mapping — meaning you can see the text but cannot select or search it.

Images in PDF are stored as XObject resources — reusable components that can be referenced multiple times within a document. An image appears on a page by being painted at a specified position and scale using a PDF content stream operator. Images can use several compression methods within PDF: DCT (JPEG) for photographic content, Flate (ZIP-based deflate) for lossless compression, JBIG2 for binary (black/white) images, and CCITT Group 4 for facsimile-style binary images. The choice of compression affects both file size and visual quality. Image resolution in PDF is measured in DPI (dots per inch) but is fundamentally about pixel dimensions relative to the display size. A 3000x2000 pixel image displayed at 10 inches by 6.67 inches has an effective resolution of 300 DPI. The same image displayed at 30 inches by 20 inches has an effective resolution of 100 DPI. Transparency — allowing images and graphics to blend with content beneath them — uses the PDF transparency model. Transparency groups and alpha channels add complexity to rendering but enable sophisticated visual effects. Transparency is one of the reasons some PDFs are slow to render or produce unexpected results when printed on older PostScript printers.

Every PDF contains metadata — information about the document stored outside the visible content. Standard metadata fields include title, author, subject, keywords, creator (application that created the document), producer (application that generated the PDF), creation date, and modification date. Metadata can be stored in two places: the Document Information Dictionary (the older method) and XMP (Extensible Metadata Platform) streams (the newer standard). Modern PDFs typically store metadata in both locations for compatibility. XMP metadata can contain additional information beyond the standard fields: copyright notices, licensing terms, GPS coordinates (relevant for scanned documents from mobile devices), revision history, and custom application-specific data. This metadata is often more revealing than document creators realize. A PDF might show the original author's name even if the document appears anonymous, the law firm's document management system version, or the GPS location where a mobile scan was taken. Stripping metadata before distribution is a privacy best practice for sensitive documents.

PDFs can contain interactive elements that go far beyond static display. Form fields allow data entry — text fields, checkboxes, radio buttons, dropdown menus, and signature fields. Form data can be submitted, exported, and imported. Annotations add comments, highlighting, sticky notes, and markup to PDF pages without modifying the underlying content. This is how PDF review workflows function — reviewers add annotations, the document owner responds to them, and the base document remains unchanged. JavaScript in PDFs enables dynamic behavior: validation of form inputs, calculations based on field values, and user interface interactions. JavaScript PDFs are also a security concern — malicious JavaScript has been used in PDF-based attacks. For this reason, many PDF readers disable JavaScript by default or prompt before execution. AcroForms is the standard PDF form specification. XFA (XML Forms Architecture) is an older Adobe-proprietary form format that is not supported by non-Adobe PDF readers and is deprecated in PDF 2.0.

PDF has evolved through multiple versions, each adding capabilities. PDF 1.0 through 1.4 added progressively richer features. PDF 1.7 became an ISO standard (ISO 32000-1) in 2008, removing Adobe's proprietary control. PDF 2.0 (ISO 32000-2) was published in 2017 with improved encryption, better accessibility support, and cleaner specification. Specialized PDF variants serve specific purposes. PDF/A is the archiving standard — it prohibits features that could impair long-term readability and requires font embedding and embedded color profiles. PDF/X is the prepress standard for print production. PDF/UA is the accessibility standard, ensuring documents are usable with screen readers. PDF/E targets engineering workflows. For most everyday use, standard PDF 1.7 compatibility is ideal — widely supported and feature-complete for common needs. PDF/A is appropriate for documents that must remain readable for years or decades.