pdf

package

v0.5.7 Latest Latest Go to latest Published: Jul 28, 2026 License: AGPL-3.0 Imports: 19 Imported by: 0

Details

Valid go.mod file
Redistributable license
Tagged version
Stable version
Learn more about best practices

Repository

github.com/get-vix/vix

Links

Open Source Insights

Documentation ¶

Overview ¶

Package pdf is a self-contained, dependency-free PDF reader that extracts a text-based PDF's content and renders it as Markdown for LLM consumption.

Scope (v1): digital/text-based PDFs. It parses the classic cross-reference table as well as PDF 1.5+ cross-reference streams and compressed object streams, inflates FlateDecode content, interprets the text-showing operators of page content streams, maps character codes to Unicode via ToUnicode CMaps and the standard simple-font encodings, then reconstructs paragraphs, headings and best-effort tables.

Out of scope: scanned/image-only PDFs (no text layer — OCR is not attempted) and encrypted PDFs.

Constants ¶

This section is empty.

Variables ¶

View Source

var ErrEncrypted = errors.New("pdf: encrypted document not supported")

ErrEncrypted is returned when the PDF is encrypted and cannot be unlocked with the empty user password — i.e. it genuinely requires a password (or uses an unsupported security handler). Empty-password/permissions-only encryption is decrypted transparently and does not surface this error.

Functions ¶

func Float ¶

func Float(o Object) (float64, bool)

Float returns the numeric value of an Integer or Real, and whether the object was numeric at all.

func Int ¶

func Int(o Object) (int, bool)

Int returns the integer value of an Integer (or truncated Real), and whether the object was numeric.

Types ¶

type Array ¶

type Array []Object

Array is a PDF array object.

type Boolean ¶

type Boolean bool

Boolean is a PDF boolean (true/false).

type Dict ¶

type Dict map[Name]Object

Dict is a PDF dictionary object.

type Document ¶

type Document struct {
	// contains filtered or unexported fields
}

Document is a parsed PDF file: its cross-reference table, trailer and a cache of resolved indirect objects.

func Parse ¶

func Parse(data []byte) (*Document, error)

Parse reads a PDF file's structure from raw bytes.

func (*Document) Resolve ¶

func (d *Document) Resolve(o Object) Object

Resolve dereferences an indirect reference (recursively for chained refs), returning the concrete object. Non-references are returned as-is.

type Integer ¶

type Integer int64

Integer is a PDF integer object.

type Name ¶

type Name string

Name is a PDF name object without the leading slash (e.g. "Type", "Font").

type Null ¶

type Null struct{}

Null is the PDF null object.

type Object ¶

type Object interface{}

Object is any PDF object. The concrete types below form a closed set; callers type-switch over them.

type Real ¶

type Real float64

Real is a PDF real (floating point) object.

type Reference ¶

type Reference struct {
	Num int
	Gen int
}

Reference is an indirect object reference ("12 0 R").

type Result ¶

type Result struct {
	Markdown string // rendered Markdown ("" when Scanned)
	Pages    int    // number of pages walked
	Scanned  bool   // true when no extractable text layer was found
}

Result is the outcome of converting a PDF to Markdown.

func ToMarkdown ¶

func ToMarkdown(data []byte) (Result, error)

ToMarkdown parses a PDF and renders its text content as Markdown. It returns ErrEncrypted for encrypted documents. For image-only/scanned PDFs it returns a Result with Scanned=true and empty Markdown (OCR is not attempted).

type Stream ¶

type Stream struct {
	Dict Dict
	Raw  []byte
}

Stream is a PDF stream object: its dictionary plus the raw (still-encoded) stream bytes. Decoding happens on demand via the filter pipeline.

type String ¶

type String []byte

String is a PDF string object, already decoded from its literal "(...)" or hexadecimal "<...>" representation into raw bytes.

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL