Aspose.PDF.FOSS 26.6.0

Aspose.PDF FOSS for .NET

A free, open-source PDF library for .NET 8+ — read, create, modify, and convert PDF documents.

Note: Some features are incomplete and the public API may change between releases — see What's not included for current gaps.

Platform note: Runs on .NET 8+ across Windows, Linux, and macOS. Core document work — parsing, editing, text, forms, encryption/signing — and page→image rendering (via the built-in managed SoftwarePageRenderer, used automatically off-Windows) are pure-managed and fully cross-platform. A few image-interop and font-measurement members, plus the optional GDI+ renderer, still depend on System.Drawing.Common and throw PlatformNotSupportedException on Linux/macOS. Removing the remaining Windows dependency is planned for a later release.

It provides an Aspose.PDF-compatible API surface for common PDF scenarios — a large part of existing Aspose.PDF for .NET code can compile and run unchanged. See What's included and What's not included below for the per-feature breakdown.

Requirements

• .NET 8.0 or later — runs on Windows, macOS, and Linux.
• One NuGet dependency: System.Drawing.Common, used only by a few image-interop members (e.g. ImageDevice.GetBitmap, StampInfo.Image). It's Windows-only on .NET 8, so those specific members throw PlatformNotSupportedException on Linux/macOS; everything else — text, forms, parsing, encryption & signing, and page→image rendering via PngDevice/JpegDevice — is pure-managed and fully cross-platform.

Installation

Build from source:

git clone https://github.com/aspose-pdf-foss/Aspose.PDF-FOSS-for-.NET.git
cd Aspose.PDF-FOSS-for-.NET
dotnet build src/Aspose.Pdf.Foss.csproj -c Release

Then reference src/Aspose.Pdf.Foss.csproj from your project (e.g. dotnet add reference), or add the built Aspose.Pdf.Foss.dll as an assembly reference.

What's included

Document core

• Open and create PDFs; full round-trip save and incremental save
• Page manipulation: add, delete, reorder, rotate, resize, copy across documents
• Bookmarks / outlines, page labels, named destinations, document info & metadata (XMP)
• Read & write document permissions, optional content (OCG) layers
• Standard 14 font handling, TrueType / OpenType embedding & subsetting, Type 1 (PostScript) and Type 3 font parsing
• Pure-managed crypto — no System.Security.Cryptography runtime dependency

Text

• Extraction via TextAbsorber, TextFragmentAbsorber, ParagraphAbsorber, TableAbsorber
• Find / replace across pages, including cross-operator and ligature-aware spans
• Build new text: TextBuilder, TextFragment, TextSegment, TextStamp, FormattedText
• Bidi (RTL) support; AGL Unicode mapping; CMap-based decoding

Forms (AcroForm)

• Read, fill, and build form fields: text, checkbox, radio, choice (combo/list), button, signature, rich text
• Field-level flatten, JSON import / export, XFDF tags, form-data import / export
• Form, Field, FormFieldBuilder, FormEditor, FormDataConverter

Annotations

• All standard annotation types: link, text, free-text, highlight, square / circle / line / polyline, ink, stamp, popup, file attachment, screen, redaction, watermark, pre-press marks, rich media
• Read, create, modify; appearance generation; flatten

Security

• Encrypt / decrypt with RC4-40, RC4-128, AES-128, AES-256
• Read & set document permissions
• Sign with PKCS#7 / CMS detached signatures; verify signatures with embedded certificates
• DocMDP / certifying signatures; appearance stream + image preview embedding

Rendering

• Page-to-image via PngDevice, JpegDevice, BmpDevice, TiffDevice, SvgDevice, TextDevice
• Pluggable IPageRenderer (default: SoftwarePageRenderer)
• Mesh shading types 4 – 7 (free-form / lattice Gouraud + Coons / tensor-product patch)
• JBIG2 decoder, JPEG / JPEG2000 / CCITT-Fax filters
• Resolution control, color depth, transparency, soft masks (incl. /TR transfer function)

Converters

• In: PDF, HTML, XML, SVG, Markdown
• Out: PDF, PDF/A (1A, 1B, 2A, 2B, 2U, 3A, 3B, 3U, 4, 4E, 4F), HTML, Markdown, SVG, plain text, image formats above
• HtmlSaveOptions, MdLoadOptions, SvgLoadOptions, HtmlLoadOptions

Facades

• PdfFileEditor — concatenate, split, resize, extract pages, edit content
• PdfFileSecurity — encrypt / decrypt / permission management
• PdfFileSignature — sign, verify, extract signature info, signature appearance
• PdfBookmarkEditor — outline / bookmark CRUD, XML / HTML export
• PdfContentEditor — add annotations, stamps, attachments, replace text
• PdfFileStamp — header / footer / page-number stamping
• PdfFileInfo, PdfFileMend, PdfAnnotationEditor, PdfPageEditor, PdfConverter, PdfExtractor, PdfViewer, FormEditor, FormDataConverter

Tagged PDF & accessibility

• Read existing /StructTreeRoot trees; walk Aspose.Pdf.LogicalStructure element hierarchy
• Build tagged content via ITaggedContent: 37 typed structure elements (Document, Sect, Part, P, Span, H1–H6, L / LI / Lbl / LBody, Table / THead / TBody / TFoot / TR / TH / TD, Figure, Form, Annot, etc.)
• Set document language, title, alternate text, expansion text, marked content

Other

• Tables (Table, Row, Cell, TableAbsorber)
• Drawing primitives (Graph, Rectangle, Circle, Ellipse, Arc, Line, Curve)
• XMP metadata + XmpPdfAExtension schema editing
• Image extraction & placement (ImagePlacementAbsorber, XImageCollection)
• Stamps (text / image / page-number / watermark / PDF-page)

What's not included (vs. Aspose.PDF for .NET)

The following Aspose.PDF for .NET features are not in the FOSS edition. Code that depends on them will throw NotImplementedException / PlatformNotSupportedException, or — for some types — exist as a reflection-shape stub that doesn't perform the operation:

Out of scope by design

• AI / LowCode workflows — Aspose.Pdf.AI.*, Aspose.Pdf.LowCode.*. Use Aspose.PDF for .NET for OpenAI / Llama-powered summarisation, OCR-aided extraction, and the LowCode façade APIs.
• Multithreading APIs — Aspose.Pdf.Multithreading.*. The FOSS edition is single-threaded for a single Document instance.
• Format converters beyond PDF / HTML / SVG / Markdown / XML — DOCX, EPUB, MHT, XPS, PCL, LaTeX, DJVU, OFD, PostScript, Comparison, Cgm.
• Printing pipeline — PdfViewer.Print* methods throw PlatformNotSupportedException. Render to image and pass the image to your printing stack instead.

Partial — basics work, advanced doesn't

• Digital signatures — basic Sign / Verify / PKCS#1 / PKCS#7 / DocMDP / certifying works. The validation-options surface (ValidationOptions, ValidationResult, OCSP, timestamping over the network, custom remote-sign delegates) is stored-but-not-active.
• Advanced PdfFileEditor features — MakeNUp imposition, MakeBooklet from Stream with non-trivial margins, ResizeContents with custom imposition matrices — accepted as input but the FOSS save path emits the simple layout.
• XFA forms — dynamic XFA can be read, and a whole XFA packet can be assigned / copied via Form.AssignXfa (it round-trips through the AcroForm /XFA); fine-grained authoring of individual XFA dataset fields is still limited.
• 3D annotations (PDF3DAnnotation) — annotation surface exists (artwork, content stream, lighting / render mode, view array) but the 3D content stream is not embedded in the saved PDF and the page renderer does not display the 3D model.
• Tagged PDF rendering fidelity — the structure tree is built and round-trips through /StructTreeRoot, but advanced PDF/UA-2 layout-aware tagging features are partial.
• Some PDF/A conversion semantics — validation against PDF/A levels works; certain PDF/A-2 / PDF/A-3 conversion fix-ups (transparency flattening, embedded-file constraints, ICC profile downgrade) are stored-but-not-applied.
• Advanced rendering of complex colour spaces — DeviceN and Separation with N > 4 channels, ICC-based group blending in CMYK — render in approximate sRGB.

Renderer fidelity

• The FOSS rendering output is pixel-close but not byte-identical to Aspose.PDF for .NET / GDI+ output. Sub-pixel glyph positioning, image dithering, anti-aliasing weight, and widget-appearance details may differ slightly. If you compare rendered PNGs against an Aspose.PDF for .NET reference, expect small per-pixel deltas.

Platform-specific notes

• System.Drawing.Common interop — methods that return System.Drawing.Image / Bitmap are tagged [SupportedOSPlatform("windows")] and throw PlatformNotSupportedException on Linux / macOS. Use the byte-array / Stream overloads instead, or read pixel data directly from the renderer's BitmapInfo.

Quick start

Open a PDF and extract text

using Aspose.Pdf;
using Aspose.Pdf.Text;

using var doc = new Document("input.pdf");
var absorber = new TextAbsorber();
absorber.Visit(doc.Pages[1]);
Console.WriteLine(absorber.Text);

Create a new PDF

using Aspose.Pdf;
using Aspose.Pdf.Text;

using var doc = new Document();
var page = doc.Pages.Add();

var fragment = new TextFragment("Hello, World!");
fragment.TextState.FontSize = 24;
page.Paragraphs.Add(fragment);

doc.Save("output.pdf");

Find and replace text

using Aspose.Pdf;
using Aspose.Pdf.Text;

using var doc = new Document("input.pdf");
var absorber = new TextFragmentAbsorber("old text");
absorber.Visit(doc);

foreach (var fragment in absorber.TextFragments)
    fragment.Text = "new text";

doc.Save("output.pdf");

Fill form fields

using Aspose.Pdf;

using var doc = new Document("form.pdf");
foreach (var field in doc.Form.Fields)
    Console.WriteLine($"{field.FullName} = {field.Value}");

Encrypt a PDF

using Aspose.Pdf;

using var doc = new Document("input.pdf");
doc.Encrypt("user-pw", "owner-pw", Permissions.PrintDocument, CryptoAlgorithm.AESx256);
doc.Save("encrypted.pdf");

Convert HTML to PDF

using Aspose.Pdf;

using var doc = new Document("input.html", new HtmlLoadOptions());
doc.Save("output.pdf");

Render a page to PNG

using Aspose.Pdf;
using Aspose.Pdf.Devices;

using var doc = new Document("input.pdf");
var device = new PngDevice(new Resolution(300));
using var stream = File.Create("page1.png");
device.Process(doc.Pages[1], stream);

Sign a PDF

using Aspose.Pdf;
using Aspose.Pdf.Facades;
using Aspose.Pdf.Forms;

using var signer = new PdfFileSignature();
signer.BindPdf("input.pdf");
var sig = new PKCS7("my.pfx", "pfx-password");
signer.Sign(1, "reason", "contact", "location", true,
    new System.Drawing.Rectangle(100, 100, 200, 60), sig);
signer.Save("signed.pdf");

Documentation

API overview

Running tests

dotnet test tests/Aspose.Pdf.Foss.Tests.csproj

Related projects

Aspose.PDF for .NET is the product from Aspose Pty Ltd that offers a broader feature set — including AI / LowCode workflows, XFA authoring, advanced digital-signature validation, multithreading, and conversion to many non-PDF formats (DOCX, EPUB, XPS, PCL, PostScript, ...). The FOSS edition shares the same public API shape, so code can usually migrate between the two with minimal changes.

License

MIT — see LICENSE.