Convert documents for RAG pipelines Leave feedback
On this page
Retrieval-augmented generation (RAG) systems need documents in a clean, structured text format for chunking and embedding. Markdown is ideal — it preserves document structure (headings, lists, tables) while being easy to parse.
flowchart LR
A["PDF / DOCX / XLSX"]
B["GroupDocs.Markdown"]
C["Markdown"]
D["Text Chunking"]
E["Vector Embeddings"]
F["LLM Query"]
A --> B --> C --> D --> E --> F
using GroupDocs.Markdown;
// Convert document to Markdown — skip images for text-only RAG
var options = new ConvertOptions
{
ImageExportStrategy = new SkipImagesStrategy(),
Flavor = MarkdownFlavor.CommonMark
};
string markdown = MarkdownConverter.ToMarkdown("knowledge-base.pdf", options);
// Split into chunks by headings
string[] chunks = markdown.Split(
new[] { "\n## ", "\n# " },
StringSplitOptions.RemoveEmptyEntries);
foreach (string chunk in chunks)
{
// Send each chunk to your embedding model
Console.WriteLine($"Chunk ({chunk.Length} chars): {chunk.Substring(0, Math.Min(80, chunk.Length))}...");
}
using GroupDocs.Markdown;
var options = new ConvertOptions
{
ImageExportStrategy = new SkipImagesStrategy()
};
string[] files = Directory.GetFiles("documents", "*.pdf");
foreach (string file in files)
{
try
{
string markdown = MarkdownConverter.ToMarkdown(file, options);
string outputPath = Path.ChangeExtension(file, ".md");
File.WriteAllText(outputPath, markdown);
Console.WriteLine($"Converted: {file}");
}
catch (GroupDocsMarkdownException ex)
{
Console.WriteLine($"Skipped {file}: {ex.Message}");
}
}
Was this page helpful?
Any additional feedback you'd like to share with us?
Please tell us how we can improve this page.
Thank you for your feedback!
We value your opinion. Your feedback will help us improve our documentation.
