PDF Redaction in C#: Remove Sensitive Data and Sanitize Documents with IronPDF

using IronPdf;

// Load the source document
PdfDocument pdf = PdfDocument.FromFile("employee-records.pdf");

// Redact an employee name from the entire document
pdf.RedactTextOnAllPages("John Smith");

// Redact a Social Security Number
pdf.RedactTextOnAllPages("123-45-6789");

// Redact an internal employee ID
pdf.RedactTextOnAllPages("EMP-2024-0042");

// Save the cleaned document
pdf.SaveAs("employee-records-redacted.pdf");

using IronPdf;
using System.Collections.Generic;

// Load the document once
PdfDocument pdf = PdfDocument.FromFile("quarterly-report.pdf");

// Define all terms that need redaction
List<string> sensitiveTerms = new List<string>
{
    "Project Titan",
    "Sarah Johnson",
    "Budget: $4.2M",
    "Q3-INTERNAL-2024",
    "sarah.johnson@company.com"
};

// Redact each term
foreach (string term in sensitiveTerms)
{
    pdf.RedactTextOnAllPages(term);
}

// Save the result
pdf.SaveAs("quarterly-report-public.pdf");

using IronPdf;

// Load the document
PdfDocument pdf = PdfDocument.FromFile("contract-bundle.pdf");

// Redact text only on page 1 (index 0)
pdf.RedactTextOnPage(0, "Client Name: Acme Corporation");

// Redact text on pages 3, 5, and 7 (indices 2, 4, 6)
int[] financialPages = { 2, 4, 6 };
pdf.RedactTextOnPages(financialPages, "Payment Terms: Net 30");

// Other pages remain untouched except for the specific redactions applied

pdf.SaveAs("contract-bundle-redacted.pdf");

using IronPdf;

// Process a large document efficiently
PdfDocument pdf = PdfDocument.FromFile("annual-report-500-pages.pdf");

// We know from document structure that:
// - Executive summary with names is on pages 1-3
// - Financial data is on pages 45-60
// - Appendix with employee info is on pages 480-495

// Redact executive names from summary section
for (int i = 0; i <= 2; i++)
{
    pdf.RedactTextOnPage(i, "CEO: Robert Williams");
    pdf.RedactTextOnPage(i, "CFO: Maria Garcia");
}

// Redact specific financial figures from the financial section
int[] financialSection = { 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59 };
pdf.RedactTextOnPages(financialSection, "Net Revenue: $847M");

// Redact employee identifiers from appendix
for (int i = 479; i <= 494; i++)
{
    pdf.RedactTextOnPage(i, "Employee ID:");
}

pdf.SaveAs("annual-report-public-release.pdf");

using IronPdf;

// Load the document
PdfDocument pdf = PdfDocument.FromFile("legal-brief.pdf");

// Case-sensitive redaction: only matches exact case
// "CLASSIFIED" will be redacted but "classified" or "Classified" will not
pdf.RedactTextOnAllPages(
    "CLASSIFIED",
    CaseSensitive: true,
    OnlyMatchWholeWords: true,
    DrawRectangles: true,
    ReplacementText: "[REDACTED]"
);

// Case-insensitive redaction: matches regardless of case
// Will redact "Secret", "SECRET", "secret", etc.
pdf.RedactTextOnAllPages(
    "secret",
    CaseSensitive: false,
    OnlyMatchWholeWords: true,
    DrawRectangles: true,
    ReplacementText: "*****"
);

// Whole word disabled: matches partial strings too
// Will redact "password", "passwords", "mypassword123", etc.
pdf.RedactTextOnAllPages(
    "password",
    CaseSensitive: false,
    OnlyMatchWholeWords: false,
    DrawRectangles: true,
    ReplacementText: "XXXXX"
);

// No visual rectangle: text is removed but no black box appears
// Useful when you want seamless removal without obvious redaction marks
pdf.RedactTextOnAllPages(
    "internal-reference-code",
    CaseSensitive: true,
    OnlyMatchWholeWords: true,
    DrawRectangles: false,
    ReplacementText: ""
);

pdf.SaveAs("legal-brief-redacted.pdf");

using IronPdf;
using IronSoftware.Drawing;

// Load a document with signature blocks and photos
PdfDocument pdf = PdfDocument.FromFile("signed-agreement.pdf");

// Define a region for a signature block
// Located 100 points from left, 650 points from bottom
// Width of 200 points, height of 50 points
RectangleF signatureRegion = new RectangleF(100, 650, 200, 50);

// Redact the signature region on all pages
pdf.RedactRegionsOnAllPages(signatureRegion);

// Define a region for a photo ID in the upper right
RectangleF photoRegion = new RectangleF(450, 700, 100, 120);
pdf.RedactRegionsOnAllPages(photoRegion);

// Save the document with regions redacted
pdf.SaveAs("signed-agreement-redacted.pdf");

using IronPdf;
using IronSoftware.Drawing;

PdfDocument pdf = PdfDocument.FromFile("form-document.pdf");

// Get dimensions of the first page
var pageInfo = pdf.Pages[0];

// Calculate regions relative to page dimensions
// Redact the bottom quarter of the page where signatures appear
float signatureAreaHeight = (float)(pageInfo.Height / 4);
RectangleF bottomQuarter = new RectangleF(
    0,                              // Start at left edge
    0,                              // Start at bottom
    (float)pageInfo.Width,          // Full page width
    signatureAreaHeight             // Quarter of page height
);

pdf.RedactRegionsOnAllPages(bottomQuarter);

// Redact a header area at the top containing letterhead with address
float headerHeight = 100;
RectangleF headerArea = new RectangleF(
    0,
    (float)(pageInfo.Height - headerHeight), // Position from bottom
    (float)pageInfo.Width,
    headerHeight
);

pdf.RedactRegionsOnAllPages(headerArea);

pdf.SaveAs("form-document-redacted.pdf");

using IronPdf;
using System;

// Load a document containing sensitive metadata
PdfDocument pdf = PdfDocument.FromFile("internal-report.pdf");

// Access current metadata properties
string author = pdf.MetaData.Author;
string title = pdf.MetaData.Title;
string subject = pdf.MetaData.Subject;
string keywords = pdf.MetaData.Keywords;
string creator = pdf.MetaData.Creator;
string producer = pdf.MetaData.Producer;
DateTime? creationDate = pdf.MetaData.CreationDate;
DateTime? modifiedDate = pdf.MetaData.ModifiedDate;

// Get all metadata keys including custom properties
var allKeys = pdf.MetaData.Keys();

using IronPdf;
using System;

PdfDocument pdf = PdfDocument.FromFile("confidential-memo.pdf");

// Replace identifying metadata with generic values
pdf.MetaData.Author = "Organization Name";
pdf.MetaData.Creator = "Document System";
pdf.MetaData.Producer = "";
pdf.MetaData.Title = "Public Document";
pdf.MetaData.Subject = "";
pdf.MetaData.Keywords = "";

// Normalize dates to remove timing information
pdf.MetaData.CreationDate = DateTime.Now;
pdf.MetaData.ModifiedDate = DateTime.Now;

// Remove specific custom metadata keys
pdf.MetaData.RemoveMetaDataKey("OriginalFilename");
pdf.MetaData.RemoveMetaDataKey("LastSavedBy");
pdf.MetaData.RemoveMetaDataKey("Company");
pdf.MetaData.RemoveMetaDataKey("Manager");

// Remove custom properties added by applications
try
{
    pdf.MetaData.CustomProperties.Remove("SourcePath");
}
catch { }

pdf.SaveAs("confidential-memo-cleaned.pdf");

using IronPdf;
using System;
using System.Collections.Generic;

public class MetadataCleaner
{
    private readonly string _defaultAuthor;
    private readonly string _defaultCreator;

    public MetadataCleaner(string organizationName)
    {
        _defaultAuthor = organizationName;
        _defaultCreator = $"{organizationName} Document System";
    }

    public void CleanMetadata(PdfDocument pdf)
    {
        // Replace standard metadata fields
        pdf.MetaData.Author = _defaultAuthor;
        pdf.MetaData.Creator = _defaultCreator;
        pdf.MetaData.Producer = "";
        pdf.MetaData.Subject = "";
        pdf.MetaData.Keywords = "";

        // Normalize timestamps
        DateTime now = DateTime.Now;
        pdf.MetaData.CreationDate = now;
        pdf.MetaData.ModifiedDate = now;

        // Get all keys and remove potentially sensitive ones
        List<string> keysToRemove = new List<string>();
        foreach (string key in pdf.MetaData.Keys())
        {
            // Keep only essential keys
            if (!IsEssentialKey(key))
            {
                keysToRemove.Add(key);
            }
        }

        foreach (string key in keysToRemove)
        {
            pdf.MetaData.RemoveMetaDataKey(key);
        }
    }

    private bool IsEssentialKey(string key)
    {
        // Keep only the basic display properties
        string[] essentialKeys = { "Title", "Author", "CreationDate", "ModifiedDate" };
        foreach (string essential in essentialKeys)
        {
            if (key.Equals(essential, StringComparison.OrdinalIgnoreCase))
            {
                return true;
            }
        }
        return false;
    }
}

// Usage
class Program
{
    static void Main()
    {
        MetadataCleaner cleaner = new MetadataCleaner("Acme Corporation");

        PdfDocument pdf = PdfDocument.FromFile("report.pdf");
        cleaner.CleanMetadata(pdf);
        pdf.SaveAs("report-clean.pdf");
    }
}

using IronPdf;

// Load a PDF that may contain active content
PdfDocument pdf = PdfDocument.FromFile("received-document.pdf");

// Sanitize using SVG conversion
// Faster processing, results in searchable text, slight layout variations possible
PdfDocument sanitizedSvg = Cleaner.SanitizeWithSvg(pdf);
sanitizedSvg.SaveAs("sanitized-svg.pdf");

// Sanitize using Bitmap conversion
// Slower processing, text becomes image (not searchable), exact visual reproduction
PdfDocument sanitizedBitmap = Cleaner.SanitizeWithBitmap(pdf);
sanitizedBitmap.SaveAs("sanitized-bitmap.pdf");

using IronPdf;

// Load the potentially unsafe document
PdfDocument pdf = PdfDocument.FromFile("untrusted-source.pdf");

// Configure rendering options for sanitization
var renderOptions = new ChromePdfRenderOptions
{
    MarginTop = 10,
    MarginBottom = 10,
    MarginLeft = 10,
    MarginRight = 10
};

// Sanitize with custom options
PdfDocument sanitized = Cleaner.SanitizeWithSvg(pdf, renderOptions);
sanitized.SaveAs("untrusted-source-safe.pdf");

using IronPdf;
using System;

public class SecureDocumentProcessor
{
    public PdfDocument ProcessUntrustedDocument(string inputPath)
    {
        // Load the document
        PdfDocument original = PdfDocument.FromFile(inputPath);

        // Step 1: Sanitize to remove active content
        PdfDocument sanitized = Cleaner.SanitizeWithSvg(original);

        // Step 2: Clean metadata
        sanitized.MetaData.Author = "Processed Document";
        sanitized.MetaData.Creator = "Secure Processor";
        sanitized.MetaData.Producer = "";
        sanitized.MetaData.CreationDate = DateTime.Now;
        sanitized.MetaData.ModifiedDate = DateTime.Now;

        // Remove all custom metadata
        foreach (string key in sanitized.MetaData.Keys())
        {
            if (key != "Title" && key != "Author" && key != "CreationDate" && key != "ModifiedDate")
            {
                sanitized.MetaData.RemoveMetaDataKey(key);
            }
        }

        return sanitized;
    }
}

// Usage
class Program
{
    static void Main()
    {
        SecureDocumentProcessor processor = new SecureDocumentProcessor();
        PdfDocument safe = processor.ProcessUntrustedDocument("email-attachment.pdf");
        safe.SaveAs("email-attachment-safe.pdf");
    }
}

using IronPdf;
using IronSoftware.Drawing;
using System;
using System.Collections.Generic;
using System.IO;
using System.Text.RegularExpressions;

public class DocumentProcessingResult
{
    public string OriginalFile { get; set; }
    public string OutputFile { get; set; }
    public bool WasSanitized { get; set; }
    public int TextRedactionsApplied { get; set; }
    public int RegionRedactionsApplied { get; set; }
    public bool MetadataCleaned { get; set; }
    public List<string> SensitiveDataTypesFound { get; set; } = new List<string>();
    public DateTime ProcessedAt { get; set; }
    public bool Success { get; set; }
    public string ErrorMessage { get; set; }
}

public class ComprehensiveDocumentProcessor
{
    // Sensitive data patterns
    private readonly Dictionary<string, string> _sensitivePatterns = new Dictionary<string, string>
    {
        { "SSN", @"\b\d{3}-\d{2}-\d{4}\b" },
        { "Credit Card", @"\b(?:\d{4}[-\s]?){3}\d{1,4}\b" },
        { "Email", @"\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b" },
        { "Phone", @"\b(?:\(\d{3}\)\s?|\d{3}[-.])\d{3}[-.]?\d{4}\b" }
    };

    // Standard regions to redact (signature areas, photo locations)
    private readonly List<RectangleF> _standardRedactionRegions = new List<RectangleF>
    {
        new RectangleF(72, 72, 200, 50),    // Bottom left signature
        new RectangleF(350, 72, 200, 50)    // Bottom right signature
    };

    private readonly string _organizationName;

    public ComprehensiveDocumentProcessor(string organizationName)
    {
        _organizationName = organizationName;
    }

    public DocumentProcessingResult ProcessDocument(
        string inputPath,
        string outputPath,
        bool sanitize = true,
        bool redactPatterns = true,
        bool redactRegions = true,
        bool cleanMetadata = true,
        List<string> additionalTermsToRedact = null)
    {
        var result = new DocumentProcessingResult
        {
            OriginalFile = inputPath,
            OutputFile = outputPath,
            ProcessedAt = DateTime.Now
        };

        try
        {
            // Load the document
            PdfDocument pdf = PdfDocument.FromFile(inputPath);

            // Step 1: Security scan
            CleanerScanResult scanResult = Cleaner.ScanPdf(pdf);

            if (scanResult.IsDetected && scanResult.Risks.Count > 10)
            {
                throw new SecurityException("Document contains too many security risks to process");
            }

            // Step 2: Sanitization (if needed or requested)
            if (sanitize || scanResult.IsDetected)
            {
                pdf = Cleaner.SanitizeWithSvg(pdf);
                result.WasSanitized = true;
            }

            // Step 3: Pattern-based text redaction
            if (redactPatterns)
            {
                string fullText = pdf.ExtractAllText();
                HashSet<string> valuesToRedact = new HashSet<string>();

                foreach (var pattern in _sensitivePatterns)
                {
                    Regex regex = new Regex(pattern.Value, RegexOptions.IgnoreCase);
                    MatchCollection matches = regex.Matches(fullText);

                    if (matches.Count > 0)
                    {
                        result.SensitiveDataTypesFound.Add($"{pattern.Key} ({matches.Count})");
                        foreach (Match match in matches)
                        {
                            valuesToRedact.Add(match.Value);
                        }
                    }
                }

                // Apply redactions
                foreach (string value in valuesToRedact)
                {
                    pdf.RedactTextOnAllPages(value);
                    result.TextRedactionsApplied++;
                }
            }

            // Step 4: Additional specific terms
            if (additionalTermsToRedact != null)
            {
                foreach (string term in additionalTermsToRedact)
                {
                    pdf.RedactTextOnAllPages(term);
                    result.TextRedactionsApplied++;
                }
            }

            // Step 5: Region-based redaction
            if (redactRegions)
            {
                foreach (RectangleF region in _standardRedactionRegions)
                {
                    pdf.RedactRegionsOnAllPages(region);
                    result.RegionRedactionsApplied++;
                }
            }

            // Step 6: Metadata cleaning
            if (cleanMetadata)
            {
                pdf.MetaData.Author = _organizationName;
                pdf.MetaData.Creator = $"{_organizationName} Document Processor";
                pdf.MetaData.Producer = "";
                pdf.MetaData.Subject = "";
                pdf.MetaData.Keywords = "";
                pdf.MetaData.CreationDate = DateTime.Now;
                pdf.MetaData.ModifiedDate = DateTime.Now;
                result.MetadataCleaned = true;
            }

            // Step 7: Save the processed document
            pdf.SaveAs(outputPath);
            result.Success = true;
        }
        catch (Exception ex)
        {
            result.Success = false;
            result.ErrorMessage = ex.Message;
        }

        return result;
    }
}

// Usage example
class Program
{
    static void Main()
    {
        var processor = new ComprehensiveDocumentProcessor("Acme Corporation");

        // Process a single document with all protections
        var result = processor.ProcessDocument(
            inputPath: "customer-application.pdf",
            outputPath: "customer-application-redacted.pdf",
            sanitize: true,
            redactPatterns: true,
            redactRegions: true,
            cleanMetadata: true,
            additionalTermsToRedact: new List<string> { "Project Alpha", "Internal Use Only" }
        );

        // Batch process multiple documents
        string[] inputFiles = Directory.GetFiles("incoming", "*.pdf");
        foreach (string file in inputFiles)
        {
            string outputFile = Path.Combine("processed", Path.GetFileName(file));
            processor.ProcessDocument(file, outputFile);
        }
    }
}