Remove text from PDF document using Aspose.PDF library?

Question

I need to delete a text from a PDF document. I am using Aspose for the purpose am currently using TextFragmentAbsorber.

FYI, I cannot use any other 3rd party library.

Below is the code I am using :

private string DeleteMachineReadableCode(string inputFilePath)
    {
        var outputFilePath = Path.Combine(Path.GetTempPath(), string.Format(@"{0}.pdf", Guid.NewGuid()));

        try
        {
            // Open document
            Document pdfDocument = new Document(inputFilePath);

            // Create TextAbsorber object to find all the phrases matching the regular expression
            TextFragmentAbsorber textFragmentAbsorber = new TextFragmentAbsorber("#START#((.|\r\n)*?)#END#"); 

            // Set text search option to specify regular expression usage
            TextSearchOptions textSearchOptions = new TextSearchOptions(true);


            textFragmentAbsorber.TextSearchOptions = textSearchOptions;

            // Accept the absorber for all pages
            pdfDocument.Pages.Accept(textFragmentAbsorber);

            // Get the extracted text fragments
            TextFragmentCollection textFragmentCollection = textFragmentAbsorber.TextFragments;

            // Loop through the fragments
            foreach (TextFragment textFragment in textFragmentCollection)
            {
                // Update text and other properties
                textFragment.Text = string.Empty;

                // Set to an instance of an object.
                textFragment.TextState.Font = FontRepository.FindFont("Verdana");
                textFragment.TextState.FontSize = 1;
                textFragment.TextState.ForegroundColor = Aspose.Pdf.Color.FromRgb(System.Drawing.Color.White);
                textFragment.TextState.BackgroundColor = Aspose.Pdf.Color.FromRgb(System.Drawing.Color.White);
            }

            pdfDocument.Save(outputFilePath);

        }
        finally
        {
            if (File.Exists(inputFilePath))
                File.Delete(inputFilePath);
        }

        return outputFilePath;
    }

I am able to replace the content if the content to be deleted is on a single page. My problem is that if the text spans over multiple pages the TextFragmentAbsorber does not recognize the text with the mentioned regex pattern ("#START#((.|\r\n)*?)#END#").

Please suggest if anything can be done on the regex or the some setting in Aspose can fix my issue.

Answer 1

As shared earlier, we can not promise earlier resolution of the issue reported by you, because of architecture limitation. However, we have modified the code snippet to meet your requirements.

The idea is to find text starting from '#START#' on the one of the document pages. Then to find text ending with '#END#' on the one of subsequent pages. And also to process all text fragments that placed on the pages between those two pages (if it exists).

    private string DeleteMachineReadableCodeUpdated(string inputFilePath)
    {
    string outputFilePath = Path.Combine(Path.GetTempPath(), string.Format(@"{0}.pdf", Guid.NewGuid()));

try
{
    // Open document
    Document pdfDocument = new Document(inputFilePath);

    // Create TextAbsorber object to find all the phrases matching the regular expression
    TextFragmentAbsorber absorber = new TextFragmentAbsorber("#START#((.|\r\n)*?)#END#");

    // Set text search option to specify regular expression usage
    TextSearchOptions textSearchOptions = new TextSearchOptions(true);

    absorber.TextSearchOptions = textSearchOptions;

    // Accept the absorber for all pages
    pdfDocument.Pages.Accept(absorber);

    // Get the extracted text fragments
    TextFragmentCollection textFragmentCollection = absorber.TextFragments;

    // If pattern found on one of the pages
    if (textFragmentCollection.Count > 0)
    {
        RemoveTextFromFragmentCollection(textFragmentCollection);
    }
    else
    {
        // In case nothing was found tries to find by parts
        string startingPattern = "#START#((.|\r\n)*?)\\z";
        string endingPattern = "\\A((.|\r\n)*?)#END#";
        bool isStartingPatternFound = false;
        bool isEndingPatternFound = false;
        ArrayList fragmentsToRemove = new ArrayList();

        foreach (Page page in pdfDocument.Pages)
        {
            // If ending pattern was already found - do nothing
            if (isEndingPatternFound)
                continue;

            // If starting pattern was already found - activate textFragmentAbsorber with ending pattern
            absorber.Phrase = !isStartingPatternFound ? startingPattern : endingPattern;

            page.Accept(absorber);
            if (absorber.TextFragments.Count > 0)
            {
                // In case something is found - add it to list
                fragmentsToRemove.AddRange(absorber.TextFragments);

                if (isStartingPatternFound)
                {
                    // Both starting and ending patterns found - the document processing
                    isEndingPatternFound = true;                        
                    RemoveTextFromFragmentCollection(fragmentsToRemove);
                }
                else
                {
                    // Only starting pattern found yet - continue
                    isStartingPatternFound = true;                        
                }
            }
            else
            {
                // In case neither starting nor ending pattern are found on current page
                // If starting pattern was found previously - get all fragments from the page
                if (isStartingPatternFound)
                {
                    absorber.Phrase = String.Empty;
                    page.Accept(absorber);
                    fragmentsToRemove.AddRange(absorber.TextFragments);
                }
                // Otherwise do nothing (continue)
            }
        }
    }

    pdfDocument.Save(outputFilePath);
}
finally
{
    if (File.Exists(inputFilePath))
        File.Delete(inputFilePath);
}

return outputFilePath;
}

private void RemoveTextFromFragmentCollection(ICollection fragmentCollection)
{
// Loop through the fragments
foreach (TextFragment textFragment in fragmentCollection)
{
    textFragment.Text = string.Empty;
}
}

Note:

• This code assumed that the only one text block starting from '#START#' and ending with '#END#' is in the document. However the above code can be easly modified to process several those blocks.
• Instead of processing text on intermediate page(s) you may store page number(s) and than delete using pdfDocument.Pages.Delete(pageNumber) before the saving document. It lets to avoid 'blank' pages if them undesirable.