How to convert images to PDF in memory?

Technical Support Apryse SDK
, ,
1

Question:
I need to convert PNG, JPEG and TIFF images to PDF all in memory. No reading or writing to disk.

Answer:

The following C# code shows how to convert an image to a PDF all in memory, using this class.

https://www.pdftron.com/api/PDFTronSDK/dotnet/pdftron.PDF.Convert.html#pdftron_PDF_Convert_StreamingPDFConversion_pdftron_Filters_Filter_pdftron_PDF_ConversionOptions_

PDFNet.Initialize(yourLicenseOrDemoKey);
// At the beginning of your process
PDFNet.SetDefaultDiskCachingEnabled(false); // disables writing temporary streams to disk

string imageExt = "png";
byte[] imageData = System.IO.File.ReadAllBytes(input_path + "butterfly.png"); // for demo purpose read from disk
using (pdftron.Filters.MemoryFilter memoryFilter = new pdftron.Filters.MemoryFilter((int)imageData.Length, false)) // false = sink
{
	pdftron.Filters.FilterWriter writer = new pdftron.Filters.FilterWriter(memoryFilter); // helper filter to allow us to write to buffer
	writer.WriteBuffer(imageData);
	writer.Flush();
	memoryFilter.SetAsInputFilter(); // switch from sink to source

	using (var newDoc = new PDFDoc())
	{
		var options = new ConversionOptions();
		options.SetFileExtension(imageExt);
		DocumentConversion documentConversion = pdftron.PDF.Convert.StreamingPDFConversion(newDoc, memoryFilter, options);
		documentConversion.Convert();
		byte[] pdfData = newDoc.Save(SDFDoc.SaveOptions.e_linearized);
		// System.IO.File.WriteAllBytes(outPath, pdfData); // if you want to test/verify the output
	}
}
1 Like
3

Hi Ryan,

I used this code myself for multi page tiffs and found that it only copied over the first page of the tiff…
I then tried to adapt the code with documentConversion.ConvertNextPage(); and documentConversion.GetConversionStatus(), but documentConversion.GetConversionStatus() fires complete after a single page, which does not work…

Do you have any advice for getting multi page tiffs to work?

1 Like
4

I actually got this to work for multi-page tiff with a different approach. I wrote the file temporarily to file then used: pdftron.PDF.Convert.ToPdf(newDoc, tempFileName); to write to the pdfDoc to then get a memory stream.

Going forward it might be nice to not have to write a file to disk to use pdftron.PDF.Convert.ToPdf… And use a memory stream/byte[] instead.

Thanks

1 Like
5

Hi @Ryan,
Without specifying “png” format, how to generalize this conversion for more different types of formats ?

1 Like
6

@Ryan Why did you use FilterWriter here. Could you please explain the usage of it ?

1 Like
7

FilterReader and FilterWriter are the main means of getting data into, or out of, a Filter such as MemoryFilter.

In this case FilterWriter is being used to copy the PNG data source into the memory owned by the MemoryFilter instance.

If the above doesn’t answer your question, please elaborate on what your are trying to resolve on your end.

2 Likes
8

Thanks. One tiny question, having “using” for FilterWriter object creation, would help to remove line ‘writer.Flush()’ here, wouldn’t it ?

1 Like
9

Hello. I’ve tried this code on a sample jpg (attached) and the resulting PDF is not what I expected. Upon inspecting the PDF, the page size defaulted to the dimensions of a letter size bond paper (8.5 x 11in) and resulted into having top and bottom white margins:

image
image1326×927 85.2 KB

I’m exploring the stream conversion approach because I am not allowed to use the server’s disk therefore cannot use other methods like Convert.ToPdf(...) that only accepts file paths.

Sample JPG in drive link due to 4MB upload limit: file0001545806234.jpg - Google Drive

1 Like
10

If you do not like the output of our APIs (other than ToPDF) then you can follow the AddImageTest sample where you will have full control over page size and image size.

Though you might need an image processing library to extract any EXIF data that is important to you, especially Orientation (as you might need to rotate the image) and DPI if you care about that.

1 Like