Hello!
Thank you for addressing us. I have investigated the issue. It should be fixed in Aspose.Pdf. I can explain what happens.
The characters that are present in the document but eaten in the PDF are long spaces. I created a document from scratch and inserted some test lines here. See the attachment. When I convert it to PDF everything goes fine. But in the XML of the original document there are the same characters, namely long spaces. Plain text editors like Notepad don’t show this character correctly. You can open the intermediate XML files in any binary editor. For instance I use MS Visual Studio binary editor. Here you could see that in both XML files long spaces are represented by code sequence “E2 80 83”. Just search for this sequence.
What I conclude is that Aspose.Words gives equal output on the segment level but Aspose.Pdf renders long space in some conditions and doesn’t render in other. Please initiate some more investigation in Aspose.Pdf.
As a workaround you can replace these long spaces with one or more non-breaking space each.
Thank you,
Viktor Sazhaev
Software Engineer,
Aspose Auckland Team