Major Word to PDF conversion defect

Aspose.Words Product Family
1

Create a Word document using Aspose.Words. Use the Save(“mydoc.pdf”) to create a PDF document. Information in the Word document is lost in the conversion. This is a major defect that impacts the results of the final PDF document. Notice in the attached files that the table in the Word header area is clipped short and the other table is completely lost! This seems like a major issue with conversion using Aspose. If I open the Word doc in Microsoft word and Save As… a PDF it converts just fine.

2

Hi Joseph,

Thanks for your inquiry.

I cannot reproduce any of the issues you described above using the latest version of Aspose.Words (9.3). You can dowload it from here. Can you please test using this and post your feedback if you find any issues with the current version

Thanks,

3

Here is are the documents produced using Words.9.3.0 when the document is created in memory then saved to a PDF filename using the doc.Save() command. As you can see there is still a major defect. The tables are completely missing in the PDF. However, if the in-memory word document is saved to a filename (test.doc) as an MS doc file then reloaded into memory using doc = new Document(“test.doc”) then saved as doc.Save(“test.pdf”) then the tables are all present! So there is a MAJOR difference between the document that is created in memory and the document that is read from disk.

Maybe there are some options that need to be set before saving the in-memory version to a PDF ?

I would prefer NOT to have to save the doc to disk then read it in before saving to PDF. This takes extra time and makes maintaining the code very dangerous.

4

Hi Joseph,

Thanks for this additional information.

I still cannot reproduce the issue using the document attached. All tables are rendered correctly to .pdf. From your further explanation it appears this issue is only occurring when you first save it to doc after generating it in memory. In this case could you please attach a sample project containing the code which generates the document, which should then allow me to reproduce this issue on my side.

Thanks,

5

The project is too complicated with too many dependencies to send. I will see if I can reproduce it with a small project. This could take me a few days.