Get your taxes done using TurboTax

raoulwatson · ‎January 7, 2022

I think this is very easy to reproduce. Im sure you can find a turbotax generated PDF from your return last year. Just try it yourself.

I had three pdf, not scanned but generated by TT and H&R block. The easy to test if a PDF is not scanned is simply open it with Adobe and try to select any text. On a scanned document (which is basically just a picture or bitmap), this wont work. So if you can select text and copy it then its a "real" pdf.

Internally, everything in the pdf file is represented by code "stream and endstream", for example:

<</Filter/FlateDecode/Length 10>>stream
xœ+ä î | (<== actual data)
endstream

I have written many C+ code to open and read PDF files so I am a little surprised. I'm not sure what the TT programmers are actually looking for in the PDF file (I hope they don't look for generator or version specific output).

So if you can find last year PDF generated by TT, just try to browse to it to import and start a new return.

What I get is the following, on the two TT pdf files (I file an original return and also amended return), I get file too big error. On the H&R block which I originally did in 2021 (but never filed), I get unable (see screen shots of dialog box).

I have resigned to having to reinput my personal information so no big deal here. I just called TT last November to just report it. Code wise its easy so not sure what the issue is. Reading input stream from PDF file in C+ is less than one page. You can even find so many free code like here https://www.codeproject.com/Articles/7056/Code-to-extract-plain-text-from-a-PDF-file

I really think at its minimum, TT should be able to read pdf files generated by TT itself.