Is there a way to convert HL7 MDM encoded base64 PDF data to text?

Here’s some code that translates from PDF to both RTF and TXT.

https://pastebin.com/vJ4czmpq

You can use “pdftotext” to convert to text, but there may be additional formatting tweaks required to get it to look nice. This code does some stuff like replacing windows extended characters with ascii equivalents for some characters (non-blocking space and hypen).

It is quite possible your receiving system could choke on the non-encoded data as HL/7 separators could now be exposed.

To answer your question and for anyone else who happens to read this, I do not think you should ever want to do this because the amount of support effort and risk of corrupting data is far outweighed by having this handled on the sending side, which should be an easy task as it is probably just some reconfiguration at most. You really should have your sender send non-encoded and HL7 escaped data if that is how you want to forward it on.

I feel like by doing it within Cloverleaf code, even when you get it working it is just a bandaid compared to having your sender do it.

With that said, to get your incoming data out of base64 is pretty simple in TCL… simply add this command (to any scripting location you want, probably even within an Xlate copy?):

set unencodedData [decode $base64encodedinput ]

That is the easy part. If you were to dump $unencodedData into a file, say “output.pdf” you would see your document.

However, as Jim says above, continuing that data as HL7 is another story. You will now have a document that needs to be prepared/escaped for HL7 before it can be forwarded (escape ampersands, backslashes, handle spacing, etc). Pretty much all of David’s code above is to handle this piece.

Good luck!

Personally I would have the receiving system become compliant with the standard. That should not be too difficult for them to do and it will make them more marketable as this is a common way to transfer data.

Since you indicate this is a prospective system to be acquired, now is the time to negotiate that change in order to be acceptable to your organization I would think. This is when you have the most leverage to get the receiving system doing what it should.

The sending system is following the standard and thus I don’t think should be enjoined to do anything.

But you have options.

Turns out I am trying to do something similar – the PDF is encoded and I want to extract it into a file structure.

This is the code I am using, and it does create a file, but the PDF reader I am using (Adobe) says the file is “damaged”.

Up to this point, I have isolated the field into a variable then decode the variable and write to file:

set unencodedData [decode $OBX5_4 ]
echo unencodedData:$unencodedData

set outfilename $OBX3_4
set fileId [open $outfilename “w”]
puts $fileId $unencodedData
close $fileId

So…what am I doing wrong?  I have taken the value in $OBX5_4 and pasted it into a BASE64 to PDF converter (https://base64.guru/converter/decode/pdf) and it created the document just fine.

Thanks,

Scott

Perhaps you need to do a fconfigure on the file using the -translation switch with binary as its argument.

Adding this line did the trick.  Thank you for the help!

set outfilename $OBX3_4
set fileId [open $outfilename “w”]
fconfigure $fileId -encoding binary -translation binary
puts $fileId $unencodedData
close $fileId