OCR Output is Corrupted...

[ajcullen]

Hey guys,

I'm having trouble implementing the OCR code using the OCRDemoCSharp as a reference guide for implementing it into our application. Everything appears to run and i don't get any errors, however, the output file appears to be corrupted and is only 4KB in size...

I've tried running the OCRDemoCSharp and using that and it also results in a corrupted file or file with no pages.

The project is a .net core 8.0 desktop application. We have already purchased a full license to the software.

Please help...

Thanks
AJ

[Daniel - PDF-XChange]

Hello, ajcullen

We have been discussing this issue via email, but I wanted to post here publicly the results of that investigation as well, incase anyone else is having the same issue:

We have managed to locate the root of the issue(s). Both are on our end, and will need to be resolved in the next release (10.8.3). It seems the issue relates to a combination of errors in the language files in place for the SDK (they are also somewhat dated - so new copies will be coming), which could directly cause the OCR process to fail on normal recognition - and a secondary bug relating to font size detection which is likely why so little/no text was recognized in your example.

I am sorry I cannot offer an on the spot solution to be implemented from your end, but you can look forward to a fix in the next release, which should hopefully only be a few weeks away at most.

We hope to have the next release publicly available in late February or early march.

Kind regards,