What to do when a PDF document is converted to garbled characters and symbols?
Some imported PDF documents may return garbled text when you view them in the parsing rule editor or process them with existing parsing rules. When you see unreadable gibberish symbols as shown in the screenshot below, you are likely dealing with a corrupted PDF file.
More specifically, your PDF document is probably missing important information about font character mapping. The reason for this can be that the document was produced incorrectly. Another common reason is that the character mapping information was deliberately obfuscated as a protection mechanism to prevent the reader to "copy & paste" the text data. Lastly, it is also possible that Optical Character Recognition (OCR) with low accuracy was applied to your document before uploading it to Docparser.
In either way, it is unfortunately technically not possible to simply "fix" the document and restore the original text. Luckily, there is a work-around in Docparser that will give you near-perfect results.
To fix unreadable text issues, go to the Preprocessing settings inside of your Document Parser (SETTINGS > PREPROCESSING) and set the option "Perform OCR" to "Yes - always perform OCR" as shown in the screenshot below.
Setting this option to "Yes - always perform OCR" will convert your documents to an image file and then apply Optical Character Recognition (OCR). This means that we create a completely new text document based on the visual appearance of your original file. The new file will contain an image of your original document alongside a new (invisible) text layer with a correct character encoding. Once you enable this option, all newly uploaded documents will be sent to our OCR engine and the text should show up correctly.
PS - If you open your original document in Adobe Reader (or Mac Preview) and attempt to copy and paste the same text, you will probably run into the same issues. If the text does not paste as gibberish, please send your document to our support staff and we'll get back to you with a more detailed analysis.