IsPdfWithTextualContentFromStream Method of Engine Object
This method finds out if any of the pages of PDF file added from stream contain text layer. If you need to convert multiple PDF files to searchable PDF, you can check for text layer, and simply copy the whole file if it contains text layer. However, note that the text layer can contain additional or irrelevant text, in which case it's still necessary to perform recognition if you need completely adequate text rendering of the document.
Syntax
HRESULT IsPdfWithTextualContentFromStream( IReadStream* Stream, IImagePasswordCallback* Callback, VARIANT_BOOL* Result );
Parameters
- Stream
- [in] Refers to the interface of the user-implemented object of the type IReadStream which contains the PDF file in which to detect the text layer.
- Callback
- [in] This variable refers to the interface of the user-implemented object of the type ImagePasswordCallback which is used to handle possible password requests for accessing PDF files. This parameter is optional and may be 0 in which case password-protected files cannot be checked.
- Result
- [out, retval] This parameter returns TRUE if at least one of the pages of the file has text layer, and FALSE if none of the pages contain text layer.
Return values
This method has no specific return values. It returns the standard return values of ABBYY FineReader Engine functions.
See also
03.07.2024 8:50:10