Question

I am using tesseract OCR for business card reading.
I have a memory leak and I can’t resolve it, I don’t know how to.

In my code…

tesseract->Recognize(NULL); 
char* utf8Text = tesseract->GetUTF8Text();

GetUTF8Text() method gives memory leak.
Here is the log in memory leak instruments:

tesseract::TessBaseAPI::GetUTF8Text()
operator new[](unsigned long) libstdc++.6.dylib
operator new(unsigned long) libstdc++.6.dylib
malloc libsystem_c.dylib

After some memory leaks, app crashes.
GetUTF8Text is in baseapi.h file. I think tessearact was written by c++. I don’t know c++.
Can anyone help? Or anyone has clean tesseract?

Answer 1

According the documentation I found in baseapi.h.

/**
 * The recognized text is returned as a char* which is coded
 * as UTF8 and must be freed with the delete [] operator.
 */
char* GetUTF8Text();

So you will need to delete [] the utf8text when you are done with it.

tesseract->Recognize(NULL); 
char* utf8Text = tesseract->GetUTF8Text();
... //use utf8Text or copy if necessary
delete [] utf8text;