Forum Discussion
Andrey_M
12 years agoContributor
Hi.
Data from images can be extracted using TC OCR feature but my own experience shows that OCR is unreliable (you may try to use different document scaling of PDF output prior to recognition).
As alternative approach you may try to use output to Generic Printer instead of XPS/PDF and save output into text file. In this case most likely you will get data dump (which looks like text garbage) but at least it can be parsed (if your are lucky) using low-level file reading. This method will defenitely fail if your application create picture images for printing.
Honestly I don't know reliable approach for printed output automated testing.
Printed outputs can be tested appropriately using manual testing only (manual tester should check not only consistency of data but common issues such as readability, font sizes, general layout, splitting tables across pages etc).
In TC this manual testing section can be implemented using manual checkpoints feature.
Data from images can be extracted using TC OCR feature but my own experience shows that OCR is unreliable (you may try to use different document scaling of PDF output prior to recognition).
As alternative approach you may try to use output to Generic Printer instead of XPS/PDF and save output into text file. In this case most likely you will get data dump (which looks like text garbage) but at least it can be parsed (if your are lucky) using low-level file reading. This method will defenitely fail if your application create picture images for printing.
Honestly I don't know reliable approach for printed output automated testing.
Printed outputs can be tested appropriately using manual testing only (manual tester should check not only consistency of data but common issues such as readability, font sizes, general layout, splitting tables across pages etc).
In TC this manual testing section can be implemented using manual checkpoints feature.