Processing scanned documentsΒΆ

In this use case we will focus on how and why use the Connector for processing scanned documents.

Getting scanned documents into M-Files and classifying the document can be very easy, and mostly is. M-Files provide powerful capabilities for it out of the box. Specialist scanning software (such as Ancora, chronoscan, ezescan, ABBYY and others) all include methods of processing scanned documents into M-Files.

What role does the Connector play then? In our experience it comes into play, usually in collaboration with the M-Files own tools, or any of the tools, when the processing in the backend requires additional intervention. The following scenarios could apply:

  • complexity of sourcing and validating the master data to file against;

  • quality and complexity of the scanned and acquired data;

  • merging the scanned data with external data;

  • advanced validating of the scanned data against data from different sources;

  • creating multiple and complex relationships between the objects in the upload process and further processing of the data

  • triggering workflows based on the characteristics and quality of the scanned data, or the availability of master data at the point of scanning.

Using the power of SQL, combined with the Connector allows for more flexible and dynamic development of the backend process to tie the scanned document to the correct objects and metadata without resorting the API development.

The mechanism is quit simple:

Process the source file through the scanning software and push all the scanned data into a SQL table. Ensure the name of the output file is unique. Capture the name of the file into the SQL table. Allow the file to be imported into M-Files using standard External File Connector functionality.

Use a state change context menu action to trigger the MFSQL Connector to:

  • get the new file object from M-Files

  • use SQL to perform all the complex processing

  • on completion update the result into M-Files which may include several new objects

The key is that the SQL process could include a whole host of operations including getting or pushing data to other applications; validating the data and matching; making the data decisions for all the options and getting or pushing related objects to M-Files.