Facilitating data type detection using existing code

A computer-implemented method for facilitating data type detection, the method comprising:

• executing a candidate type-detection function using positive example values and negative example values associated with a data type;
• for each execution of the candidate type-detection function, capturing an execution trace that represents states of the executed candidate type-detection function;
• identifying a set of features within the execution traces;
• utilizing the set of features to generate a logical explanation for the candidate type-detection function, the logical explanation representing a distinction in execution of the positive example values from the negative example values;
• and
• using the logical explanation for the candidate type-detection function to rank the candidate type-detection function among other type-detection functions based on extents to which corresponding logical explanations distinguish execution of the positive example values from the negative example values.

One or more computer storage media having computer-executable instructions embodied thereon that, when executed by one or more processors, cause the one or more processors to perform a method for facilitating data transformations, the method comprising:

• receiving a target data type and a set of positive example values;
• searching an existing repository having functions to identify a set of candidate functions related to the target data type;
• using at least a portion of the set of positive example values to generate a set of negative example values;
• analyzing the candidate functions using the positive example values and the negative example values to identify one or more type-detection functions to provide as a function suggestion for detecting the target data type;
• and
• providing the one or more type-detection functions as a function suggestion for detecting the target data type.