Automated software testing using natural language-based form completion

A method for automated software testing, the method comprising:

• executing software under test;
• determining that a user interface of the software includes a textual input field;
• identifying a label of the textual input field;
• identifying additional information regarding the textual input field, the additional information comprising hover-over text that is displayed when a cursor is over the textual input field;
• obtaining a first input text from a textual input database based on the label;
• inputting the first input text into the textual input field;
• determining that a first response of the software to the first input text includes a first error message related to the textual input field, the first error message comprising a constraint for the textual input field;
• inputting into a natural language processing system the label, information about the first error message, and the additional information as a query that includes an instruction to generate a second sample input for the textual input field subject to the constraint;
• receiving, from the natural language processing system in response to the query, a second input text comprising the second sample input;
• inputting the second input text into the textual input field;
• and
• recording a second response of the software to the second input text.

A system having a memory, computer readable instructions, and a processing system for executing the computer readable instructions, the computer readable instructions control the processing system to perform operations comprising:

• executing software under test;
• determining that a user interface of the software includes a textual input field;
• identifying a label of the textual input field;
• identifying additional information regarding the textual input field, the additional information comprising hover-over text that is displayed when a cursor is over the textual input field;
• inputting into a natural language processing system the label and the additional information as a first query, the first query including an instruction to generate a sample input for the textual input field;
• receiving, from the natural language processing system in response to the first query, a first input text comprising the sample input;
• inputting the first input text into the textual input field;
• determining that a first response of the software includes a first error message related to the textual input field, the first error message comprising a constraint for the textual input field;
• inputting into the natural language processing system the label and information about the first error message as a second query, the second query including an instruction to generate a second sample input for the textual input field subject to the constraint;
• receiving, from the natural language processing system in response to the second query, a second input text comprising the second sample input;
• inputting the second input text into the textual input field;
• and
• recording a second response of the software to the second input text.