Microsoft Technology Licensing, LLC

Gustavo Araujo Soares

Sumit Gulwani

Mark Alistair Wilson-Thomas

Arjun Radhakrishna

Vu Minh Le

Ashish Tiwari

Peter Groenewegen

David Ellis Pugh

Piyush Arora

German David Obando Chacon

Titus Barik

Ivan Radicek

Ameya Sanjay Ketkar

Wode Ni

Daniel Morris

Andrew R Sterland

Nachiappan Nagappan

Abhishek Udupa

Gregory Bernard Miskelly

Jonathan Preston Carter

José Pablo Cambronero Sánchez

Gust Ben Anneloes Verbruggen

Shraddha Govind Barke

Jonathan Keith Simmons

Xiang Gao

Aaron Chak Hei Yim

Alan Thomas Leung

Mohammad Raza

Kiarash Rahmani

Anders Miltner

Jialu Zhang

Some embodiments generate prompts and submit them in queries to a language model trained on code to perform automated program repair. Some embodiments fix syntactic mistakes and semantic mistakes by combining multimodal prompts, iterative querying, test-case-based selection of few-shots, and program chunking. In some cases, edit distance is minimized between an initial flawed program and the automatically created improved version of that program. The initial flawed program is obtained from a programming student, or from a source code generator.

Program improvement using large language models

Edit automation functionality generalizes edits performed by a user in a document, locates similar text, and recommends or applies transforms while staying within a current workflow. Source code edits such as refactoring are automated. The functionality uses or provides anchor target lists, temporal edit patterns, edit graphs, automatable edit sequence libraries, and other data structures and computational techniques for identifying locations appropriate for particular edits, for getting transforms, for selecting optimal transforms, for leveraging transforms in an editing session or later, and for displaying transform recommendations and results. The edit automation functionality enhances automation subtool generation, discoverability, and flexibility, for refactoring, snippet insertion, quick actions in an integrated development environment, and other automatable edit sequences.

Edit automation using an anchor target list

Embodiments use a multi-modal approach to generate software programs that match a solution program description. The solution program description may include natural language, input-output examples, partial source code, desired operators, or other hints. Some embodiments use optimized prompts to a pre-trained language model to obtain initial candidate programs. Maximal program components are extracted and then recombined variously using component-based synthesis. Beam search reduces a solution program search space by discarding some candidates from a given synthesis iteration. Relevance metrics, string similarity metrics, operator frequency distributions, token rareness scores, and other optimizations may be employed. By virtue of optimizations and the multi-modal approach, a solution program may be obtained after fewer iterations than by use of a language model alone. The multi-modal approach is domain agnostic, as illustrated by examples using regular expression and cascading style sheet selector domain specific languages.

Multi-modal program inference

Edit automation using a temporal edit pattern

Methods, systems, and computer program products for deploying a static code analyzer based on program synthesis from input-output examples. A computer system uses program synthesis on a set of input-output examples of source code edits to generate a rewrite rule that defines a transformation pattern. Based on a determined static code analyzer format, the computer system generates a static code analyzer from the rewrite rule. The static code analyzer includes a find portion that indicates a matching condition for identifying a portion of source code text, and a fix portion that indicates a textual replacement to apply to the portion of source code text matched by the find portion. The computer system deploys the static code analyzer to a development environment, including configuring the static code analyzer to be executable against a source code file within the development environment.

Deploying a static code analyzer based on program synthesis from input-output examples

Embodiments automate surfacing of underutilized development tool features, thereby enhancing the discoverability of subtools, commands, shortcuts, settings, visualizers, and other tool features. After spotting an inefficiency in the user's interaction with one or more tools, the feature surfacing functionality offers the user an interaction optimization suggestion. A mapping structure correlates detected interaction patterns with objectively better interaction optimizations. Several examples of mappings are discussed. The user can accept a suggestion, have the suggested optimization applied by an enhanced tool, and thereby reduce the number of user gestures utilized to accomplish a desired result, reduce the number of tools utilized, increase security, reduce risk of error, or get to the desired result faster, for example. Interaction optimizations also help the user stay focused, by reducing or avoiding departures from the user's current primary workflow. Other aspects of tool feature surfacing functionality are also described herein.

Surfacing underutilized tool features

A synthesis procedure learns program transformations for a text document, on-the-fly during an edit session, from examples of concrete edits made during the edit session and from an unsupervised set of additional inputs. The additional inputs are derived from explicit feedback from the user and inferred feedback from the user's behavior during the edit session. A reward score, based on anti-unification and provenance analysis, is used to classify the additional inputs as either a positive input or a negative input. Outputs are generated for the positive inputs that are consistent with the existing examples and then used to synthesize a new program transformation. The program transformations are then used to generate code edit suggestions during the edit session.

Feedback-driven semi-supervised synthesis of program transformations

Assisting automation of repeated edits of code by automated generation of rules that, when applied, perform code transformations. The transformations are synthesized while observing developers make repeated code edits, and automatically perform similar modifications as those observed. This synthesized transformation defines an initial state of code to which the transformation can be applied, and defines a modification from that initial state. A rule is then generated that includes a detector mechanism that, when selected, is configured to find locations in code that have the defined corresponding initial state of the corresponding transformation. Thus, the transformation may be applied to any code to which the rule is exposed.

Rules generation using learned repetitive code edits

Automatically identifying context-specific repeated transformations (such as repeated edit tasks) that are based on observation of the developer drafting or modifying code. As the developer modifies the code, the code passes through a series of states, one after the other. The computing system observes the series of states of the code. It is based on this observation that the computing system identifies repeated transformations of the code for potentially offering to continue performing the repeated transformations for the user. This alleviates the developer from having to manually perform the remainder of the repeated transformations.

Automatic repetition of context-specific code edits

Systems and techniques for structured query language (SQL) query formatting by examples are described herein. In an example, a SQL query formatter is adapted to receive a SQL statement, such that the SQL statement is formed using SQL tokens. The SQL query formatter may be further adapted to use a SQL formatter model to format the SQL statement, which results in a formatted SQL statement. The SQL formatter model may be trained using at least one previously executed SQL statement. The SQL formatter model may include format definitions for SQL tokens. The SQL query formatter may be further adapted to output the formatted SQL statement.

SQL query formatting by examples

Growing community of inventors

Seattle, WA, United States of America

Gustavo Araujo Soares