Artem Y Livshits

US Patent:

7076486, Jul 11, 2006

Filed:

Jan 14, 2005

Appl. No.:

11/034965

Inventors:

Artem Y. Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 17/30
G06F 12/00
G06K 9/00

US Classification:

707 6, 707 3, 707200, 382130

Abstract:

Methods and data structures are disclosed for carrying out identifying differences between large files comprising many lines (or other units of comparison such as rows, words, paragraphs, sentences, etc. ). The disclosed methods and data structures facilitate and carry out a streamlined, yet thorough comparison of two files to identify differences between them. The streamlining is achieved by pre-processing the files prior to submitting them to any known longest common subsequence (LCS) search engine. The output of the LCS generator is post-processed to compensate for changes to the sequences introduced by the pre-processing stage.

US Patent:

7290251, Oct 30, 2007

Filed:

May 16, 2003

Appl. No.:

10/440339

Inventors:

Artem Y. Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 9/44
G06F 15/00

US Classification:

717143, 715511, 715512

Abstract:

A method and system provides an improved representation of merge conflicts in a three-way merge operation of three sequences: Original, Version, and Version. A merge program first compares the three sequences to identify a list of difference chunks of elements between the Original and Version, and a list of difference chunks between the Original and Version. From the two sets of difference chunks, the merge program creates an annotated list of chunks. Each chunk in the annotated list is identified by a conflict flag that indicates whether the chunk is a part of a conflict and if it is, what sequence(s) being merged it belongs to. A merged document is then constructed based on the annotated list of chunks, with the chunks identified in the merged document with conflict markers for viewing by a user.

US Patent:

7299450, Nov 20, 2007

Filed:

Jun 17, 2003

Appl. No.:

10/463152

Inventors:

Artem Y. Livshits - Bellevue WA, US
Christopher J. Antos - Redmond WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 9/44

US Classification:

717121, 717122

Abstract:

Carrying out maintaining a software configuration management system are disclosed. In particular, a way to reverse prior revisions without removing their history in the server are described. When a client submits an undo command to reverse the effects of one or more revisions within a stated revision range of a file, the revision history is updated by submitting a new revision that effectively removes the changes embodied within the range of undone revisions. Implementing undoing prior revisions preserves the revision history for a file, in case there is a need to access the undone revisions at a later time.

US Patent:

7502792, Mar 10, 2009

Filed:

Apr 26, 2006

Appl. No.:

11/411549

Inventors:

Artem Yevgenyevich Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 7/00

US Classification:

707 8, 707203

Abstract:

The present invention extends to methods, systems, computer program products, and data structures for managing database snapshot storage. Spatial properties database transactions are used to manage database snapshot storage allocation and reclamation. When a read transaction and subsequent write transactions refer to disjoint portions of a database, the read transaction can continue to read portions of data directly from the database. When a read transaction and subsequent write transactions refer to intersecting portions of a database, the read transaction can be transitioned to “snapshot” mode. The write transaction can publish a database snapshot can configure the read transaction to read contested portions of data from the published snapshot.

US Patent:

7653643, Jan 26, 2010

Filed:

Mar 24, 2005

Appl. No.:

11/088378

Inventors:

Artem Y. Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 7/00
G06F 12/00

US Classification:

707101, 707 4, 707200

Abstract:

A configuration management system that uses a data compression method to compress entries in a data set. An entry is selected as a prefix value and prefix compression of the data set is performed. The entry to serve as the prefix value is quickly selected using an iterative approach. In each iteration, subgroups of entries are formed from groups formed in prior iterations based on the values of characters at successive positions in the entries. The approach is readily implemented using data structures represented as lists.

US Patent:

8306956, Nov 6, 2012

Filed:

Jul 31, 2007

Appl. No.:

11/888272

Inventors:

Artem Y. Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 7/00
G06F 17/00
H03M 5/00
H03M 7/00

US Classification:

707693, 341 55, 341 79

Abstract:

A configuration management system that uses a data compression method to compress entries in a data set. An entry is selected as a prefix value and prefix compression of the data set is performed. The entry to serve as the prefix value is quickly selected using an iterative approach. In each iteration, subgroups of entries are formed from groups formed in prior iterations based on the values of characters at successive positions in the entries. The approach is readily implemented using data structures represented as lists.

US Patent:

20060123033, Jun 8, 2006

Filed:

Jan 6, 2006

Appl. No.:

11/275473

Inventors:

Artem Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 7/00

US Classification:

707101000

Abstract:

A method and system provides an improved representation of merge conflicts in a three-way merge operation of three sequences: Original, Version1, and Version2. A merge program first compares the three sequences to identify a list of difference chunks of elements between the Original and Version1, and a list of difference chunks between the Original and Version2. From the two sets of difference chunks, the merge program creates an annotated list of chunks. Each chunk in the annotated list is identified by a conflict flag that indicates whether the chunk is a part of a conflict and if it is, what sequence(s) being merged it belongs to. A merged document is then constructed based on the annotated list of chunks, with the chunks identified in the merged document with conflict markers for viewing by a user.

US Patent:

6904430, Jun 7, 2005

Filed:

Apr 26, 2002

Appl. No.:

10/133520

Inventors:

Artem Y. Livshits - Bellevue WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F017/30

US Classification:

707 6, 707102, 382130

Abstract:

Methods and data structures are disclosed for carrying out identifying differences between large files comprising many lines (or other units of comparison such as rows, words, paragraphs, sentences, etc. ). The disclosed methods and data structures facilitate and carry out a streamlined, yet thorough comparison of two files to identify differences between them. The streamlining is achieved by pre-processing the files prior to submitting them to any known longest common subsequence (LCS) search engine. The output of the LCS generator is post-processed to compensate for changes to the sequences introduced by the pre-processing stage.