code review
{{short description|Activity where one or more people check a program's code}}
{{Software development process}}
Code review (sometimes referred to as peer review) is a software quality assurance activity in which one or more people examine the source code of a computer program, either after implementation or during the development process. The persons performing the checking, excluding the author, are called "reviewers". At least one reviewer must not be the code's author.{{cite book |last1=Baum |first1=Tobias |last2=Liskin |first2=Olga |last3=Niklas |first3=Kai |last4=Schneider |first4=Kurt |title= 2016 IEEE International Conference on Software Quality, Reliability and Security (QRS)|pages=74–85 |date=2016 |doi=10.1109/QRS.2016.19|isbn=978-1-5090-4127-5 |chapter=A Faceted Classification Scheme for Change-Based Industrial Code Review Processes |s2cid=9569007 }}{{cite book | last = Kolawa | first = Adam |author2=Huizinga, Dorota | title = Automated Defect Prevention: Best Practices in Software Management | url = http://www.wiley.com/WileyCDA/WileyTitle/productCd-0470042125.html | year = 2007 | publisher = Wiley-IEEE Computer Society Press | page=260 | isbn = 978-0-470-04212-0 }}
Code review differs from related software quality assurance techniques like static code analysis, self-checks, testing, and pair programming. Static analysis relies primarily on automated tools, self-checks involve only the author, testing requires code execution, and pair programming is performed continuously during development rather than as a separate step.
Goal
Although direct discovery of quality problems is often the main goal,{{cite book |last1=Baum |first1=Tobias |title=Product-Focused Software Process Improvement |last2=Leßmann |first2=Hendrik |last3=Schneider |first3=Kurt |date=2017 |isbn=978-3-319-69925-7 |series=Lecture Notes in Computer Science |volume=10611 |pages=111–127 |chapter=The Choice of Code Review Process: A Survey on the State of the Practice |doi=10.1007/978-3-319-69926-4_9}} code reviews are usually performed to reach a combination of goals:{{cite book |last1=Baum |first1=Tobias |title=Proceedings of the 2016 24th ACM SIGSOFT International Symposium on Foundations of Software Engineering - FSE 2016 |last2=Liskin |first2=Olga |last3=Niklas |first3=Kai |last4=Schneider |first4=Kurt |date=2016 |isbn=9781450342186 |pages=85–96 |chapter=Factors Influencing Code Review Processes in Industry |doi=10.1145/2950290.2950323 |s2cid=15467294}}
- Improving code quality {{snd}}Improve internal code quality and maintainability through better readability, uniformity, and understandability
- Detecting defects{{snd}}Improve quality regarding external aspects, especially correctness, but also find issues such as performance problems, security vulnerabilities, and injected malware
- Learning/Knowledge transfer{{snd}}Sharing codebase knowledge, solution approaches, and quality expectations, both to the reviewers and the author
- Increase sense of mutual responsibility{{snd}}Increase a sense of collective code ownership and solidarity
- Finding better solutions{{snd}}Generate ideas for new and better solutions and ideas beyond the specific code at hand
- Complying to QA guidelines, ISO/IEC standards{{snd}}Code reviews are mandatory in some contexts, such as air traffic software and safety-critical software
Review types
Several variations of code review processes exist, with additional types specified in IEEE 1028.{{Cite book|title=IEEE Standard for Software Reviews and Audits|publisher=IEEE STD 1028-2008 |date=August 2008 |pages=1–53 |url=https://ieeexplore.ieee.org/document/4601584|doi=10.1109/ieeestd.2008.4601584|isbn=978-0-7381-5768-9 }}
- Management reviews
- Technical reviews
- Inspections
- Walk-throughs
- Audits
= Inspection (formal) =
Historically, the first code review process that was studied and described in detail was called "Inspection" by its inventor, Michael Fagan.{{cite journal |last1=Fagan |first1=Michael |title=Design and code inspections to reduce errors in program development |journal=IBM Systems Journal |date=1976 |volume=15 |issue=3 |pages=182–211|doi=10.1147/sj.153.0182 }} Fagan inspection is a formal process that involves a careful and detailed execution with multiple participants and phases. In formal code reviews, software developers attend a series of meetings to examine code line by line, often using printed copies. Research has shown formal inspections to be extremely thorough and highly effective at identifying defects.
= Regular change-based code review (Walk-throughs) =
Software development teams typically adopt a more lightweight review process in which the scope of each review is on the changes to the codebase performed in a ticket, user story, commit, or some other unit of work.{{cite book |last1=Rigby |first1=Peter |title=Proceedings of the 2013 9th Joint Meeting on Foundations of Software Engineering |last2=Bird |first2=Christian |date=2013 |isbn=9781450322379 |pages=202–212 |chapter=Convergent contemporary software peer review practices |citeseerx=10.1.1.641.1046 |doi=10.1145/2491411.2491444 |s2cid=11163811}} Furthermore, there are rules or conventions that integrate the review task into the development workflow through conventions like mandatory review of all tickets, commonly as part of a pull request, instead of explicitly planning each review. Such a process is called "regular, change-based code review". There are many variations of this basic process.
A 2017 survey of 240 development teams found that 90% of teams using code review followed a change-based process, with 60% specifically using regular change-based review.
Major software corporations known to use changed-based code review include Microsoft,{{cite journal |last1=MacLeod |first1=Laura |last2=Greiler |first2=Michaela |last3=Storey |first3=Margaret-Anne|last4=Bird |first4=Christian|last5=Czerwonka |first5=Jacek|title=Code Reviewing in the Trenches: Challenges and Best Practices |journal= IEEE Software |volume=35 |issue=4 |pages=34 |date=2017 |doi=10.1109/MS.2017.265100500 |s2cid=49651487 | url=https://www.michaelagreiler.com/wp-content/uploads/2019/03/Code-Reviewing-in-the-Trenches-Understanding-Challenges-Best-Practices-and-Tool-Needs.pdf | access-date=2020-11-28 }}
and Facebook.
Efficiency and effectiveness
Ongoing research by Capers Jones analyzing over 12,000 software development projects found formal inspections had a latent defect discovery rate of 60-65%, while informal inspections detected fewer than 50% of defects. The latent defect discovery rate for most forms of testing is about 30%.{{cite web | title=Measuring Defect Potentials and Defect Removal Efficiency | first=Capers | last=Jones | publisher=Crosstalk, The Journal of Defense Software Engineering | date=June 2008 | url=http://www.crosstalkonline.org/storage/issue-archives/2008/200806/200806-0-Issue.pdf | access-date=2010-10-05 | archive-url=https://web.archive.org/web/20120806092322/http://www.crosstalkonline.org/storage/issue-archives/2008/200806/200806-0-Issue.pdf | archive-date=2012-08-06 | url-status=dead }}{{cite journal|title=Embedded Software: Facts, Figures, and Future | journal=Computer | volume=42 | issue=4 | pages=42–52 | first1=Capers | last1=Jones | first2=Christof | last2=Ebert | date=April 2009 | doi=10.1109/MC.2009.118 | s2cid=14008049 }} A code review case study published in the book Best Kept Secrets of Peer Code Review contradicted the Capers Jones study, finding that lightweight reviews can uncover as many bugs as formal reviews while being more efficient in terms of cost and money{{cite book | author = Jason Cohen | title = Best Kept Secrets of Peer Code Review (Modern Approach. Practical Advice.) | publisher = Smart Bear Inc. | year = 2006 | isbn = 978-1-59916-067-2 | url-access = registration | url = https://archive.org/details/bestkeptsecretso00jaso }}
Studies indicate that up to 75% of code review comments affect software evolvability and maintainability rather than functionality,{{cite book | doi=10.1109/ICSE.2015.131 | url=https://www.michaelagreiler.com/wp-content/uploads/2019/02/Code-Reviews-Do-Not-Find-Bugs.-How-the-Current-Code-Review-Best-Practice-Slows-Us-Down.pdf | access-date=2020-11-28 | volume=2 | pages=27–28 | year=2015 | last1=Czerwonka | first1=Jacek | last2=Greiler | first2=Michaela | last3=Tilford | first3=Jack | title=2015 IEEE/ACM 37th IEEE International Conference on Software Engineering | chapter=Code Reviews do Not Find Bugs. How the Current Code Review Best Practice Slows Us Down | isbn=978-1-4799-1934-5 | s2cid=29074469 }}{{cite journal |doi=10.1109/TSE.2008.71 | citeseerx=10.1.1.188.5757 | url=http://lib.tkk.fi/Diss/2009/isbn9789512298570/article5.pdf | access-date=2012-03-21| title=What Types of Defects Are Really Discovered in Code Reviews? | journal=IEEE Transactions on Software Engineering | volume=35 | issue=3 | pages=430–448 | year=2009 | last1=Mantyla | first1=M.V. | last2=Lassenius | first2=C. | s2cid=17570489 }}{{cite web|title=Expectations, outcomes, and challenges of modern code review | first1=A| last1=Bacchelli| first2=C| last2=Bird| publisher= Proceedings of the 35th IEEE/ACM International Conference On Software Engineering (ICSE 2013)| date=May 2013| url=http://sback.it/publications/icse2013.pdf | access-date=2015-09-02}}{{cite web|title=Modern code reviews in open-source projects: which problems do they fix? | first1=M| last1=Beller| first2=A| last2=Bacchelli| first3=A| last3=Zaidman| first4=E| last4=Juergens| publisher= Proceedings of the 11th Working Conference on Mining Software Repositories (MSR 2014)| date=May 2014| url=http://sback.it/publications/msr2014.pdf | access-date=2015-09-02}} suggesting that code reviews are an excellent tool for software companies with long product or system life cycles.{{cite web | title=Does the Modern Code Inspection Have Value? | first1=Harvey | last1=Siy | first2=Lawrence | last2=Votta | url=http://csalpha.ist.unomaha.edu/~hsiy/research/sm.pdf | date=2004-12-01 | access-date=2015-02-17 | website=unomaha.edu | url-status=dead | archive-url=https://web.archive.org/web/20150428192217/http://csalpha.ist.unomaha.edu/~hsiy/research/sm.pdf | archive-date=2015-04-28 }} Therefore, less than 15% of issues discussed in code reviews relate directly to bugs.{{cite web |title=Characteristics of Useful Code Reviews: An Empirical Study at Microsoft | first1=Amiangshu| last1=Bosu | first2=Michaela | last2=Greiler | first3=Chris | last3=Bird | publisher= 2015 IEEE/ACM 12th Working Conference on Mining Software Repositories | date=May 2015| url=https://www.michaelagreiler.com/wp-content/uploads/2019/02/Characteristics-Of-Useful-Comments.pdf | access-date=2020-11-28}}
= Guidelines =
Research indicates review effectiveness correlates with review speed. Optimal code review rates range from 200 to 400 lines of code per hour.{{cite journal|last1=Kemerer|first1=C.F.|last2=Paulk|first2=M.C.|title=The Impact of Design and Code Reviews on Software Quality: An Empirical Study Based on PSP Data|journal=IEEE Transactions on Software Engineering|date=2009-04-17|volume=35|issue=4|pages=534–550|doi=10.1109/TSE.2009.27|hdl=11059/14085 |s2cid=14432409|hdl-access=free}}{{cite web|title=Code Review Metrics|url=https://www.owasp.org/index.php/Code_Review_Metrics#Inspection_Rate|website=Open Web Application Security Project|access-date=9 October 2015|archive-url=https://web.archive.org/web/20151009202719/https://www.owasp.org/index.php/Code_Review_Metrics|archive-date=2015-10-09}}{{cite web|title=Best Practices for Peer Code Review|url=http://smartbear.com/all-resources/articles/best-practices-for-peer-code-review/|website=Smart Bear|publisher=Smart Bear Software|access-date=9 October 2015|archive-url=https://web.archive.org/web/20151009202810/http://smartbear.com/all-resources/articles/best-practices-for-peer-code-review/|archive-date=2015-10-09}}{{cite journal|last1=Bisant|first1=David B.|title=A Two-Person Inspection Method to Improve Programming Productivity|journal=IEEE Transactions on Software Engineering|date=October 1989|volume=15|issue=10|pages=1294–1304|doi=10.1109/TSE.1989.559782|s2cid=14921429|url=http://dl.acm.org/citation.cfm?id=77604|access-date=9 October 2015|url-access=subscription}} Inspecting and reviewing more than a few hundred lines of code per hour for critical software (such as safety critical embedded software) may be too fast to find errors.{{cite web|title=A Guide to Code Inspections | first=Jack | last=Ganssle | publisher= The Ganssle Group | date=February 2010 | url=http://www.ganssle.com/inspections.pdf | access-date=2010-10-05}}
= Supporting tools =
Static code analysis software assist reviewers by automatically checking source code for known vulnerabilities and defect patterns, particularly for large chunks of code.{{Cite book | doi=10.1109/ICSE.2013.6606642 | isbn=978-1-4673-3076-3 | chapter=Reducing human effort and improving quality in peer code reviews using automatic static analysis and reviewer recommendation | title=2013 35th International Conference on Software Engineering (ICSE) | pages=931–940 | year=2013 | last1=Balachandran | first1=Vipin | s2cid=15823436 }} A 2012 study by VDC Research reports that 17.6% of the embedded software engineers surveyed currently use automated tools to support peer code review and 23.7% plan to use them within two years.{{cite web|title=Automated Defect Prevention for Embedded Software Quality | last=VDC Research| publisher = VDC Research| date=2012-02-01 | url=http://alm.parasoft.com/embedded-software-vdc-report/ | access-date=2012-04-10}}
See also
External links
- [https://blogs.oracle.com/javamagazine/five-code-review-antipatterns Five Code Review Antipatterns]
- [https://blogs.oracle.com/javamagazine/the-best-of-2020-the-10-most-popular-java-magazine-articles Java Magazine, Best of 2020]