Version Control Systems: SVN and GIT How do VCS support SW ...

Version Control Systems:SVN and GIT

How do VCS support SW development teams?

CS 435/535

The College of William and Mary

Chapter 3 Agile software development

Agile manifesto

We are uncovering better ways of developing software by doing it and helping others do it. Through this work we have come to value:

Individuals and interactions over processes and toolsWorking software over comprehensive documentation Customer collaboration over contract negotiation Responding to change over following a plan

That is, while there is value in the items on the right, we value the items on the left more.

2

What is needed?


Plan-driven and agile specification

Requirementsspecification

Requirementsengineering

Design andimplementation

Requirements changerequests

Plan-based development

Agile development

Requirementsengineering

Design andimplementation

3


The extreme programming release cycle

Break downstories to tasks

Select userstories for this

releasePlan release

Releasesoftware

Evaluatesystem

Develop/integrate/test software

4


The Scrum process

Outline planningand architectural

designProject closure

Assess Select

Review Develop

Sprint cycle

5


Scrum benefits

The product is broken down into a set of manageable and understandable chunks.

Unstable requirements do not hold up progress.The whole team have visibility of everything and

consequently team communication is improved.Customers see on-time delivery of increments and gain

feedback on how the product works.Trust between customers and developers is established

and a positive culture is created in which everyone expects the project to succeed.

6

Software Engineering is Team Work

• Enabling technology for productivity

• must support parallelization

• must support communication

• Documentation as preserved communication

• must support management of tasks & people

• What needs to be done? When? By whom?

• What has been done? By whom?

Remember SVN from CS 301?

What does it support?

Version Control Systems

Centralized •CVS – 1990•SVN - 2000

Distributed •Bitkeeper - 1997•Git – 2005•Bazaar – 2005•Mercurial - 2005

*More VCS at http://en.wikipedia.org/wiki/Comparison_of_revision_control_software

Version Control Systems

• Version control system

• supports concurrent software development on shared code base

• keeps track of changes,

• integrates versions / recognizes conflicts,

• allows for recovery, documentation of changes

• Common set up:

• IDE as front end, VCS as back end (shared, persistent storage)

Subclipse: eclipse plugin for SVN

Subclipse: eclipse plugin for SVN

http://subclipse.tigris.org/update_1.8.x

EGit: eclipse plugin for Git


http://download.eclipse.org/egit/updates


http://eclipsesource.com/blogs/tutorials/egit-tutorial/

http://wiki.eclipse.org/EGit/User_Guide#Overview

Centralized vs Distributed Version Control Systems

Image from http://git-scm.com/book/en/Getting-Started-About-Version-Control

Centralized Architecture:

Distributed Version Control Systems

Image from http://git-scm.com/book/en/Getting-Started-About-Version-Control

Distributed Architecture:

Centralized vs Distributed VCS

• What are the pros & cons?

• Software engineering is much about scalability:

• Project size in # of developers

• about 10

• up to 100

• more than 100

Workflows: Centralized

Image from http://git-scm.com/book/en/Distributed-Git-Distributed-Workflows

• Small teams• Typical workflow for SVN

and CVS• Repository is a single point

of failure

Workflows: Integration - Manager


• Supported by CVS and SVN using branches

• More easily supported by distributed version control systems

Workflows: Director and Lieutenants




• Generally used by huge projects (e.g., Linux kernel)

Versions, Revisions, and Snapshots

• CVS: each commit generates a new version for each file modified

• SVN: each commit generates new state of the file system tree, called a revision

• GIT: same than SVN; keeps a snapshot of the system but instead of saving the deltas it saves the changed files and references to the unchanged ones

Git follows idea of a file system with snapshots

CHAPTER 1 GETTING STARTED

so on) think of the information they keep as a set of files and the changes made to each

file over time, as illustrated in Figure 1.4.

Figure 1.4: Other systems tend to store data as changes to a base version of each file.

Git doesn’t think of or store its data this way. Instead, Git thinks of its data more

like a set of snapshots of a mini filesystem. Every time you commit, or save the state

of your project in Git, it basically takes a picture of what all your files look like at

that moment and stores a reference to that snapshot. To be efficient, if files have not

changed, Git doesn’t store the file again—just a link to the previous identical file it has

already stored. Git thinks about its data more like Figure 1.5.

Figure 1.5: Git stores data as snapshots of the project over time.

This is an important distinction between Git and nearly all other VCSs. It makes

Git reconsider almost every aspect of version control that most other systems copied

from the previous generation. This makes Git more like a mini filesystem with some

incredibly powerful tools built on top of it, rather than simply a VCS. We’ll explore

some of the benefits you gain by thinking of your data this way when we cover Git

branching in Chapter 3.

1.3.2 Nearly Every Operation Is Local

Most operations in Git only need local files and resources to operate generally no

information is needed from another computer on your network. If you’re used to a

CVCS where most operations have that network latency overhead, this aspect of Git

will make you think that the gods of speed have blessed Git with unworldly powers.

Because you have the entire history of the project right there on your local disk, most

operations seem almost instantaneous.

5

SVN et al:

CHAPTER 1 GETTING STARTED

so on) think of the information they keep as a set of files and the changes made to each

file over time, as illustrated in Figure 1.4.

Figure 1.4: Other systems tend to store data as changes to a base version of each file.

Git doesn’t think of or store its data this way. Instead, Git thinks of its data more

like a set of snapshots of a mini filesystem. Every time you commit, or save the state

of your project in Git, it basically takes a picture of what all your files look like at

that moment and stores a reference to that snapshot. To be efficient, if files have not

changed, Git doesn’t store the file again—just a link to the previous identical file it has

already stored. Git thinks about its data more like Figure 1.5.

Figure 1.5: Git stores data as snapshots of the project over time.

This is an important distinction between Git and nearly all other VCSs. It makes

Git reconsider almost every aspect of version control that most other systems copied

from the previous generation. This makes Git more like a mini filesystem with some

incredibly powerful tools built on top of it, rather than simply a VCS. We’ll explore

some of the benefits you gain by thinking of your data this way when we cover Git

branching in Chapter 3.

1.3.2 Nearly Every Operation Is Local

Most operations in Git only need local files and resources to operate generally no

information is needed from another computer on your network. If you’re used to a

CVCS where most operations have that network latency overhead, this aspect of Git

will make you think that the gods of speed have blessed Git with unworldly powers.

Because you have the entire history of the project right there on your local disk, most

operations seem almost instantaneous.

5

Versions, Revisions, and Snapshots

Image from http://svnbook.red-bean.com/en/1.7/svn.basic.in-action.html

SVN and Git use global revision numbers

Operations and states (CVS and SVN)

Workspace Repository

Checkout

Commit

Operations and states (Git)

Workspace Repository

Checkout

Commit

Staging area (Index)

Stage

Operations and commands - Git

http://osteele.com/posts/2008/05/commit-policies

Workflows: Integration - Manager




Operations and commands

Operation CVS SVN Git

Init init create init

Import import import commit

Checkout checkout checkout clone

Checkout branch checkout checkout checkout

Commit/Checkin commit commit commit, push

Update update update fetch, pull

Operations and commands - SVN+Eclipse






Workflows and issues

• Workflow: 1) get code base 2) make changes 3) deliver changes

• Issue: Read/write access to remote repository

• Protected: User authentication, registration, account/pw necessary in communication, IDE stores/uses account/pw for convenience

• Issue: Conflicts

• Changes do not fit together, automatically recognized at some level of granularity (same file, same method, same line of code)

• Automatically recognized, manually fixed

• Issue: Documentation / Communication

• What changed, how trustworthy are the changes, what needs to be changed as an effect

• Finding the right historical version to undo some changes

Tagging

• Useful for marking specific points in history, in particular: Releases

• Two types: lightweight vs annotated

• annotated: full objects in Git DB, check summed, contain tagger name, email, date, tagging message, can be signed & verifiedCHAPTER 2 GIT BASICS

$ git show v1.4

tag v1.4

Tagger: Scott Chacon <[email protected]>

Date: Mon Feb 9 14:45:11 2009 -0800

my version 1.4

commit 15027957951b64cf874c3557a0f3547bd83b3ff6

Merge: 4a447f7... a6b4c97...

Author: Scott Chacon <[email protected]>

Date: Sun Feb 8 19:02:46 2009 -0800

Merge branch ’experiment’

That shows the tagger information, the date the commit was tagged, and the anno-tation message before showing the commit information.

2.6.4 Signed Tags

You can also sign your tags with GPG, assuming you have a private key. All you haveto do is use -s instead of -a:

$ git tag -s v1.5 -m ’my signed 1.5 tag’

You need a passphrase to unlock the secret key for

user: "Scott Chacon <[email protected]>"

1024-bit DSA key, ID F721C45A, created 2009-02-09

If you run git show on that tag, you can see your GPG signature attached to it:

$ git show v1.5

tag v1.5

Tagger: Scott Chacon <[email protected]>

Date: Mon Feb 9 15:22:20 2009 -0800

my signed 1.5 tag

-----BEGIN PGP SIGNATURE-----

Version: GnuPG v1.4.8 (Darwin)

iEYEABECAAYFAkmQurIACgkQON3DxfchxFr5cACeIMN+ZxLKggJQf0QYiQBwgySN

Ki0An2JeAVUCAiJ7Ox6ZEtK+NvZAj82/

=WryJ

-----END PGP SIGNATURE-----

commit 15027957951b64cf874c3557a0f3547bd83b3ff6

Merge: 4a447f7... a6b4c97...

Author: Scott Chacon <[email protected]>

Date: Sun Feb 8 19:02:46 2009 -0800

Merge branch ’experiment’

A bit later, you’ll learn how to verify signed tags.

37

• $ git tag -a v1.4 -m ’my version 1.4’

Branching

• CVS: simple process for creating branches on the repository

• SVN: has no internal concept of a branch; branches are managed as copies of a directory.

• GIT: very simple process for creating local and remote branches

Merging

1 2 3 5 6

4Branch

1 2 3 5 6

4Branch

7

Merge

Branching in SVN

Branching in SVN

Branching in SVN

Branching in SVN

Branching in SVN

Branching in SVN

Branching in SVN

Branching in Git

• Branches are lightweight movable pointers to commits

• The default branch is the MASTER (trunk)

Images from http://git-scm.com/book/en/Git-Branching-Basic-Branching-and-Merging

Branching in Git

Initial layout for three commits

New branch pointer (iss53)


Branching in Git

New commit on the branch

Hot fix branch on master


Merging in Git


Merging in Git


Branching & Merging in Git

• Key concept, well supported

• Local workflow

• people create branches for any issue / task /assignment they deal with, sometimes called “topic” branch

• optional: rebase instead of merge to obtain a linear history

• only recommended for local repository

• Remote repository: merge with master

• For integration manager with blessed repository: pull request

What is missing so far?

• Documentation of problems, bug reports

• Work assignments, who does what and till when

Issue tracking

Github Issue Tracker

• Filter by open and closed issues, assignees, labels, and milestones.

• Sort by issue age, number of comments, and update time.

• Milestones / labels

Github Workflow: Code review & Pull request

• Pull request starts conversation around proposed changes. Additional commits may add to branch before merging into master.

• Pull Request = Code + Issue + Code Comments

Software Engineering is Team Work

• Enabling technology for productivity

• must support parallelization

• must support communication

• Documentation as preserved communication

• must support management of tasks & people

• What needs to be done? When? By whom?

• What has been done? By whom?

Version Control Systems: SVN and GIT How do VCS support SW ...

Documents