book

Version Control with Git, 2nd Edition

by Jon Loeliger, Matthew McCullough

August 2012

Intermediate to advanced

452 pages

12h 39m

English

O'Reilly Media, Inc.

Read now

Unlock full access

AudienceAssumed FrameworkBook Layout and OmissionsConventions Used in This BookUsing Code ExamplesSafari® Books OnlineHow to Contact UsAcknowledgmentsAttributions
BackgroundThe Birth of GitPrecedentsTimelineWhat’s in a Name?
Using Linux Binary DistributionsDebian/UbuntuOther Binary DistributionsObtaining a Source ReleaseBuilding and InstallingInstalling Git on WindowsInstalling the Cygwin Git PackageInstalling Standalone Git (msysGit)
The Git Command LineQuick Introduction to Using GitCreating an Initial RepositoryAdding a File to Your RepositoryConfiguring the Commit AuthorMaking Another CommitViewing Your CommitsViewing Commit DifferencesRemoving and Renaming Files in Your RepositoryMaking a Copy of Your RepositoryConfiguration FilesConfiguring an AliasInquiry
Basic ConceptsRepositoriesGit Object TypesIndexContent-Addressable NamesGit Tracks ContentPathname Versus ContentPack FilesObject Store PicturesGit Concepts at WorkInside the .git DirectoryObjects, Hashes, and BlobsFiles and TreesA Note on Git’s Use of SHA1Tree HierarchiesCommitsTags
It’s All About the IndexFile Classifications in GitUsing git addSome Notes on Using git commitUsing git commit --allWriting Commit Log MessagesUsing git rmUsing git mvA Note on Tracking RenamesThe .gitignore FileA Detailed View of Git’s Object Model and Files
Atomic ChangesetsIdentifying CommitsAbsolute Commit Namesrefs and symrefsRelative Commit NamesCommit HistoryViewing Old CommitsCommit GraphsUsing gitk to View the Commit GraphCommit RangesFinding CommitsUsing git bisectUsing git blameUsing Pickaxe
Reasons for Using BranchesBranch NamesDos and Don’ts in Branch NamesUsing BranchesCreating BranchesListing Branch NamesViewing BranchesChecking out BranchesA Basic Example of Checking out a BranchChecking out When You Have Uncommitted ChangesMerging Changes into a Different BranchCreating and Checking out a New BranchDetached HEAD BranchesDeleting Branches
Forms of the git diff CommandSimple git diff Examplegit diff and Commit Rangesgit diff with Path LimitingComparing How Subversion and Git Derive diffs
Merge ExamplesPreparing for a MergeMerging Two BranchesA Merge with a ConflictWorking with Merge ConflictsLocating Conflicted FilesInspecting Conflictsgit diff with conflictsgit log with conflictsHow Git Keeps Track of ConflictsFinishing Up a Conflict ResolutionAborting or Restarting a MergeMerge StrategiesDegenerate MergesNormal MergesRecursive mergesOctopus mergesSpecialty MergesApplying Merge StrategiesMerge DriversHow Git Thinks About MergesMerges and Git’s Object ModelSquash MergesWhy Not Just Merge Each Change One by One?

Caution About Altering HistoryUsing git resetUsing git cherry-pickUsing git revertreset, revert, and checkoutChanging the Top CommitRebasing CommitsUsing git rebase -irebase Versus merge
The StashThe Reflog
Repository ConceptsBare and Development RepositoriesRepository ClonesRemotesTracking BranchesReferencing Other RepositoriesReferring to Remote RepositoriesThe refspecExample Using Remote RepositoriesCreating an Authoritative RepositoryMake Your Own Origin RemoteDeveloping in Your RepositoryPushing Your ChangesAdding a New DeveloperGetting Repository UpdatesThe fetch stepThe merge or rebase stepShould you merge or rebase?Remote Repository Development Cycle in PicturesCloning a RepositoryAlternate HistoriesNon–Fast-Forward PushesFetching the Alternate HistoryMerging HistoriesMerge ConflictsPushing a Merged HistoryRemote ConfigurationUsing git remoteUsing git configUsing Manual EditingWorking with Tracking BranchesCreating Tracking BranchesAhead and BehindAdding and Deleting Remote BranchesBare Repositories and git push
A Word About ServersPublishing RepositoriesRepositories with Controlled AccessRepositories with Anonymous Read AccessPublishing repositories using git-daemonPublishing repositories using an HTTP daemonPublishing a repository using Smart HTTPPublishing via Git and HTTP daemonsRepositories with Anonymous Write AccessPublishing Your Repository to GitHubRepository Publishing AdviceRepository StructureThe Shared Repository StructureDistributed Repository StructureRepository Structure ExamplesLiving with Distributed DevelopmentChanging Public HistorySeparate Commit and Publish StepsNo One True HistoryKnowing Your PlaceUpstream and Downstream FlowsThe Maintainer and Developer RolesMaintainer–Developer InteractionRole DualityWorking with Multiple RepositoriesYour Own WorkspaceWhere to Start Your RepositoryConverting to a Different Upstream RepositoryUsing Multiple Upstream RepositoriesForking ProjectsIsn’t forking a project bad?Reconciling forksForking projects at GitHub
Why Use Patches?Generating PatchesPatches and Topological SortsMailing PatchesApplying PatchesBad PatchesPatching Versus Merging
Installing HooksExample HooksCreating Your First HookAvailable HooksCommit-Related HooksPatch-Related HooksPush-Related HooksOther Local Repository Hooks
The Old Solution: Partial CheckoutsThe Obvious Solution: Import the Code into Your ProjectImporting Subprojects by CopyingImporting Subprojects with git pull -s subtreeSubmitting Your Changes UpstreamThe Automated Solution: Checking out Subprojects Using Custom ScriptsThe Native Solution: gitlinks and git submoduleGitlinksThe git submodule Command
Submodule CommandsWhy Submodules?Submodules PreparationWhy Read Only?Why Not Read Only?Examining the Hashes of Submodule CommitsCredential ReuseUse CasesMultilevel Nesting of ReposSubmodules on the Horizon
Example: A Shallow Clone of a Single BranchMaking Your Changes in GitFetching Before CommittingCommitting Through git svn rebasePushing, Pulling, Branching, and Merging with git svnKeeping Your Commit IDs StraightCloning All the BranchesSharing Your RepositoryMerging Back into SubversionHow dcommit handles mergesMiscellaneous Notes on Working with Subversionsvn:ignore Versus .gitignoreReconstructing the git-svn Cache
Using git filter-branchExamples Using git filter-branchUsing git filter-branch to expunge a fileUsing filter-branch to edit a commit messagefilter-branch PitfallsHow I Learned to Love git rev-listDate-Based CheckoutDate-based checkout cautionsRetrieve Old Version of a FileInteractive Hunk StagingRecovering a Lost CommitThe git fsck CommandReconnecting a Lost Commit
Interactive Rebase with a Dirty Working DirectoryRemove Left-Over Editor FilesGarbage CollectionSplit a RepositoryTips for Recovering CommitsSubversion Conversion TipsGeneral AdviceRemove a Trunk After an SVN ImportRemoving SVN Commit IDsManipulating Branches from Two RepositoriesRecovering from an Upstream RebaseMake Your Own Git CommandQuick Overview of ChangesCleaning UpUsing git-grep to Search a RepositoryUpdating and Deleting refsFollowing Files that MovedKeep, But Don’t Track, This FileHave You Been Here Before?
Repo for Public CodeCreating a GitHub RepositorySocial Coding on Open SourceWatchersNews FeedForksCreating Pull RequestsManaging Pull RequestsNotificationsFinding Users, Projects, and CodeWikisGitHub Pages (Git for Websites)In-Page Code EditorSubversion BridgeTags Automatically Becoming ArchivesOrganizationsREST APISocial Coding on Closed SourceEventual Open SourcingCoding ModelsGitHub EnterpriseGitHub in Sum

Content preview from Version Control with Git, 2nd Edition

Chapter 11. The Stash and the Reflog

The Stash

Do you ever feel overwhelmed in your daily development cycle when the constant interruptions, demands for bug fixes, and requests from coworkers or managers all pile up and clutter the real work you are trying to do? If so, the stash was designed to help you!

The stash is a mechanism for capturing your work in progress, allowing you to save it and return to it later when convenient. Sure, you can already do that using the existing branch and commit mechanisms within Git, but the stash is a quick convenience mechanism that allows a complete and thorough capturing of your index and working directory in one simple command. It leaves your repository clean, uncluttered, and ready for an alternate development direction. Another single command restores that index and working directory state completely, allowing you to resume where you left off.

Let’s see how the stash works with the canonical use case: the so-called “interrupted work flow.”

In this scenario, you are happily working in your Git repository and have changed several files and maybe even staged a few in the index. Then, some interruption happens. Perhaps a critical bug is discovered and lands on your plate and must be fixed immediately. Perhaps your team lead has suddenly prioritized a new feature over everything else and insists you drop everything to work on it. Whatever the circumstance, you realize you must stash everything, clean your slate and work tree, and start afresh. This ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Start your free trial

Publisher Resources

ISBN: 9781449345037Errata Page

Version Control with Git, 2nd Edition

by Jon Loeliger, Matthew McCullough

Chapter 11. The Stash and the Reflog

The Stash

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

You might also like