使用Subversion进行版本控制

For Subversion 1.5(Compiled from r3133)

Ben Collins-Sussman

Brian W. Fitzpatrick

C. Michael Pilato

本作品使用共同创造许可证,可以访问http://creativecommons.org/licenses/by/2.0/或发送邮件到Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.查看本许可证。

(TBA)


目录

前言
序言
读者
How to Read This Book
本书约定
排版习惯
Tips and Warnings
本书的结构
This Book Is Free
致谢
来自 Ben Collins-Sussman
来自 Brian W. Fitzpatrick
来自 C. Michael Pilato
What Is Subversion?
Is Subversion the Right Tool?
Subversion的历史
Subversion的特性
Subversion的架构
Subversion的组件
What's New in Subversion
1. 基本概念
版本库
版本模型
The Problem of File Sharing
锁定-修改-解锁 方案
拷贝-修改-合并 方案
Subversion实践
Subversion版本库URL
工作拷贝
修订版本
工作拷贝怎样跟踪版本库
混合修订版本的工作拷贝
总结
2. 基本使用
求助!
Getting Data into Your Repository
svn import
Recommended Repository Layout
初始化检出
禁用密码缓存
用其它身份认证
基本的工作周期
更新你的工作拷贝
修改你的工作拷贝
检查你的修改
取消本地修改
解决冲突(合并别人的修改)
提交你的修改
检验历史
Generating a List of Historical Changes
Examining the Details of Historical Changes
Browsing the Repository
Fetching Older Repository Snapshots
有时你只需要清理
Disposing of a Working Copy
Recovering from an Interruption
总结
3. 高级主题
版本清单
修订版本关键字
版本日期
属性
为什么需要属性?
操作属性
属性和 Subversion 工作流程
自动设置属性
文件移植性
文件内容类型
文件的可执行性
行结束字符串
忽略未版本控制的条目
关键字替换
Sparse Directories
锁定
Creating Locks
Discovering Locks
Breaking and Stealing Locks
锁定交流
外部定义
Peg和实施修订版本
Changelists
Creating and Modifying Changelists
Changelists as Operation Filters
Changelist Limitations
网络模型
请求和响应
客户端凭证缓存
4. 分支与合并
什么是分支?
使用分支
创建分支
在分支上工作
The Key Concepts Behind Branching
Basic Merging
Changesets
Keeping a Branch in Sync
Mergeinfo and Previews
取消修改
找回删除的项目
Advanced Merging
Cherrypicking
Merge Syntax: Full Disclosure
Merges Without Mergeinfo
More on Merge Conflicts
Blocking Changes
Merge-Sensitive Logs and Annotations
关注还是忽视祖先
合并和移动
Blocking Merge-Unaware Clients
使用分支
标签
建立简单标签
建立复杂标签
分支维护
版本库布局
数据的生命周期
常用分支模式
发布分支
特性分支
Vendor Branches
常规的供方分支管理过程
svn_load_dirs.pl
总结
5. 版本库管理
Subversion 版本库的定义
版本库开发策略
规划你的版本库结构
决定在哪里与如何部署你的版本库
选择数据存储格式
创建和配置你的版本库
创建版本库
实现版本库钩子
Berkeley DB 配置
版本库维护
管理员的工具箱
修正提交消息
管理磁盘空间
Berkeley DB 恢复
版本库数据的移植
过滤版本库历史
版本库复制
版本库备份
Managing Repository UUIDs
Moving and Removing Repositories
总结
6. 服务配置
概述
选择一个服务器配置
svnserve服务器
svnserve使用SSH通道
Apache 的 HTTP 服务器
推荐
svnserve, a Custom Server
调用服务器
Built-in Authentication and Authorization
Using svnserve with SASL
SSH 隧道
SSH 配置技巧
httpd, the Apache HTTP Server
先决条件
基本的 Apache 配置
认证选项
授权选项
额外的糖果
基于路径的授权
支持多种版本库访问方法
7. 定制你的Subversion体验
运行配置区
配置区布局
配置和Windows注册表
配置选项
本地化
Understanding Locales
Subversion's Use of Locales
Using External Editors
Using External Differencing and Merge Tools
外置 diff
外置 diff3
8. 嵌入Subversion
分层的库设计
版本库层
版本库访问层
客户端层
进入工作拷贝的管理区
条目文件
原始拷贝和属性文件
使用API
Apache可移植运行库
URL 和路径需求
使用 C 和 C++ 以外的语言
代码样例
9. Subversion 完全参考
The Subversion Command-Line Client: svn
svn选项
svn子命令
svnadmin
svnadmin选项
svnadmin子命令
svnlook
svnlook选项
svnlook子命令
svnsync
svnsync选项
svnsync子命令
svnserve
svnserve选项
svnversion
mod_dav_svn
Subversion Properties
版本控制的属性
未版本控制的属性
版本库钩子
A. Subversion 快速入门指南
安装 Subversion
快速指南
B. CVS用户的Subversion指南
版本号现在不同了
目录的版本
更多离线操作
区分状态和更新
状态
更新
分支和标签
元数据属性
解决冲突
二进制文件和行结束标记转换
版本化的模块
认证
迁移CVS版本库到Subversion
C. WebDAV和自动版本
What Is WebDAV?
自动版本化
客户端交互性
Standalone WebDAV Applications
File-Explorer WebDAV Extensions
WebDAV Filesystem Implementation
D. Copyright
索引

插图清单

1. Subversion's architecture
1.1. 一个典型的客户/服务器系统
1.2. 需要避免的问题
1.3. 锁定-修改-解锁 方案
1.4. 拷贝-修改-合并 方案
1.5. 拷贝-修改-合并 方案(续)
1.6. 版本库的文件系统
1.7. 版本库
4.1. 分支与开发
4.2. 开始规划版本库
4.3. 版本库与复制
4.4. 一个文件的分支历史
8.1. 二维的文件和目录
8.2. 版本时间—第三维!

表格清单

1.1.
4.1.
5.1.
6.1.
C.1.

范例清单

5.1. txn-info.sh (reporting outstanding transactions)
5.2. 镜像版本库的 pre-revprop-change 钩子
5.3. 镜像版本库的 start-commit 钩子
6.1. 匿名访问的配置实例。
6.2. 一个认证访问的配置实例。
6.3. A sample configuration for mixed authenticated/anonymous access
6.4. 禁用所有的路径检查
7.1. Sample registration entries (.reg) file.
7.2. diffwrap.sh
7.3. diffwrap.bat
7.4. diff3wrap.sh
7.5. diff3wrap.bat
8.1. 使用版本库层
8.2. 使用 Python 处理版本库层
8.3. A Python status crawler

前言

Karl Fogel

芝加哥,2004年3月14日

A bad Frequently Asked Questions (FAQ) sheet is one that is composed not of the questions people actually ask, but of the questions the FAQ's author wishes people would ask. Perhaps you've seen the type before:

Q:怎样使用Glorbosoft XYZ最大程度的提高团队生产率?

A: Many of our customers want to know how they can maximize productivity through our patented office groupware innovations. The answer is simple. First, click on the File menu, scroll down to Increase Productivity, then…

The problem with such FAQs is that they are not, in a literal sense, FAQs at all. No one ever called the tech support line and asked, “How can we maximize productivity?”. Rather, people asked highly specific questions, such as “How can we change the calendaring system to send reminders two days in advance instead of one?” and so on. But it's a lot easier to make up imaginary Frequently Asked Questions than it is to discover the real ones. Compiling a true FAQ sheet requires a sustained, organized effort: over the lifetime of the software, incoming questions must be tracked, responses monitored, and all gathered into a coherent, searchable whole that reflects the collective experience of users in the wild. It calls for the patient, observant attitude of a field naturalist. No grand hypothesizing, no visionary pronouncements here—open eyes and accurate note-taking are what's needed most.

我很喜欢这本书,因为它正是按照这种精神建立起来的,这种精神体现在本书的每一页中。这是作者与用户直接交流的结果。而这一切是源于Ben Collins-Sussman's对于Subversion常见问题邮件列表的研究。他发现人们总是在邮件列表中重复询问一些基本问题:使用subversion的一般程序是怎样的?分支与标签同其它版本控制系统的工作方式一样吗?我怎样知道某一处修改是谁做的?

Frustrated at seeing the same questions day after day, Ben worked intensely over a month in the summer of 2002 to write The Subversion Handbook, a 60 page manual that covered all the basics of using Subversion. The manual made no pretense of being complete, but it was distributed with Subversion and got users over that initial hump in the learning curve. When O'Reilly decided to publish a full-length Subversion book, the path of least resistance was obvious: just expand the Subversion handbook.

The three coauthors of the new book were thus presented with an unusual opportunity. Officially, their task was to write a book top-down, starting from a table of contents and an initial draft. But they also had access to a steady stream—indeed, an uncontrollable geyser—of bottom-up source material. Subversion was already in the hands of thousands of early adopters, and those users were giving tons of feedback, not only about Subversion, but about its existing documentation.

在写这本书的过程里,Ben,Mike 和 Brian一直像鬼魂一样游荡在Subversion邮件列表和聊天室中,仔细的研究用户实际遇到的问题。监视这些反馈也是他们在CollabNet工作的一部分,这给他们撰写Subversion文档提供了巨大的便利。这本书建立在丰富的使用经验,而非在流沙般脆弱的想象之上,它结合了用户手册和FAQ的优点。初次阅读时,这种二元性的优势并不明显,按照顺序,从前到后,这本书只是简单的从头到尾描述了软件的细节。书中的内容包括一章概述,一章必不可少的快速指南,一章关于管理配置,一些高级主题,当然还包括命令参考手册和故障排除指南。而当你过一段时间之后,再次翻开本书查找一些特定问题的解决方案时,这种二元性才得以显现:这些生动的细节一定来自不可预测的实际用例的提炼,大多是源于用户的需要和视点。

Of course, no one can promise that this book will answer every question you have about Subversion. Sometimes, the precision with which it anticipates your questions will seem eerily telepathic; yet occasionally, you will stumble into a hole in the community's knowledge and come away empty-handed. When this happens, the best thing you can do is email and present your problem. The authors are still there and still watching, and the authors include not just the three listed on the cover, but many others who contributed corrections and original material. From the community's point of view, solving your problem is merely a pleasant side effect of a much larger project—namely, slowly adjusting this book, and ultimately Subversion itself, to more closely match the way people actually use it. They are eager to hear from you, not only because they can help you, but because you can help them. With Subversion, as with all active free software projects, you are not alone.

让这本书将成为你的第一个伙伴。

序言

 

即使你能确认什么是完美,也不要让完美成为好的敌人,更何况你不能确认。因为落入过去陷阱的不悦,你会在设计时因为担心自己的缺陷而无所作为。

 
  --Greg Hudson, Subversion developer

In the world of open source software, the Concurrent Versions System (CVS) was the tool of choice for version control for many years. And rightly so. CVS was open source software itself, and its nonrestrictive modus operandi and support for networked operation allowed dozens of geographically dispersed programmers to share their work. It fit the collaborative nature of the opensource world very well. CVS and its semi-chaotic development model have since become cornerstones of open source culture.

But CVS was not without its flaws, and simply fixing those flaws promised to be an enormous effort. Enter Subversion. Designed to be a successor to CVS, Subversion's originators set out to win the hearts of CVS users in two ways—by creating an open source system with a design (and “look and feel”) similar to CVS, and by attempting to avoid most of CVS's noticeable flaws. While the result isn't necessarily the next great evolution in version control design, Subversion is very powerful, very usable, and very flexible. And for the most part, almost all newly started open source projects now choose Subversion instead of CVS.

This book is written to document the 1.5 series of the Subversion version control system. We have made every attempt to be thorough in our coverage. However, Subversion has a thriving and energetic development community, so there are already a number of features and improvements planned for future versions that may change some of the commands and specific notes in this book.

读者

This book is written for computer-literate folk who want to use Subversion to manage their data. While Subversion runs on a number of different operating systems, its primary user interface is command-line-based. That command-line tool (svn), and some auxiliary programs, are the focus of this book.

For consistency, the examples in this book assume that the reader is using a Unix-like operating system and is relatively comfortable with Unix and command-line interfaces. That said, the svn program also runs on non-Unix platforms such as Microsoft Windows. With a few minor exceptions, such as the use of backward slashes (\) instead of forward slashes (/) for path separators, the input to and output from this tool when run on Windows are identical to its Unix counterpart.

大多数读者可能是那些需要跟踪代码变化的程序员或者系统管理员,这是Subversion最普遍的用途,因此这个场景贯穿于整本书的例子中。但是Subversion可以用来管理任何类型的数据:图像、音乐、数据库、文档等等。对于Subversion,数据就是数据而已。

While this book is written with the assumption that the reader has never used a version control system, we've also tried to make it easy for users of CVS (and other systems) to make a painless leap into Subversion. Special sidebars may mention other version control systems from time to time, and Appendix B summarizes many of the differences between CVS and Subversion.

Note also that the source code examples used throughout the book are only examples. While they will compile with the proper compiler incantations, they are intended to illustrate a particular scenario and not necessarily serve as examples of good programming style or practices.

How to Read This Book

Technical books always face a certain dilemma: whether to cater to top-down or to bottom-up learners. A top-down learner prefers to read or skim documentation, getting a large overview of how the system works; only then does she actually start using the software. A bottom-learner is a “learn by doing” person—someone who just wants to dive into the software and figure it out as she goes, referring to book sections when necessary. Most books tend to be written for one type of person or the other, and this book is undoubtedly biased towards top-down learners. (And if you're actually reading this section, you're probably already a top-down learner yourself!) However, if you're a bottom-up person, don't despair. While the book may be laid out as a broad survey of Subversion topics, the content of each section tends to be heavy with specific examples that you can try-by-doing. For the impatient folks who just want to get going, you can jump right to 附录 A, Subversion 快速入门指南.

Regardless of your learning style, this book aims to be useful to people of widely different backgrounds—from those with no previous experience in version control to experienced system administrators. Depending on your own background, certain chapters may be more or less important to you. The following can be considered a “recommended reading list” for various types of readers:

资深系统管理员

The assumption here is that you've probably used version control before and are dying to get a Subversion server up and running ASAP. 第 5 章 版本库管理 and 第 6 章 服务配置 will show you how to create your first repository and make it available over the network. After that's done, 第 2 章 基本使用 and 附录 B, CVS用户的Subversion指南 are the fastest routes to learning the Subversion client.

新用户

如果管理员已经为你准备好了Subversion服务,你所需要的是学习如何使用客户端。如果你没有使用版本控制系统(像CVS)的经验,那么第 1 章 基本概念第 2 章 基本使用是重要的入门教程,其中介绍了版本控制的重要思想。

高级用户

Whether you're a user or administrator, eventually your project will grow larger. You're going to want to learn how to do more advanced things with Subversion, such as how to use Subversion's property support (第 3 章 高级主题), how to use branches and perform merges (第 4 章 分支与合并), how to configure runtime options (第 7 章 定制你的Subversion体验), and other things. These chapters aren't critical at first, but be sure to read them once you're comfortable with the basics.

开发者

你应该已经很熟悉Subversion了,并且想扩展它或使用它的API开发新软件。第 8 章 嵌入Subversion将最适合你。

本书以参考材料作为结束—第 9 章 Subversion 完全参考是一部Subversion全部命令的详细指南,此外,在附录中还有许多很有意义的主题。阅读完本书后,这些章节将会是你经常查阅的内容。

本书约定

本节描述了本书中使用的各种约定。

排版习惯

The following typographic conventions are used in this book:

等宽字体

用于命令,命令输出和选项

等宽字体

用于代码和文本中的可替换部分

斜体

Used for file and directory names as well as for new terms

Tips and Warnings

注意

此图标表示旁边的文本内容需特别注意。

提示

此图标表示旁边的文本描述了一个有用的小技巧。

警告

此图标表示旁边的文本是警告信息。

本书的结构

以下是各个章节的内容介绍:

序言

回顾了Subversion的历史,描述了Subversion的特性、架构、组件。

第 1 章 基本概念

介绍了版本控制的基础知识及不同的版本模型,同时讲述了Subversion版本库,工作拷贝和修订版本的概念。

第 2 章 基本使用

引领你开始一个Subversion用户的工作。示范怎样使用Subversion获得、修改和提交数据。

第 3 章 高级主题

覆盖了许多普通用户最终要面对的复杂特性,例如版本化的元数据、文件锁定和peg修订版本。

第 4 章 分支与合并

讨论分支、合并与标签,包括最佳实践的介绍,常见用例的描述,怎样取消修改,以及怎样从一个分支转到另一个分支。

第 5 章 版本库管理

讲述Subversion版本库的基本概念,怎样建立、配置和维护版本库,以及哪些工具可以完成上述的工作。

第 6 章 服务配置

Explains how to configure your Subversion server and offers different ways to access your repository: HTTP, the svn protocol, and local disk access. It also covers the details of authentication, authorization and anonymous access.

第 7 章 定制你的Subversion体验

研究了Subversion的客户端配置文件,对国际化字符的处理,以及Subversion如何与外置工具交互。

第 8 章 嵌入Subversion

介绍了Subversion的核心部件、Subversion的文件系统,以及程序员眼中的工作拷贝管理区域,展示了如何使用公共API编写Subversion应用程序。最重要的内容是,如何为Subversion的开发贡献力量。

第 9 章 Subversion 完全参考

以大量的实例,详细描述了svnsvnadminsvnlook的所有子命令。

附录 A, Subversion 快速入门指南

因为缺乏耐心,我们会立刻解释如何安装和使用Subversion,我们已经告诉你了。

附录 B, CVS用户的Subversion指南

Covers the similarities and differences between Subversion and CVS, with numerous suggestions on how to break all the bad habits you picked up from years of using CVS. Included are descriptions of Subversion revision numbers, versioned directories, offline operations, update versus status, branches, tags, metadata, conflict resolution, and authentication.

附录 C, WebDAV和自动版本

Describes the details of WebDAV and DeltaV and how you can configure your Subversion repository to be mounted read/write as a DAV share.

附录 D, Copyright

A copy of the Creative Commons Attribution License., under which this book is licensed.

This Book Is Free

This book started out as bits of documentation written by Subversion project developers, which were then coalesced into a single work and rewritten. As such, it has always been under a free license (see 附录 D, Copyright). In fact, the book was written in the public eye, originally as a part of Subversion project itself. This means two things:

  • 总可以在Subversion的版本库里找到本书的最新版本。

  • 可以任意分发或修改本书—它在免费许可证的控制之下,你的唯一限制是必须保留正确的最初作者。当然,与其独自发布私有版本,不如向Subversion开发社区提供反馈和修正信息。

本书的在线主页在http://svnbook.red-bean.com,有许多志愿的翻译工作。在网站上,你可以找到许多本书最新快照和标签版本的链接,也可以访问到本书的Subversion版本库(存放了DocBook XML源文件)。我们欢迎反馈—也愿意接受鼓励。请将所有的评论、抱怨和对本书源文件的补丁发送到。本书的中文版主要是由Subversion中文站的志愿者翻译的,可以在http://www.subversion.org.cn/看到本书的最新版本和其他资料,也要感谢i18n-zh的朋友的一些支持。

致谢

This book would not be possible (nor very useful) if Subversion did not exist. For that, the authors would like to thank Brian Behlendorf and CollabNet for the vision to fund such a risky and ambitious new open source project; Jim Blandy for the original Subversion name and design—we love you, Jim; and Karl Fogel for being such a good friend and a great community leader, in that order. [1]

Thanks to O'Reilly and our editors, Linda Mui and Tatiana Apandi, for their patience and support.

Finally, we thank the countless people who contributed to this book with informal reviews, suggestions, and fixes. While this is undoubtedly not a complete list, this book would be incomplete and incorrect without the help of: David Anderson, Jani Averbach, Ryan Barrett, Francois Beausoleil, Jennifer Bevan, Matt Blais, Zack Brown, Martin Buchholz, Brane Cibej, John R. Daily, Peter Davis, Olivier Davy, Robert P. J. Day, Mo DeJong, Brian Denny, Joe Drew, Nick Duffek, Ben Elliston, Justin Erenkrantz, Shlomi Fish, Julian Foad, Chris Foote, Martin Furter, Vlad Georgescu, Dave Gilbert, Eric Gillespie, David Glasser, Matthew Gregan, Art Haas, Eric Hanchrow, Greg Hudson, Alexis Huxley, Jens B. Jorgensen, Tez Kamihira, David Kimdon, Mark Benedetto King, Andreas J. Koenig, Nuutti Kotivuori, Matt Kraai, Scott Lamb, Vincent Lefevre, Morten Ludvigsen, Paul Lussier, Bruce A. Mah, Philip Martin, Feliciano Matias, Patrick Mayweg, Gareth McCaughan, Jon Middleton, Tim Moloney, Christopher Ness, Mats Nilsson, Joe Orton, Amy Lyn Pilato, Kevin Pilch-Bisson, Dmitriy Popkov, Michael Price, Mark Proctor, Steffen Prohaska, Daniel Rall, Jack Repenning, Tobias Ringstrom, Garrett Rooney, Joel Rosdahl, Christian Sauer, Larry Shatzer, Russell Steicke, Sander Striker, Erik Sjoelund, Johan Sundstroem, John Szakmeister, Mason Thomas, Eric Wadsworth, Colin Watson, Alex Waugh, Chad Whitacre, Josef Wolf, Blair Zajac, and the entire Subversion community.

来自 Ben Collins-Sussman

Thanks to my wife Frances, who, for many months, got to hear, “But honey, I'm still working on the book,” rather than the usual, “But honey, I'm still doing email.” I don't know where she gets all that patience! She's my perfect counterbalance.

感谢我的家人对我的鼓励,无论他们是否真的对我的课题感兴趣。(你知道的,一个人说 “哇,你正在写一本书?”,然后当他知道你是写一本计算机书时,那种惊讶就变得没有那么多了。)

感谢我身边让我富有的朋友,不要那样看我—你们知道你们是谁。

Thanks to my parents for the perfect low-level formatting and being unbelievable role models. Thanks to my kids for the opportunity to pass that on.

来自 Brian W. Fitzpatrick

Huge thanks to my wife Marie for being incredibly understanding, supportive, and most of all, patient. Thank you to my brother Eric who first introduced me to Unix programming way back when. Thanks to my Mom and Grandmother for all their support, not to mention enduring a Christmas holiday where I came home and promptly buried my head in my laptop to work on the book.

Mike和Ben:与你们一起工作非常快乐,Heck,我们在一起工作很愉快!

感谢所有在Subversion和Apache软件基金会的人们给我机会与你们在一起,没有一天我不从你们那里学到知识。

最后,感谢我的祖父,他一直跟我说“自由等于责任”,我深信不疑。

来自 C. Michael Pilato

Special thanks to Amy, my best friend and wife of nearly ten incredible years, for her love and patient support, for putting up with the late nights, and for graciously enduring the version control processes I've imposed on her. Don't worry, Sweetheart—you'll be a TortoiseSVN wizard in no time!

Gavin, you can probably read half of the words in this book yourself now; sadly, it's the other half that provide the key concepts. But when you've finally gotten a handle on the written form of this crazy language we speak, I hope you're as proud of your Daddy as he is of you.

Aidan, what can I say? I'm sorry this book doesn't have any pictures or stories of locomotives. I still love you, son. (And I recommend the works of Rev. W. V. Awdry to fuel your current passion.)

妈妈和爸爸,感谢你们的支持和热情,岳父岳母,以同样的理由感谢你们,还要感谢你们难以置信的女儿。

Hats off to Shep Kendall, through whom the world of computers was first opened to me; Ben Collins-Sussman, my tour-guide through the open source world; Karl Fogel, you are my .emacs; Greg Stein, for oozing practical programming know-how; Brian Fitzpatrick, for sharing this writing experience with me. To the many folks from whom I am constantly picking up new knowledge—keep dropping it!

最后,对所有为我展现完美卓越创造力的人们—感谢。

What Is Subversion?

Subversion is a free/open source version control system. That is, Subversion manages files and directories, and the changes made to them, over time. This allows you to recover older versions of your data or examine the history of how your data changed. In this regard, many people think of a version control system as a sort of “time machine.

Subversion的版本库可以通过网络访问,从而使用户可以在不同的电脑上进行操作。从某种程度上来说,允许用户在各自的空间里修改和管理同一组数据可以促进团队协作。因为修改不再是单线进行,开发速度会更快。此外,由于所有的工作都已版本化,也就不必担心由于错误的更改而影响软件质量—如果出现不正确的更改,只要撤销那一次更改操作即可。

Some version control systems are also software configuration management (SCM) systems. These systems are specifically tailored to manage trees of source code and have many features that are specific to software development—such as natively understanding programming languages, or supplying tools for building software. Subversion, however, is not one of these systems. It is a general system that can be used to manage any collection of files. For you, those files might be source code—for others, anything from grocery shopping lists to digital video mixdowns and beyond.

Is Subversion the Right Tool?

If you're a user or system administrator pondering the use of Subversion, the first question you should ask yourself is: "Is this the right tool for the job?" Subversion is a fantastic hammer, but be careful not to view every problem as a nail.

If you need to archive old versions of files and directories, possibly resurrect them, or examine logs of how they've changed over time, then Subversion is exactly the right tool for you. If you need to collaborate with people on documents (usually over a network) and keep track of who made which changes, then Subversion is also appropriate. This is why Subversion is so often used in software development environments— programming is an inherently social activity, and Subversion makes it easy to collaborate with other programmers. Of course, there's a cost to using Subversion as well: administrative overhead. You'll need to manage a data repository to store the information and all its history, and be diligent about backing it up. When working with the data on a daily basis, you won't be able to copy, move, rename, or delete files the way you usually do. Instead, you'll have to do all of those things through Subversion.

Assuming you're fine with the extra workflow, you should still make sure you're not using Subversion to solve a problem that other tools solve better. For example, because Subversion replicates data to all the collaborators involved, a common misuse is to treat it as a generic distribution system. People will sometimes use Subversion to distribute huge collections of photos, digital music, or software packages. The problem is that this sort of data usually isn't changing at all. The collection itself grows over time, but the individual files within the collection aren't being changed. In this case, using Subversion is “overkill.[2] There are simpler tools that efficiently replicate data without the overhead of tracking changes, such as rsync or unison.

Subversion的历史

In early 2000, CollabNet, Inc. (http://www.collab.net) began seeking developers to write a replacement for CVS. CollabNet offers a collaboration software suite called CollabNet Enterprise Edition (CEE), of which one component is version control. Although CEE used CVS as its initial version control system, CVS's limitations were obvious from the beginning, and CollabNet knew it would eventually have to find something better. Unfortunately, CVS had become the de facto standard in the open source world largely because there wasn't anything better, at least not under a free license. So CollabNet determined to write a new version control system from scratch, retaining the basic ideas of CVS, but without the bugs and misfeatures.

In February 2000, they contacted Karl Fogel, the author of Open Source Development with CVS (Coriolis, 1999), and asked if he'd like to work on this new project. Coincidentally, at the time Karl was already discussing a design for a new version control system with his friend Jim Blandy. In 1995, the two had started Cyclic Software, a company providing CVS support contracts, and although they later sold the business, they still used CVS every day at their jobs. Their frustration with CVS had led Jim to think carefully about better ways to manage versioned data, and he'd already come up with not only the name “Subversion,” but also with the basic design of the Subversion data store. When CollabNet called, Karl immediately agreed to work on the project, and Jim got his employer, Red Hat Software, to essentially donate him to the project for an indefinite period of time. CollabNet hired Karl and Ben Collins-Sussman, and detailed design work began in May 2000. With the help of some well-placed prods from Brian Behlendorf and Jason Robbins of CollabNet, and from Greg Stein (at the time an independent developer active in the WebDAV/DeltaV specification process), Subversion quickly attracted a community of active developers. It turned out that many people had encountered the same frustrating experiences with CVS and welcomed the chance to finally do something about it.

The original design team settled on some simple goals. They didn't want to break new ground in version control methodology, they just wanted to fix CVS. They decided that Subversion would match CVS's features and preserve the same development model, but not duplicate CVS's most obvious flaws. And although it did not need to be a drop-in replacement for CVS, it should be similar enough that any CVS user could make the switch with little effort.

After 14 months of coding, Subversion became “self-hosting” on August 31, 2001. That is, Subversion developers stopped using CVS to manage Subversion's own source code and started using Subversion instead.

While CollabNet started the project, and still funds a large chunk of the work (it pays the salaries of a few full-time Subversion developers), Subversion is run like most open source projects, governed by a loose, transparent set of rules that encourage meritocracy. CollabNet's copyright license is fully compliant with the Debian Free Software Guidelines. In other words, anyone is free to download, modify, and redistribute Subversion as he pleases; no permission from CollabNet or anyone else is required.

Subversion的特性

在讲解Subversion为版本控制领域带来的特性时,我们会经常通过Subversion对CVS的改进进行说明。如果不熟悉CVS,了解所有Subversion的特性会有一定的困难。而如果根本就不熟悉版本控制,你就只有干瞪眼的份儿了。因此,最好首先阅读一下第 1 章 基本概念,这一章简单介绍了一些版本控制的基本思想和概念。

Subversion支持:

版本化的目录

CVS tracks only the history of individual files, but Subversion implements a “virtual” versioned filesystem that tracks changes to whole directory trees over time. Files and directories are versioned.

真实的版本历史

由于只能跟踪单个文件的变更,CVS无法支持如文件拷贝和改名这些常见的操作—这些操作改变了目录的内容。同样,在CVS中,一个目录下的文件只要名字相同即拥有相同的历史,即使这些同名文件在历史上毫无关系。而在Subversion中,可以对文件或目录进行增加、拷贝和改名操作,也解决了同名而无关的文件之间的历史联系问题。

原子提交

A collection of modifications either goes into the repository completely or not at all. This allows developers to construct and commit changes as logical chunks and prevents problems that can occur when only a portion of a set of changes is successfully sent to the repository.

版本化的元数据

每一个文件和目录都有自己的一组属性—键和它们的值。可以根据需要建立并存储任何键/值对。和文件本身的内容一样,属性也在版本控制之下。

可选的网络层

Subversion has an abstracted notion of repository access, making it easy for people to implement new network mechanisms. Subversion can plug into the Apache HTTP Server as an extension module. This gives Subversion a big advantage in stability and interoperability, and instant access to existing features provided by that server—authentication, authorization, wire compression, and so on. A more lightweight, standalone Subversion server process is also available. This server speaks a custom protocol that can be easily tunneled via SSH.

一致的数据操作

Subversion用一个二进制差异算法描述文件的变化,对于文本(可读)和二进制(不可读)文件其操作方式是一致的。这两种类型的文件压缩存储在版本库中,而差异信息则在网络上双向传递。

高效的分支和标签操作

The cost of branching and tagging need not be proportional to the project size. Subversion creates branches and tags by simply copying the project, using a mechanism similar to a hard link. Thus these operations take only a very small, constant amount of time.

可修改性

Subversion没有历史负担,它以一系列优质的共享C程序库的方式实现,具有定义良好的API。这使得Subversion非常容易维护,和其它语言的互操作性很强。

Subversion的架构

图 1 “Subversion's architecture”给出了Subversion设计总体上的“俯视图”。

图 1. Subversion's architecture


图中的一端是保存所有版本数据的Subversion版本库,另一端是Subvesion的客户程序,管理着所有版本数据的本地影射(称为“工作拷贝”),在这两极之间是各种各样的版本库访问(RA)层,某些使用电脑网络通过网络服务器访问版本库,某些则绕过网络服务器直接访问版本库。

Subversion的组件

安装好的Subversion由几个部分组成,下面将简单的介绍一下这些组件。下文的描述或许过于简略,不易理解,但不用担心—本书后面的章节中会用更多的内容来详细阐述这些组件。

svn

命令行客户端程序。

svnversion

此工具用来显示工作拷贝的状态(用术语来说,就是当前项目的修订版本)。

svnlook

直接查看Subversion版本库的工具。

svnadmin

A tool for creating, tweaking, or repairing a Subversion repository.

svndumpfilter

过滤Subversion版本库转储数据流的工具。

mod_dav_svn

Apache HTTP服务器的一个插件,使版本库可以通过网络访问。

svnserve

一个单独运行的服务器程序,可以作为守护进程或由SSH调用。这是另一种使版本库可以通过网络访问的方式。

svnsync

一个通过网络增量镜像版本库的程序。

What's New in Subversion

The first edition of this book was released in 2004, shortly after Subversion had reached 1.0. Over the following four years Subversion released five major new versions, fixing bugs and adding major new features. While we've managed to keep the online version of this book up to date, we're thrilled that the second edition from O'Reilly now covers Subversion up through release 1.5, a major milestone for the project. Here's a quick summary of major new changes since Subversion 1.0. Note that this is not a complete list; for full details, please visit Subversion's web site at http://subversion.tigris.org.

Subversion 1.1 (September 2004)

Release 1.1 introduced FSFS, a flat-file repository storage option for the repository. While the BerkeleyDB back-end is still widely used and supported, FSFS has since become the “default” choice for newly-created repositories due to its low barrier to entry and minimal maintenance requirements. Also in this release came the ability to put symbolic links under version control, auto-escaping of URLs, and a localized user interface.

Subversion 1.2 (May 2005)

Release 1.2 introduced the ability to create server-side “locks” on files, thus serializing commit access to certain resources. While Subversion is still a fundamentally concurrent version control system, certain types of binary files (art assets, for example) cannot be merged together. The locking feature fulfills the need to version and protect such resources. With locking also came a complete “auto-versioning” implementation, allowing Subversion repositories to be mounted as network folders. Finally, Subversion 1.2 began using a new, faster binary-differencing algorithm to compress and retrieve old versions of files.

Subversion 1.3 (December 2005)

Release 1.3 brought path-based authorization controls to the svnserve server, matching a feature formerly found only in the Apache server. The Apache server, however, gained some new logging features of its own, and Subversion's API bindings to other languages also made great leaps forward.

Subversion 1.4 (September 2006)

Release 1.4 introduced a whole new tool—svnsync— for doing one-way repository replication over a network. Major parts of the working copy metadata were revamped to no longer use XML (resulting in client side speed gains), while the BerkeleyDB repository back-end gained the ability to automatically “recover” itself after a server crash.

Subversion 1.5 (June 2008)

Release 1.5 took much longer to finish than prior releases, but the headliner feature was gigantic: semi-automated tracking of branching and merging. This was a huge boon for users, and pushed Subversion far beyond the abilities of CVS and into the ranks of commercial competitors such as Perforce and Clearcase. Subversion 1.5 also introduced a bevy of other user-focused features, such as interactive resolution of file conflicts, partial checkouts, client side management of changelists, powerful new syntax for the svn:externals feature, and SASL authentication support for the svnserve server.



[1] 噢,还要感谢Karl为了本书所付出的辛勤工作。

[2] Or as a friend puts it, “swatting a fly with a Buick.

基本概念

本章主要为那些不熟悉版本控制技术的入门者提供一个简单扼要的、非系统的介绍。我们将从版本控制的基本概念开始,随后阐述Subversion的独特理念,并演示一些使用Subversion的例子。

虽然我们在本章中以分享程序源代码作为例子,但是记住Subversion可以管理任何类型的文件集—它并非是程序员专用的。

版本库

Subversion是一个“集中式”的信息共享系统。版本库是Subversion的核心部分,是数据的中央仓库。版本库以典型的文件和目录结构形式文件系统树来保存信息。任意数量的客户端连接到Subversion版本库,读取、修改这些文件。客户端通过写数据将信息分享给其他人,通过读取数据获取别人共享的信息。图 1.1 “一个典型的客户/服务器系统”展示了这种系统:

图 1.1. 一个典型的客户/服务器系统


So why is this interesting? So far, this sounds like the definition of a typical file server. And indeed, the repository is a kind of file server, but it's not your usual breed. What makes the Subversion repository special is that it remembers every change ever written to it—every change to every file, and even changes to the directory tree itself, such as the addition, deletion, and rearrangement of files and directories.

When a client reads data from the repository, it normally sees only the latest version of the filesystem tree. But the client also has the ability to view previous states of the filesystem. For example, a client can ask historical questions such as “What did this directory contain last Wednesday?” or “Who was the last person to change this file, and what changes did he make?” These are the sorts of questions that are at the heart of any version control system: systems that are designed to track changes to data over time.

版本模型

版本控制系统的核心任务是实现协作编辑和数据共享,但是不同的系统使用不同的策略实现这个目的。我们有许多理由要去理解这些策略的区别,首先,如果你遇到了其他类似Subversion的系统,可以帮助你比较现有的版本控制系统。此外,可以帮助你更有效的使用Subversion,因为Subversion本身支持不同的工作方式。

The Problem of File Sharing

所有的版本控制系统都需要解决这样一个基础问题:怎样让系统允许用户共享信息,而不会让他们因意外而互相干扰?版本库里意外覆盖别人的更改非常的容易。

Consider the scenario shown in 图 1.2 “需要避免的问题”. Suppose we have two coworkers, Harry and Sally. They each decide to edit the same repository file at the same time. If Harry saves his changes to the repository first, then it's possible that (a few moments later) Sally could accidentally overwrite them with her own new version of the file. While Harry's version of the file won't be lost forever (because the system remembers every change), any changes Harry made won't be present in Sally's newer version of the file, because she never saw Harry's changes to begin with. Harry's work is still effectively lost—or at least missing from the latest version of the file—and probably by accident. This is definitely a situation we want to avoid!

图 1.2. 需要避免的问题


锁定-修改-解锁 方案

Many version control systems use a lock-modify-unlock model to address the problem of many authors clobbering each other's work. In this model, the repository allows only one person to change a file at a time. This exclusivity policy is managed using locks. Harry must “lock” a file before he can begin making changes to it. If Harry has locked a file, then Sally cannot also lock it, and therefore cannot make any changes to that file. All she can do is read the file and wait for Harry to finish his changes and release his lock. After Harry unlocks the file, Sally can take her turn by locking and editing the file. 图 1.3 “锁定-修改-解锁 方案” demonstrates this simple solution.

图 1.3. 锁定-修改-解锁 方案


The problem with the lock-modify-unlock model is that it's a bit restrictive and often becomes a roadblock for users:

  • 锁定可能导致管理问题。有时候Harry会锁住文件然后忘了此事,这就是说Sally一直等待解锁来编辑这些文件,她在这里僵住了。然后Harry去旅行了,现在Sally只好去找管理员放开锁,这种情况会导致不必要的耽搁和时间浪费。

  • 锁定可能导致不必要的线性化开发。如果Harry编辑一个文件的开始,Sally想编辑同一个文件的结尾,这种修改不会冲突,设想修改可以正确的合并到一起,他们可以轻松的并行工作而没有太多的坏处,没有必要让他们轮流工作。

  • Locking may create a false sense of security. Suppose Harry locks and edits file A, while Sally simultaneously locks and edits file B. But what if A and B depend on one another, and the changes made to each are semantically incompatible? Suddenly A and B don't work together anymore. The locking system was powerless to prevent the problem—yet it somehow provided a false sense of security. It's easy for Harry and Sally to imagine that by locking files, each is beginning a safe, insulated task, and thus they need not bother discussing their incompatible changes early on. Locking often becomes a substitute for real communication.

拷贝-修改-合并 方案

Subversion, CVS, and many other version control systems use a copy-modify-merge model as an alternative to locking. In this model, each user's client contacts the project repository and creates a personal working copy—a local reflection of the repository's files and directories. Users then work simultaneously and independently, modifying their private copies. Finally, the private copies are merged together into a new, final version. The version control system often assists with the merging, but ultimately, a human being is responsible for making it happen correctly.

Here's an example. Say that Harry and Sally each create working copies of the same project, copied from the repository. They work concurrently and make changes to the same file A within their copies. Sally saves her changes to the repository first. When Harry attempts to save his changes later, the repository informs him that his file A is out-of-date. In other words, that file A in the repository has somehow changed since he last copied it. So Harry asks his client to merge any new changes from the repository into his working copy of file A. Chances are that Sally's changes don't overlap with his own; once he has both sets of changes integrated, he saves his working copy back to the repository. 图 1.4 “拷贝-修改-合并 方案” and 图 1.5 “拷贝-修改-合并 方案(续)” show this process.

图 1.4. 拷贝-修改-合并 方案


图 1.5. 拷贝-修改-合并 方案(续)


But what if Sally's changes do overlap with Harry's changes? What then? This situation is called a conflict, and it's usually not much of a problem. When Harry asks his client to merge the latest repository changes into his working copy, his copy of file A is somehow flagged as being in a state of conflict: he'll be able to see both sets of conflicting changes and manually choose between them. Note that software can't automatically resolve conflicts; only humans are capable of understanding and making the necessary intelligent choices. Once Harry has manually resolved the overlapping changes—perhaps after a discussion with Sally—he can safely save the merged file back to the repository.

拷贝-修改-合并模型感觉有一点混乱,但在实践中,通常运行的很平稳,用户可以并行的工作,不必等待别人,当工作在同一个文件上时,也很少会有交迭发生,冲突并不频繁,处理冲突的时间远比等待解锁花费的时间少。

最后,一切都要归结到一条重要的因素:用户交流。当用户交流贫乏,语法和语义的冲突就会增加,没有系统可以强制用户完美的交流,没有系统可以检测语义上的冲突,所以没有任何证据能够承诺锁定系统可以防止冲突,实践中,锁定除了约束了生产力,并没有做什么事。

Subversion实践

是时候从抽象转到具体了,在本小节,我们会展示一个Subversion真实使用的例子。

Subversion版本库URL

正如我们在整本书里描述的,Subversion使用URL来识别Subversion版本库中的版本化资源,通常情况下,这些URL使用标准的语法,允许服务器名称和端口作为URL的一部分:

$ svn checkout http://svn.example.com:9834/repos
…

但是Subversion处理URL的一些细微的不同之处需要注意,例如,使用file:访问方法的URL(用来访问本地版本库)必须与习惯一致,可以包括一个localhost服务器名或者没有服务器名:

$ svn checkout file:///var/svn/repos
…
$ svn checkout file://localhost/var/svn/repos
…

Also, users of the file:// scheme on Windows platforms will need to use an unofficially “standard” syntax for accessing repositories that are on the same machine, but on a different drive than the client's current working drive. Either of the two following URL path syntaxes will work, where X is the drive on which the repository resides:

C:\> svn checkout file:///X:/var/svn/repos
…
C:\> svn checkout "file:///X|/var/svn/repos"
…

在第二个语法里,你需要使用引号包含整个URL,这样竖线字符才不会被解释为管道。当然,也要注意URL使用普通的斜线而不是Windows本地(不是URL)的反斜线。

注意

也必须意识到Subversion的file: URL不能在普通的web服务器中工作。当你尝试在web服务器查看一个file:的URL时,它会通过直接检测文件系统读取和显示那个位置的文件内容,但是Subversion的资源存在于虚拟文件系统(见“版本库层”一节)中,你的浏览器不会理解怎样读取这个文件系统。

Finally, it should be noted that the Subversion client will automatically encode URLs as necessary, just like a web browser does. For example, if a URL contains a space or upper-ASCII character as in the following:

$ svn checkout "http://host/path with space/project/españa"

then Subversion will escape the unsafe characters and behave as if you had typed:

$ svn checkout http://host/path%20with%20space/project/espa%C3%B1a

如果URL包含空格,一定要使用引号,这样你的脚本才会把它做一个单独的svn参数。

工作拷贝

你已经阅读过了关于工作拷贝的内容;现在我们要讲一讲客户端怎样建立和使用它。

一个Subversion工作拷贝是你本地机器上的一个普通目录,保存着一些文件,你可以任意的编辑文件,而且如果是源代码文件,你可以像平常一样编译,你的工作拷贝是你的私有工作区,在你明确的做了特定操作之前,Subversion不会把你的修改与其他人的合并,也不会把你的修改展示给别人,你甚至可以拥有同一个项目的多个工作拷贝。

After you've made some changes to the files in your working copy and verified that they work properly, Subversion provides you with commands to “publish” your changes to the other people working with you on your project (by writing to the repository). If other people publish their own changes, Subversion provides you with commands to merge those changes into your working copy (by reading from the repository).

A working copy also contains some extra files, created and maintained by Subversion, to help it carry out these commands. In particular, each directory in your working copy contains a subdirectory named .svn, also known as the working copy's administrative directory. The files in each administrative directory help Subversion recognize which files contain unpublished changes, and which files are out of date with respect to others' work.

一个典型的Subversion的版本库经常包含许多项目的文件(或者说源代码),通常每一个项目都是版本库的子目录,在这种布局下,一个用户的工作拷贝往往对应版本库的的一个子目录。

举一个例子,你的版本库包含两个软件项目,paintcalc。每个项目在它们各自的顶级子目录下,见图 1.6 “版本库的文件系统”

图 1.6. 版本库的文件系统


To get a working copy, you must check out some subtree of the repository. (The term check out may sound like it has something to do with locking or reserving resources, but it doesn't; it simply creates a private copy of the project for you.) For example, if you check out /calc, you will get a working copy like this:

$ svn checkout http://svn.example.com/repos/calc
A    calc/Makefile
A    calc/integer.c
A    calc/button.c
Checked out revision 56.

$ ls -A calc
Makefile  button.c integer.c .svn/

The list of letter As in the left margin indicates that Subversion is adding a number of items to your working copy. You now have a personal copy of the repository's /calc directory, with one additional entry—.svn—which holds the extra information needed by Subversion, as mentioned earlier.

假定你修改了button.c,因为.svn目录记录着文件的修改日期和原始内容,Subversion可以告诉你已经修改了文件,然而,在你明确告诉它之前,Subversion不会将你的改变公开,将改变公开的操作被叫做提交(committing,或者是checking in)修改到版本库。

将你的修改发布给别人,你可以使用Subversion的提交(commit)命令。

$ svn commit button.c -m "Fixed a typo in button.c."
Sending        button.c
Transmitting file data .
Committed revision 57.

这时你对button.c的修改已经提交到了版本库,其中包含了关于此次提交的日志信息(例如是修改了拼写错误)。如果其他人取出了/calc的一个工作拷贝,他们会看到这个文件最新的版本。

Suppose you have a collaborator, Sally, who checked out a working copy of /calc at the same time you did. When you commit your change to button.c, Sally's working copy is left unchanged; Subversion modifies working copies only at the user's request.

To bring her project up to date, Sally can ask Subversion to update her working copy, by using the update command. This will incorporate your changes into her working copy, as well as any others that have been committed since she checked it out.

$ pwd
/home/sally/calc

$ ls -A
Makefile button.c integer.c .svn/

$ svn update
U    button.c
Updated to revision 57.

The output from the svn update command indicates that Subversion updated the contents of button.c. Note that Sally didn't need to specify which files to update; Subversion uses the information in the .svn directory as well as further information in the repository, to decide which files need to be brought up to date.

修订版本

An svn commit operation publishes changes to any number of files and directories as a single atomic transaction. In your working copy, you can change files' contents; create, delete, rename, and copy files and directories; then commit a complete set of changes as an atomic transaction.

By atomic transaction, we mean simply this: either all of the changes happen in the repository, or none of them happen. Subversion tries to retain this atomicity in the face of program crashes, system crashes, network problems, and other users' actions.

Each time the repository accepts a commit, this creates a new state of the filesystem tree, called a revision. Each revision is assigned a unique natural number, one greater than the number of the previous revision. The initial revision of a freshly created repository is numbered 0 and consists of nothing but an empty root directory.

图 1.7 “版本库”可以更形象的描述版本库,想象有一组修订号,从0开始,从左到右,每一个修订号有一个目录树挂在它下面,每一个树好像是一次提交后的版本库“快照”。

图 1.7. 版本库


需要特别注意的是,工作拷贝并不一定对应版本库中的单个修订版本,他们可能包含多个修订版本的文件。举个例子,你从版本库检出一个工作拷贝,最近的修订号是4:

calc/Makefile:4
     integer.c:4
     button.c:4

此刻,工作目录与版本库的修订版本4完全对应,然而,你修改了button.c并且提交之后,假设没有别的提交出现,你的提交会在版本库建立修订版本5,你的工作拷贝会是这个样子的:

calc/Makefile:4
     integer.c:4
     button.c:5

假设此刻,Sally提交了对integer.c的修改,建立修订版本6,如果你使用svn update来更新你的工作拷贝,你会看到:

calc/Makefile:6
     integer.c:6
     button.c:6

Sally对integer.c的改变会出现在你的工作拷贝,你对button.c的改变还在,在这个例子里,Makefile在4、5、6修订版本都是一样的,但是Subversion会把他的Makefile的修订号设为6来表明它是最新的,所以你在工作拷贝顶级目录作一次干净的更新,会使得所有内容对应版本库的同一修订版本。

工作拷贝怎样跟踪版本库

对于工作拷贝的每一个文件,Subversion在管理区域.svn/记录两项关键的信息:

  • What revision your working file is based on (this is called the file's working revision) and

  • A timestamp recording of when the local copy was last updated by the repository

给定这些信息,通过与版本库通讯,Subversion可以告诉我们工作文件是处于如下四种状态的那一种:

未修改且是当前的

文件在工作目录里没有修改,在工作修订版本之后没有修改提交到版本库。svn commit操作不做任何事情,svn update不做任何事情。

本地已修改且是当前的

在工作目录已经修改,从基本修订版本之后没有修改提交到版本库。本地修改没有提交,因此svn commit会成功提交,svn update不做任何事情。

未修改且不是当前的了

The file has not been changed in the working directory, but it has been changed in the repository. The file should eventually be updated in order to make it current with the latest public revision. An svn commit of the file will do nothing, and an svn update of the file will fold the latest changes into your working copy.

本地已修改且不是最新的

The file has been changed both in the working directory and in the repository. An svn commit of the file will fail with an “out-of-date” error. The file should be updated first; an svn update command will attempt to merge the public changes with the local changes. If Subversion can't complete the merge in a plausible way automatically, it leaves it to the user to resolve the conflict.

这看起来需要记录很多事情,但是svn status命令可以告诉你工作拷贝中文件的状态,关于此命令更多的信息,请看“查看你的修改概况”一节

混合修订版本的工作拷贝

作为一个普遍原理,Subversion努力做到尽可能的灵活,一个特殊的灵活特性就是让工作拷贝包含不同工作修订版本的文件和目录,不幸的是,这个灵活性会让许多新用户感到迷惑。如果上一个混合修订版本的例子让你感到困惑,这里是一个为何有这种特性和如何利用这个特性的基础介绍。

Updates and commits are separate

One of the fundamental rules of Subversion is that a “push” action does not cause a “pull,” nor the other way around. Just because you're ready to submit new changes to the repository doesn't mean you're ready to receive changes from other people. And if you have new changes still in progress, then svn update should gracefully merge repository changes into your own, rather than forcing you to publish them.

The main side effect of this rule is that it means a working copy has to do extra bookkeeping to track mixed revisions as well as be tolerant of the mixture. It's made more complicated by the fact that directories themselves are versioned.

For example, suppose you have a working copy entirely at revision 10. You edit the file foo.html and then perform an svn commit, which creates revision 15 in the repository. After the commit succeeds, many new users would expect the working copy to be entirely at revision 15, but that's not the case! Any number of changes might have happened in the repository between revisions 10 and 15. The client knows nothing of those changes in the repository, since you haven't yet run svn update, and svn commit doesn't pull down new changes. If, on the other hand, svn commit were to automatically download the newest changes, then it would be possible to set the entire working copy to revision 15—but then we'd be breaking the fundamental rule of “push” and “pull” remaining separate actions. Therefore, the only safe thing the Subversion client can do is mark the one file—foo.html—as being at revision 15. The rest of the working copy remains at revision 10. Only by running svn update can the latest changes be downloaded and the whole working copy be marked as revision 15.

混合修订版本很常见

The fact is, every time you run svn commit, your working copy ends up with some mixture of revisions. The things you just committed are marked as having larger working revisions than everything else. After several commits (with no updates in between), your working copy will contain a whole mixture of revisions. Even if you're the only person using the repository, you will still see this phenomenon. To examine your mixture of working revisions, use the svn status --verbose command (see “查看你的修改概况”一节 for more information.)

通常,新用户对于工作拷贝的混合修订版本一无所知,这会让人糊涂,因为许多客户端命令对于所检验条目的修订版本很敏感。例如svn log命令显示一个文件或目录的历史修改信息(见“Generating a List of Historical Changes”一节),当用户对一个工作拷贝对象调用这个命令,他们希望看到这个对象的整个历史信息。但是如果这个对象的修订版本已经相当老了(通常因为很长时间没有运行svn update),此时会显示比这个对象更老的历史。

混合版本很有用

If your project is sufficiently complex, you'll discover that it's sometimes nice to forcibly backdate (or, update to a revision older than the one you already have) portions of your working copy to an earlier revision; you'll learn how to do that in 第 2 章 基本使用. Perhaps you'd like to test an earlier version of a submodule contained in a subdirectory, or perhaps you'd like to figure out when a bug first came into existence in a specific file. This is the “time machine” aspect of a version control system—the feature that allows you to move any portion of your working copy forward and backward in history.

混合版本有限制

无论你如何在工作拷贝中利用混合修订版本,这种灵活性还是有限制的。

First, you cannot commit the deletion of a file or directory that isn't fully up to date. If a newer version of the item exists in the repository, your attempt to delete will be rejected in order to prevent you from accidentally destroying changes you've not yet seen.

Second, you cannot commit a metadata change to a directory unless it's fully up to date. You'll learn about attaching “properties” to items in 第 3 章 高级主题. A directory's working revision defines a specific set of entries and properties, and thus committing a property change to an out-of-date directory may destroy properties you've not yet seen.

总结

我们在这一章里学习了许多Subversion的基本概念:

  • 我们介绍了中央版本库、客户工作拷贝和版本修订树的概念。

  • 我们介绍了两个协作者如何使用Subversion通过“拷贝-修改-合并”模型发布和获得对方的修改。

  • 我们讨论了一些Subversion跟踪和管理工作拷贝信息的方式。

现在,你一定对Subversion在多数情形下的工作方式有了很好的认识,有了这些知识的武装,你一定已经准备好跳到下一章去了,一个关于Subversion命令与特性的详细教程。

基本使用

现在,我们将要深入到Subversion的使用细节当中,完成本章时,你将学会所有Subversion日常使用的命令,你将从把数据导入到Subversion开始,接着是初始化的检出(check out),然后是做出修改并检查,你也将会学到如何在工作拷贝中获取别人的修改,检查他们,并解决所有可能发生的冲突。

Note that this chapter is not meant to be an exhaustive list of all Subversion's commands—rather, it's a conversational introduction to the most common Subversion tasks that you'll encounter. This chapter assumes that you've read and understood 第 1 章 基本概念 and are familiar with the general model of Subversion. For a complete reference of all commands, see 第 9 章 Subversion 完全参考.

求助!

在继续阅读之前,需要知道Subversion使用中最重要的命令:svn help,Subversion命令行工具是一个自文档的工具—在任何时候你可以运行svn help SUBCOMMAND来查看子命令的语法、参数以及行为方式。

$ svn help import
import: Commit an unversioned file or tree into the repository.
usage: import [PATH] URL

  Recursively commit a copy of PATH to URL.
  If PATH is omitted '.' is assumed.
  Parent directories are created as necessary in the repository.
  If PATH is a directory, the contents of the directory are added
  directly under URL.
  Unversionable items such as device files and pipes are ignored
  if --force is specified.

Valid options:
  -q [--quiet]             : print nothing, or only summary information
  -N [--non-recursive]     : obsolete; try --depth=files or --depth=immediates
  --depth ARG              : limit operation by depth ARG ('empty', 'files',
                             'immediates', or 'infinity')
…

Getting Data into Your Repository

There are two ways to get new files into your Subversion repository: svn import and svn add. We'll discuss svn import now and will discuss svn add later in this chapter when we review a typical day with Subversion.

svn import

svn import是将未版本化文件导入版本库的最快方法,会根据需要创建中介目录。svn import不需要一个工作拷贝,你的文件会直接提交到版本库,这通常用在你希望将一组文件加入到Subversion版本库时,例如:

$ svnadmin create /var/svn/newrepos
$ svn import mytree file:///var/svn/newrepos/some/project \
             -m "Initial import"
Adding         mytree/foo.c
Adding         mytree/bar.c
Adding         mytree/subdir
Adding         mytree/subdir/quux.h

Committed revision 1.

在上一个例子里,将会拷贝目录mytree到版本库的some/project下:

$ svn list file:///var/svn/newrepos/some/project
bar.c
foo.c
subdir/

注意,在导入之后,原来的目录树并没有转化成工作拷贝,为了开始工作,你还是需要运行svn checkout导出一个工作拷贝。

Recommended Repository Layout

While Subversion's flexibility allows you to lay out your repository in any way that you choose, we recommend that you create a trunk directory to hold the “main line” of development, a branches directory to contain branch copies, and a tags directory to contain tag copies—for example:

$ svn list file:///var/svn/repos
/trunk
/branches
/tags

You'll learn more about tags and branches in 第 4 章 分支与合并. For details and how to set up multiple projects, see “版本库布局”一节 and “规划你的版本库结构”一节 to read more about project roots.

初始化检出

大多数时候,你会使用checkout从版本库取出一个新拷贝开始使用Subversion,这样会在本机创建一个项目的“本地拷贝”,这个拷贝包括了命令行指定版本库中的HEAD(最新的)版本:

$ svn checkout http://svn.collab.net/repos/svn/trunk
A    trunk/Makefile.in
A    trunk/ac-helpers
A    trunk/ac-helpers/install.sh
A    trunk/ac-helpers/install-sh
A    trunk/build.conf
…
Checked out revision 8810.

尽管上面的例子取出了trunk目录,你也完全可以通过输入特定URL取出任意深度的子目录:

$ svn checkout \
      http://svn.collab.net/repos/svn/trunk/subversion/tests/cmdline/
A    cmdline/revert_tests.py
A    cmdline/diff_tests.py
A    cmdline/autoprop_tests.py
A    cmdline/xmltests
A    cmdline/xmltests/svn-test.sh
…
Checked out revision 8810.

Since Subversion uses a “copy-modify-merge” model instead of “lock-modify-unlock” (see “版本模型”一节), you can start right in making changes to the files and directories in your working copy. Your working copy is just like any other collection of files and directories on your system. You can edit and change them, move them around, even delete the entire working copy and forget about it.

警告

While your working copy is “just like any other collection of files and directories on your system,” you can edit files at will, but you must tell Subversion about everything else that you do. For example, if you want to copy or move an item in a working copy, you should use svn copy or svn move instead of the copy and move commands provided by your operating system. We'll talk more about them later in this chapter.

Unless you're ready to commit the addition of a new file or directory or changes to existing ones, there's no need to further notify the Subversion server that you've done anything.

因为你可以使用版本库的URL作为唯一参数取出一个工作拷贝,你也可以在版本库URL之后指定一个目录,这样会将你的工作目录放到你的新目录,举个例子:

$  svn checkout http://svn.collab.net/repos/svn/trunk subv
A    subv/Makefile.in
A    subv/ac-helpers
A    subv/ac-helpers/install.sh
A    subv/ac-helpers/install-sh
A    subv/build.conf
…
Checked out revision 8810.

这样将把你的工作拷贝放到subv而不是和前面那样放到trunk,如果subv不存在,将会自动创建。

禁用密码缓存

When you perform a Subversion operation that requires you to authenticate, by default Subversion caches your authentication credentials on disk. This is done for convenience, so that you don't have to continually re-enter your password for future operations. If you're concerned about caching your Subversion passwords, [3] you can disable caching either permanently or on a case-by-case basis.

To disable password caching for a particular one-time command, pass the --no-auth-cache option on the command line. To permanently disable caching, you can add the line store-passwords = no to your local machine's Subversion configuration file. See “客户端凭证缓存”一节 for details.

用其它身份认证

Since Subversion caches auth credentials by default (both username and password), it conveniently remembers who you were acting as the last time you modified you working copy. But sometimes that's not helpful—particularly if you're working in a shared working copy such as a system configuration directory or a webserver document root. In this case, just pass the --username option on the command line, and Subversion will attempt to authenticate as that user, prompting you for a password if necessary.

基本的工作周期

Subversion has numerous features, options, bells, and whistles, but on a day-to-day basis, odds are that you will only use a few of them. In this section, we'll run through the most common things that you might find yourself doing with Subversion in the course of a day's work.

典型的工作周期是这样的:

  1. 更新你的工作拷贝。

    • svn update

  2. Make changes.

    • svn add

    • svn delete

    • svn copy

    • svn move

  3. Examine your changes.

    • svn status

    • svn diff

  4. Possibly undo some changes.

    • svn revert

  5. Resolve conflicts (merge others' changes).

    • svn update

    • svn resolve

  6. Commit your changes.

    • svn commit

更新你的工作拷贝

When working on a project with a team, you'll want to update your working copy to receive any changes made since your last update by other developers on the project. Use svn update to bring your working copy into sync with the latest revision in the repository:

$ svn update
U  foo.c
U  bar.c
Updated to revision 2.

In this case, it appears that someone checked in modifications to both foo.c and bar.c since the last time you updated, and Subversion has updated your working copy to include those changes.

When the server sends changes to your working copy via svn update, a letter code is displayed next to each item to let you know what actions Subversion performed to bring your working copy up-to-date. To find out what these letters mean, run svn help update.

修改你的工作拷贝

现在你可以开始工作并且修改你的工作拷贝了,你很容易决定作出一个修改(或者是一组),像写一个新的特性,修正一个错误等等。这时可以使用的Subversion命令包括svn addsvn deletesvn copysvn move。如果你只是修改版本库中已经存在的文件,在你提交之前,不必使用上面的任何一个命令。

There are two kinds of changes you can make to your working copy: file changes and tree changes. You don't need to tell Subversion that you intend to change a file; just make your changes using your text editor, word processor, graphics program, or whatever tool you would normally use. Subversion automatically detects which files have been changed, and in addition, handles binary files just as easily as it handles text files—and just as efficiently too. For tree changes, you can ask Subversion to “mark” files and directories for scheduled removal, addition, copying, or moving. These changes may take place immediately in your working copy, but no additions or removals will happen in the repository until you commit them.

下面是Subversion用来修改目录树结构的五个子命令。

svn add foo

Schedule file, directory, or symbolic link foo to be added to the repository. When you next commit, foo will become a child of its parent directory. Note that if foo is a directory, everything underneath foo will be scheduled for addition. If you want only to add foo itself, pass the --depth empty option.

svn delete foo

预定将文件、目录或者符号链foo从版本库中删除,如果foo是文件,它马上从工作拷贝中删除,如果是目录,不会被删除,但是Subversion准备好删除了,当你提交你的修改,foo就会在你的工作拷贝和版本库中被删除。[4]

svn copy foo bar

Create a new item bar as a duplicate of foo and automatically schedule bar for addition. When bar is added to the repository on the next commit, its copy history is recorded (as having originally come from foo). svn copy does not create intermediate directories unless you pass the --parents.

svn move foo bar

This command is exactly the same as running svn copy foo bar; svn delete foo. That is, bar is scheduled for addition as a copy of foo, and foo is scheduled for removal. svn move does not create intermediate directories unless you pass the --parents.

svn mkdir blort

这个命令同运行 mkdir blort; svn add blort相同,也就是创建一个叫做blort的文件,并且预定添加到版本库。

检查你的修改

当你完成修改,你需要提交他们到版本库,但是在此之前,检查一下做过什么修改是个好主意,通过提交前的检查,你可以整理一份精确的日志信息,你也可以发现你不小心修改的文件,给了你一次恢复修改的机会。此外,这是一个审查和仔细察看修改的好机会,你可通过命令svn status浏览所做的修改,通过svn diff检查修改的详细信息。

Subversion has been optimized to help you with this task, and it is able to do many things without communicating with the repository. In particular, your working copy contains a hidden cached “pristine” copy of each version controlled file within the .svn area. Because of this, Subversion can quickly show you how your working files have changed or even allow you to undo your changes without contacting the repository.

查看你的修改概况

为了浏览修改的内容,你会使用这个svn status命令,在所有Subversion命令里,svn status可能会是你用的最多的命令。

如果你在工作拷贝的顶级目录运行不带参数的svn status命令,它会检测你做的所有的文件或目录的修改,以下的例子是来展示svn status可能返回的状态码(注意,#之后的不是svn status打印的)。

?       scratch.c           # file is not under version control
A       stuff/loot/bloo.h   # file is scheduled for addition
C       stuff/loot/lump.c   # file has textual conflicts from an update
D       stuff/fish.c        # file is scheduled for deletion
M       bar.c               # the content in bar.c has local modifications

In this output format, svn status prints six columns of characters, followed by several whitespace characters, followed by a file or directory name. The first column tells the status of a file or directory and/or its contents. The codes we listed are:

A item

预定加入到版本库的文件、目录或符号链的item

C item

The file item is in a state of conflict. That is, changes received from the server during an update overlap with local changes that you have in your working copy (and weren't resolved during the update). You must resolve this conflict before committing your changes to the repository.

D item

文件、目录或是符号链item预定从版本库中删除。

M item

文件item的内容被修改了。

如果你传递一个路径给svn status,它只给你这个项目的信息:

$ svn status stuff/fish.c
D      stuff/fish.c

svn status also has a --verbose (-v) option, which will show you the status of every item in your working copy, even if it has not been changed:

$ svn status -v
M               44        23    sally     README
                44        30    sally     INSTALL
M               44        20    harry     bar.c
                44        18    ira       stuff
                44        35    harry     stuff/trout.c
D               44        19    ira       stuff/fish.c
                44        21    sally     stuff/things
A                0         ?     ?        stuff/things/bloo.h
                44        36    harry     stuff/things/gloo.c

This is the “long form” output of svn status. The letters in the first column mean the same as before, but the second column shows the working revision of the item. The third and fourth columns show the revision in which the item last changed, and who changed it.

None of the prior invocations to svn status contact the repository—instead, they compare the metadata in the .svn directory with the working copy. Finally, there is the --show-updates (-u) option, which contacts the repository and adds information about things that are out of date:

$ svn status -u -v
M      *        44        23    sally     README
M               44        20    harry     bar.c
       *        44        35    harry     stuff/trout.c
D               44        19    ira       stuff/fish.c
A                0         ?     ?        stuff/things/bloo.h
Status against revision:   46

Notice the two asterisks: if you were to run svn update at this point, you would receive changes to README and trout.c. This tells you some very useful information—you'll need to update and get the server changes on README before you commit, or the repository will reject your commit for being out of date (more on this subject later).

svn status can display much more information about the files and directories in your working copy than we've shown here—for an exhaustive description of svn status and its output, see svn status.

检查你的本地修改的详情

另一种检查修改的方式是svn diff命令,你可以通过不带参数的svn diff精确的找出你所做的修改,这会输出统一区别格式的区别信息:

$ svn diff
Index: bar.c
===================================================================
--- bar.c	(revision 3)
+++ bar.c	(working copy)
@@ -1,7 +1,12 @@
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#include <stdio.h>

 int main(void) {
-  printf("Sixty-four slices of American Cheese...\n");
+  printf("Sixty-five slices of American Cheese...\n");
 return 0;
 }

Index: README
===================================================================
--- README	(revision 3)
+++ README	(working copy)
@@ -193,3 +193,4 @@
+Note to self:  pick up laundry.

Index: stuff/fish.c
===================================================================
--- stuff/fish.c	(revision 1)
+++ stuff/fish.c	(working copy)
-Welcome to the file known as 'fish'.
-Information on fish will be here soon.

Index: stuff/things/bloo.h
===================================================================
--- stuff/things/bloo.h	(revision 8)
+++ stuff/things/bloo.h	(working copy)
+Here is a new file to describe
+things about bloo.

The svn diff command produces this output by comparing your working files against the cached “pristine” copies within the .svn area. Files scheduled for addition are displayed as all added text, and files scheduled for deletion are displayed as all deleted text.

Output is displayed in unified diff format. That is, removed lines are prefaced with -, and added lines are prefaced with +. svn diff also prints filename and offset information useful to the patch program, so you can generate “patches” by redirecting the diff output to a file:

$ svn diff > patchfile

举个例子,你可以把补丁文件发送邮件到其他开发者,在提交之前审核和测试。

Subversion uses its internal diff engine, which produces unified diff format, by default. If you want diff output in a different format, specify an external diff program using --diff-cmd and pass any flags you'd like to it using the --extensions (-x) option. For example, to see local differences in file foo.c in context output format while ignoring case differences, you might run svn diff --diff-cmd /usr/bin/diff --extensions '-i' foo.c.

取消本地修改

假定我们在看svn diff的输出,你发现对某个文件的所有修改都是错误的,或许你根本不应该修改这个文件,或者是从开头重新修改会更加容易。

这是使用svn revert的好机会:

$ svn revert README
Reverted 'README'

Subversion reverts the file to its premodified state by overwriting it with the cached “pristine” copy from the .svn area. But also note that svn revert can undo any scheduled operations—for example, you might decide that you don't want to add a new file after all:

$ svn status foo
?      foo

$ svn add foo
A         foo

$ svn revert foo
Reverted 'foo'

$ svn status foo
?      foo

注意

svn revertITEM的效果与删除ITEM然后执行svn update -r BASEITEM完全一样,但是,如果你使用svn revert它不必通知版本库就可以恢复文件。

或许你不小心删除了一个文件:

$ svn status README

$ svn delete README
D         README

$ svn revert README
Reverted 'README'

$ svn status README

解决冲突(合并别人的修改)

我们可以使用svn status -u来预测冲突,当你运行svn update一些有趣的事情发生了:

$ svn update
U  INSTALL
G  README
Conflict discovered in 'bar.c'.
Select: (p) postpone, (df) diff-full, (e) edit,
        (h) help for more options:

UG没必要关心,文件干净的接受了版本库的变化,文件标示为U表明本地没有修改,文件已经根据版本库更新。G标示合并,标示本地已经修改过,与版本库没有重迭的地方,已经合并。

But the next two lines are part of a feature (new in Subversion 1.5) called interactive conflict resolution. This means that the changes from the server overlapped with your own, and you have the opportunity to resolve this conflict. The most commonly used options are displayed, but you can see all of the options by typing h:

…
  (p)  postpone    - mark the conflict to be resolved later
  (df) diff-full   - show all changes made to merged file
  (e)  edit        - change merged file in an editor
  (r)  resolved    - accept merged version of file
  (mf) mine-full   - accept my version of entire file (ignore their changes)
  (tf) theirs-full - accept their version of entire file (lose my changes)
  (l)  launch      - launch external tool to resolve conflict
  (h)  help        - show this list

Let's briefly review each of these options before we go into detail on what each option means.

(p)ostpone

Leave the file in a conflicted state for you to resolve after your update is complete.

(d)iff

Display the differences between the base revision and the conflicted file itself in unified diff format.

(e)dit

Open the file in conflict with your favorite editor, as set in the environment variable EDITOR.

(r)esolved

After editing a file, tell svn that you've resolved the conflicts in the file and that it should accept the current contents—basically that you've “resolved” the conflict.

(m)ine-(f)ull

Discard the newly received changes from the server and use only your local changes for the file under review.

(t)heirs-(f)ull

Discard your local changes to the file under review and use only the newly received changes from the server.

(l)aunch

Launch an external program to perform the conflict resolution. This requires a bit of preparation beforehand.

(h)elp

Show the list of all possible commands you can use in interactive conflict resolution.

We'll cover these commands in more detail now, grouping them together by related functionality.

Viewing conflict differences interactively

Before deciding how to attack a conflict interactively, odds are that you'd like to see what exactly is in conflict, and the diff command (d) is what you'll use for this:

…
Select: (p) postpone, (df) diff-full, (e) edit,
        (h)elp for more options : d
--- .svn/text-base/sandwich.txt.svn-base      Tue Dec 11 21:33:57 2007
+++ .svn/tmp/tempfile.32.tmp     Tue Dec 11 21:34:33 2007
@@ -1 +1,5 @@
-Just buy a sandwich.
+<<<<<<< .mine
+Go pick up a cheesesteak.
+=======
+Bring me a taco!
+>>>>>>> .r32
…

The first line of the diff content shows the previous contents of the working copy (the BASE revision), the next content line is your change, and the last content line is the change that was just received from the server (usually the HEAD revision). With this information in hand, you're ready to move on to the next action.

Resolving conflict differences interactively

There are four different ways to resolve conflicts interactively—two of which allow you to selectively merge and edit changes, and two of which allow you to simply pick a version of the file and move along.

If you wish to choose some combination of your local changes, you can use the “edit” command (e) to manually edit the file with conflict markers in a text editor (determined by the EDITOR environment variable). Editing the file by hand in your favorite text editor is a somewhat low-tech way of remedying conflicts (see “Merging conflicts by hand”一节 for a walkthrough), so some people like to use fancy graphical merge tools instead.

In order to use a merge tool, you need to either set the SVN_MERGE environment variable or define the merge-tool-cmd option in your Subversion configuration file (see “配置选项”一节 for more details). Subversion will pass four arguments to the merge tool: The BASE revision of the file, the revision of the file received from the server as part of the update, the copy of the file containing your local edits, and lastly, the merged copy of the file (which contains conflict markers). If your merge tool is expecting arguments in a different order or format, you'll need to write a wrapper script for Subversion to invoke. After you've edited the file, if you're satisfied with the changes you've made, you can tell Subversion that the edited file is no longer in conflict by using the “resolve” command (r).

If you decide that you don't need to merge any changes, but just want to accept one version of the file or the other, you can either choose your changes (aka “mine”) by using the “mine-full” command (mf) or choose theirs by using the “theirs-full” command (tf).

Postponing conflict resolution

This may sound like an appropriate section for avoiding marital disagreements, but it's actually still about Subversion, so read on. If you're doing an update and encounter a conflict that you're not prepared to review or resolve, you can type p to postpone resolving a conflict on a file-by-file basis when you run svn update. If you're running an update and don't want to resolve any conflicts, you can pass the --non-interactive option to svn update, and any file in conflict will be marked with a C automatically.

The C stands for conflict. This means that the changes from the server overlapped with your own, and now you have to manually choose between them after the update has completed. When you postpone a conflict resolution, svn typically does three things to assist you in noticing and resolving that conflict:

  • Subversion prints a C during the update and remembers that the file is in a state of conflict.

  • If Subversion considers the file to be mergeable, it places conflict markers—special strings of text that delimit the “sides” of the conflict—into the file to visibly demonstrate the overlapping areas. (Subversion uses the svn:mime-type property to decide if a file is capable of contextual, line-based merging. See “文件内容类型”一节 to learn more.)

  • 对于每一个冲突的文件,Subversion放置三个额外的未版本化文件到你的工作拷贝:

    filename.mine

    你更新前的文件,没有冲突标志,只是你最新更改的内容。(如果Subversion认为这个文件不可以合并,.mine文件不会创建,因为它和工作文件相同。)

    filename.rOLDREV

    这是你的做更新操作以前的BASE版本文件,就是你在上次更新之后未作更改的版本。

    filename.rNEWREV

    这是你的Subversion客户端从服务器刚刚收到的版本,这个文件对应版本库的HEAD版本。

    Here OLDREV is the revision number of the file in your .svn directory, and NEWREV is the revision number of the repository HEAD.

For example, Sally makes changes to the file sandwich.txt in the repository. Harry has just changed the file in his working copy and checked it in. Sally updates her working copy before checking in and she gets a conflict, which she postpones:

$ svn update
Conflict discovered in 'sandwich.txt'.
Select: (p) postpone, (df) diff-full, (e) edit,
        (h)elp for more options : p
C  sandwich.txt
Updated to revision 2.
$ ls -1
sandwich.txt
sandwich.txt.mine
sandwich.txt.r1
sandwich.txt.r2

At this point, Subversion will not allow Sally to commit the file sandwich.txt until the three temporary files are removed.

$ svn commit -m "Add a few more things"
svn: Commit failed (details follow):
svn: Aborting commit: '/home/sally/svn-work/sandwich.txt' remains in conflict

If you've postponed a conflict, you need to resolve the conflict before Subversion will allow you to commit your changes. You'll do this with the svn resolve command and one of several arguments to the --accept option.

If you want to choose the version of the file that you last checked out before making your edits, choose the base argument.

If you want to choose the version that contains only your edits, choose the mine-full argument.

If you want to choose the version that your most recent update pulled from the server (and thus discarding your edits entirely), choose the theirs-full argument.

However, if you want to pick and choose from your changes and the changes that your update fetched from the server, merge the conflicted text “by hand” (by examining and editing the conflict markers within the file) and then choose the working argument.

svn resolve removes the three temporary files, accepts the version of the file that you specified with the --accept option, and Subversion no longer considers the file to be in a state of conflict.

$ svn resolve --accept working sandwich.txt
Resolved conflicted state of 'sandwich.txt'

Merging conflicts by hand

第一次尝试解决冲突让人感觉很害怕,但经过一点训练,它简单的像是骑着车子下坡。

这里一个简单的例子,由于不良的交流,你和同事Sally,同时编辑了sandwich.txt。Sally提交了修改,当你准备更新你的工作拷贝,冲突发生了,我们不得不去修改sandwich.txt来解决这个问题。首先,看一下这个文件:

$ cat sandwich.txt
Top piece of bread
Mayonnaise
Lettuce
Tomato
Provolone
<<<<<<< .mine
Salami
Mortadella
Prosciutto
=======
Sauerkraut
Grilled Chicken
>>>>>>> .r2
Creole Mustard
Bottom piece of bread

The strings of less-than signs, equal signs, and greater-than signs are conflict markers and are not part of the actual data in conflict. You generally want to ensure that those are removed from the file before your next commit. The text between the first two sets of markers is composed of the changes you made in the conflicting area:

<<<<<<< .mine
Salami
Mortadella
Prosciutto
=======

后两组之间的是Sally提交的修改冲突:

=======
Sauerkraut
Grilled Chicken
>>>>>>> .r2

Usually you won't want to just delete the conflict markers and Sally's changes—she's going to be awfully surprised when the sandwich arrives and it's not what she wanted. So this is where you pick up the phone or walk across the office and explain to Sally that you can't get sauerkraut from an Italian deli. [6] Once you've agreed on the changes you will check in, edit your file and remove the conflict markers.

Top piece of bread
Mayonnaise
Lettuce
Tomato
Provolone
Salami
Mortadella
Prosciutto
Creole Mustard
Bottom piece of bread

Now run svn resolve, and you're ready to commit your changes:

$ svn resolve --accept working sandwich.txt
Resolved conflicted state of 'sandwich.txt'
$ svn commit -m "Go ahead and use my sandwich, discarding Sally's edits."

Note that svn resolve, unlike most of the other commands we deal with in this chapter, requires that you explicitly list any filenames that you wish to resolve. In any case, you want to be careful and run svn resolve only when you're certain that you've fixed the conflict in your file—once the temporary files are removed, Subversion will let you commit the file even if it still contains conflict markers.

记住,如果你修改冲突时感到混乱,你可以参考subversion生成的三个文件—包括你未作更新的文件。你也可以使用三方交互合并工具检验这三个文件。

Discarding your changes in favor of a newly fetched revision

If you get a conflict and decide that you want to throw out your changes, you can run svn resolve --accept theirs-full and Subversion will discard your edits and remove the temporary files:

$ svn update
Conflict discovered in 'sandwich.txt'.
Select: (p) postpone, (df) diff-full, (e) edit,
        (h) help for more options: p
C    sandwich.txt
Updated to revision 2.
$ ls sandwich.*
sandwich.txt  sandwich.txt.mine  sandwich.txt.r2  sandwich.txt.r1
$ svn resolve --accept theirs-full sandwich.txt
Resolved conflicted state of 'sandwich.txt'

Punting: using svn revert

If you decide that you want to throw out your changes and start your edits again (Whether this occurs after a conflict or anytime), just revert your changes:

$ svn revert sandwich.txt
Reverted 'sandwich.txt'
$ ls sandwich.*
sandwich.txt

Note that when you revert a conflicted file, you don't have to run svn resolve.

提交你的修改

最后!你的修改结束了,你合并了服务器上所有的修改,你准备好提交修改到版本库。

svn commit命令发送所有的修改到版本库,当你提交修改时,你需要提供一些描述修改的日志信息,你的信息会附到这个修订版本上,如果信息很简短,你可以在命令行中使用--message(或-m)选项:

$ svn commit -m "Corrected number of cheese slices."
Sending        sandwich.txt
Transmitting file data .
Committed revision 3.

However, if you've been composing your log message as you work, you may want to tell Subversion to get the message from a file by passing the filename with the --file (-F) option:

$ svn commit -F logmsg
Sending        sandwich.txt
Transmitting file data .
Committed revision 4.

如果你没有指定--message或者--file选项,Subversion会自动地启动你最喜欢的编辑器(见“配置”一节editor-cmd部分)来编辑日志信息。

提示

如果你使用编辑器撰写日志信息时希望取消提交,你可以直接关掉编辑器,不要保存,如果你已经做过保存,只要简单的删掉所有的文本并再次保存,然后退出。

$ svn commit
Waiting for Emacs...Done

Log message unchanged or not specified
(a)bort, (c)ontinue, (e)dit
a
$

The repository doesn't know or care if your changes make any sense as a whole; it checks only to make sure that nobody else has changed any of the same files that you did when you weren't looking. If somebody has done that, the entire commit will fail with a message informing you that one or more of your files is out of date:

$ svn commit -m "Add another rule"
Sending        rules.txt
svn: Commit failed (details follow):
svn: File '/sandwich.txt' is out of date
…

(错误信息的精确措辞依赖于网络协议和你使用的服务器,但对于所有的情况,其思想完全一样。)

此刻,你需要运行svn update来处理所有的合并和冲突,然后再尝试提交。

我们已经覆盖了Subversion基本的工作周期,还有许多其它特性可以管理你得版本库和工作拷贝,但是只使用前面介绍的命令你就可以很进行日常工作了,我们还会覆盖更多用的还算频繁的命令。

检验历史

Your Subversion repository is like a time machine. It keeps a record of every change ever committed and allows you to explore this history by examining previous versions of files and directories as well as the metadata that accompanies them. With a single Subversion command, you can check out the repository (or restore an existing working copy) exactly as it was at any date or revision number in the past. However, sometimes you just want to peer into the past instead of going into it.

有许多命令可以为你提供版本库历史:

svn log

Shows you broad information: log messages with date and author information attached to revisions and which paths changed in each revision.

svn diff

显示特定修改的行级详细信息。

svn cat

Retrieves a file as it existed in a particular revision number and displays it on your screen.

svn list

显示一个目录在某一版本存在的文件。

Generating a List of Historical Changes

To find information about the history of a file or directory, use the svn log command. svn log will provide you with a record of who made changes to a file or directory, at what revision it changed, the time and date of that revision, and—if it was provided—the log message that accompanied the commit.

$ svn log
------------------------------------------------------------------------
r3 | sally | Mon, 15 Jul 2002 18:03:46 -0500 | 1 line

Added include lines and corrected # of cheese slices.
------------------------------------------------------------------------
r2 | harry | Mon, 15 Jul 2002 17:47:57 -0500 | 1 line

Added main() methods.
------------------------------------------------------------------------
r1 | sally | Mon, 15 Jul 2002 17:40:08 -0500 | 1 line

Initial import
------------------------------------------------------------------------

Note that the log messages are printed in reverse chronological order by default. If you wish to see a different range of revisions in a particular order or just a single revision, pass the --revision (-r) option:

$ svn log -r 5:19    # shows logs 5 through 19 in chronological order

$ svn log -r 19:5    # shows logs 5 through 19 in reverse order

$ svn log -r 8       # shows log for revision 8

你也可以检查单个文件或目录的日志历史,举个例子:

$ svn log foo.c
…
$ svn log http://foo.com/svn/trunk/code/foo.c
…

这样只会显示这个工作文件(或者URL)做过修订的版本的日志信息。

If you want even more information about a file or directory, svn log also takes a --verbose (-v) option. Because Subversion allows you to move and copy files and directories, it is important to be able to track path changes in the filesystem. So, in verbose mode, svn log will include a list of changed paths in a revision in its output:

$ svn log -r 8 -v
------------------------------------------------------------------------
r8 | sally | 2002-07-14 08:15:29 -0500 | 1 line
Changed paths:
   M /trunk/code/foo.c
   M /trunk/code/bar.h
   A /trunk/code/doc/README

Frozzled the sub-space winch.

------------------------------------------------------------------------

svn log也有一个--quiet (-q)选项,会禁止日志信息的主要部分,当与--verbose结合使用,仅会显示修改的文件名。

Examining the Details of Historical Changes

我们已经看过svn diff—使用标准区别文件格式显示区别,它在提交前用来显示本地工作拷贝与版本库的区别。

事实上,svn diff种不同的用法:

  • 比较本地修改

  • 比较工作拷贝与版本库

  • 比较版本库与版本库

检查本地修改

像我们看到的,不使用任何参数调用时,svn diff将会比较你的工作文件与缓存在.svn的“原始”拷贝:

$ svn diff
Index: rules.txt
===================================================================
--- rules.txt	(revision 3)
+++ rules.txt	(working copy)
@@ -1,4 +1,5 @@
 Be kind to others
 Freedom = Responsibility
 Everything in moderation
-Chew with your mouth open
+Chew with your mouth closed
+Listen when others are speaking
$

Comparing working copy to repository

如果传递一个--revision(-r)参数,你的工作拷贝会与指定的版本比较。

$ svn diff -r 3 rules.txt
Index: rules.txt
===================================================================
--- rules.txt	(revision 3)
+++ rules.txt	(working copy)
@@ -1,4 +1,5 @@
 Be kind to others
 Freedom = Responsibility
 Everything in moderation
-Chew with your mouth open
+Chew with your mouth closed
+Listen when others are speaking
$

比较版本库与版本库

If two revision numbers, separated by a colon, are passed via --revision (-r), then the two revisions are directly compared:

$ svn diff -r 2:3 rules.txt
Index: rules.txt
===================================================================
--- rules.txt	(revision 2)
+++ rules.txt	(revision 3)
@@ -1,4 +1,4 @@
 Be kind to others
-Freedom = Chocolate Ice Cream
+Freedom = Responsibility
 Everything in moderation
 Chew with your mouth open
$

A more convenient way of comparing a revision to the previous revision is to use the --change (-c) option:

$ svn diff -c 3 rules.txt
Index: rules.txt
===================================================================
--- rules.txt	(revision 2)
+++ rules.txt	(revision 3)
@@ -1,4 +1,4 @@
 Be kind to others
-Freedom = Chocolate Ice Cream
+Freedom = Responsibility
 Everything in moderation
 Chew with your mouth open
$

最后,即使你在本机没有工作拷贝,还是可以比较版本库的修订版本,只需要在命令行中输入合适的URL:

$ svn diff -c 5 http://svn.example.com/repos/example/trunk/text/rules.txt
…
$

Browsing the Repository

通过svn catsvn list,你可以在未修改工作修订版本的情况下查看文件和目录的内容,实际上,你甚至也不需要有一个工作拷贝。

svn cat

如果你只是希望检查一个过去的版本而不希望察看它们的区别,使用svn cat

$ svn cat -r 2 rules.txt
Be kind to others
Freedom = Chocolate Ice Cream
Everything in moderation
Chew with your mouth open
$

你可以重定向输出到一个文件:

$ svn cat -r 2 rules.txt > rules.txt.v2
$

svn list

svn list可以在不下载文件到本地目录的情况下来察看目录中的文件:

$ svn list http://svn.collab.net/repos/svn
README
branches/
clients/
tags/
trunk/

If you want a more detailed listing, pass the --verbose (-v) flag to get output like this:

$ svn list -v http://svn.collab.net/repos/svn
  20620 harry            1084 Jul 13  2006 README
  23339 harry                 Feb 04 01:40 branches/
  21282 sally                 Aug 27 09:41 developer-resources/
  23198 harry                 Jan 23 17:17 tags/
  23351 sally                 Feb 05 13:26 trunk/

这些列告诉你文件和目录最后修改的修订版本、做出修改的用户、如果是文件还会有文件的大小,最后是修改日期和项目的名字。

警告

The svn list with no arguments defaults to the repository URL of the current working directory, not the local working copy directory. After all, if you want a listing of your local directory, you could use just plain ls (or any reasonable non-Unixy equivalent).

Fetching Older Repository Snapshots

In addition to all of the previous commands, you can use svn update and svn checkout with the --revision option to take an entire working copy “back in time”: [7]

$ svn checkout -r 1729 # Checks out a new working copy at r1729
…
$ svn update -r 1729 # Updates an existing working copy to r1729
…

提示

Many Subversion newcomers attempt to use the preceding svn update example to “undo” committed changes, but this won't work as you can't commit changes that you obtain from backdating a working copy if the changed files have newer revisions. See “找回删除的项目”一节 for a description of how to “undo” a commit.

最后,如果你构建了一个版本,并且希望从Subversion打包文件,但是你不希望有讨厌的.svn目录,这时你可以导出版本库的一部分文件而没有.svn目录。就像svn updatesvn checkout,你也可以传递--revision选项给svn export

$ svn export http://svn.example.com/svn/repos1 # Exports latest revision
…
$ svn export http://svn.example.com/svn/repos1 -r 1729
# Exports revision r1729
…

有时你只需要清理

Now that we've covered the day-to-day tasks that you'll frequently use Subversion for, we'll review a few administrative tasks relating to your working copy.

Disposing of a Working Copy

Subversion doesn't track either the state or existence of working copies on the server, so there's no server overhead to keeping working copies around. Likewise, there's no need to let the server know that you're going to delete a working copy.

If you're likely to use a working copy again, there's nothing wrong with just leaving it on disk until you're ready to use it again, at which point all it takes is an svn update to bring it up to date and ready for use.

However, if you're definitely not going to use a working copy again, you can safely delete the entire thing, but you'd be well served to take a look through the working copy for unversioned files. To find these files, run svn status and review any files that are prefixed by a ? to make certain that they're not of importance. After you're done reviewing, you can safely delete your working copy.

Recovering from an Interruption

When Subversion modifies your working copy (or any information within .svn), it tries to do so as safely as possible. Before changing the working copy, Subversion writes its intentions to a log file. Next, it executes the commands in the log file to apply the requested change, holding a lock on the relevant part of the working copy while it works—to prevent other Subversion clients from accessing the working copy mid-change. Finally, Subversion removes the log file. Architecturally, this is similar to a journaled filesystem. If a Subversion operation is interrupted (if the process is killed or if the machine crashes, for example), the log files remain on disk. By re-executing the log files, Subversion can complete the previously started operation, and your working copy can get itself back into a consistent state.

And this is exactly what svn cleanup does: it searches your working copy and runs any leftover logs, removing working copy locks in the process. If Subversion ever tells you that some part of your working copy is “locked,” then this is the command that you should run. Also, svn status will display an L next to locked items:

$ svn status
  L    somedir
M      somedir/foo.c

$ svn cleanup
$ svn status
M      somedir/foo.c

不要将工作拷贝锁与Subversion用户使用并发版本控制的“锁定-修改-解锁”模型创建的锁混淆;更多细节见The Three Meanings of “Lock

总结

我们已经覆盖了大多数Subversion的客户端命令,引人注目的例外是处理分支与合并(见第 4 章 分支与合并)以及属性(见“属性”一节)的命令,然而你也许会希望跳到第 9 章 Subversion 完全参考来察看所有不同的命令—怎样利用它们使你的工作更容易。



[3] Of course, you're not terribly worried—first because you know that you can't really delete anything from Subversion and, secondly, because your Subversion password isn't the same as any of the other 3 million passwords you have, right? Right?

[4] Of course, nothing is ever totally deleted from the repository—just from the HEAD of the repository. You can get back anything you delete by checking out (or updating your working copy to) a revision earlier than the one in which you deleted it. Also see “找回删除的项目”一节.

[5] 而且你也没有WAN卡,考虑到你得到我们,哈!

[6] 如果你向他们询问,他们非常有理由把你带到城外的铁轨上。

[7] 看到了吧?我们说过Subversion是一个时间机器。

高级主题

If you've been reading this book chapter by chapter, from start to finish, you should by now have acquired enough knowledge to use the Subversion client to perform the most common version control operations. You understand how to check out a working copy from a Subversion repository. You are comfortable with submitting and receiving changes using the svn commit and svn update functions. You've probably even developed a reflex that causes you to run the svn status command almost unconsciously. For all intents and purposes, you are ready to use Subversion in a typical environment.

But the Subversion feature set doesn't stop at “common version control operations.” It has other bits of functionality besides just communicating file and directory changes to and from a central repository.

This chapter highlights some of Subversion's features that, while important, aren't part of the typical user's daily routine. It assumes that you are familiar with Subversion's basic file and directory versioning capabilities. If you aren't, you'll want to first read 第 1 章 基本概念 and 第 2 章 基本使用. Once you've mastered those basics and consumed this chapter, you'll be a Subversion power user!

版本清单

As we described in “修订版本”一节, revision numbers in Subversion are pretty straightforward—integers that keep getting larger as you commit more changes to your versioned data. Still, it doesn't take long before you can no longer remember exactly what happened in each and every revision. Fortunately, the typical Subversion workflow doesn't often demand that you supply arbitrary revisions to the Subversion operations you perform. For operations that do require a revision specifier, you generally supply a revision number that you saw in a commit email, in the output of some other Subversion operation, or in some other context that would give meaning to that particular number.

But occasionally, you need to pinpoint a moment in time for which you don't already have a revision number memorized or handy. So besides the integer revision numbers, svn allows as input some additional forms of revision specifiers: revision keywords and revision dates.

注意

当用来指定修订版本范围时,不同形式的Subversion修订版本可以混合匹配。例如,你可以REV1是修订版本关键字,REV2是修订版本号,或者是REV1是日期,而REV2是修订版本关键字,等等。不同的修订版本指定符是等价的,所以你可以在冒号两边任意使用。

修订版本关键字

The Subversion client understands a number of revision keywords. These keywords can be used instead of integer arguments to the --revision (-r) option, and are resolved into specific revision numbers by Subversion:

HEAD

版本库中最新的(或者是“最年轻的”)版本。

BASE

The revision number of an item in a working copy. If the item has been locally modified, this refers to the way the item appears without those local modifications.

COMMITTED

项目最近修改的修订版本,与BASE相同或更早。

PREV

一个项目最后修改版本之前的那个版本,技术上可以认为是COMMITTED -1。

因为可以从描述中得到,关键字PREVBASECOMMITTED只在引用工作拷贝路径时使用,而不能用于版本库URL,而关键字HEAD则可以用于两种路径类型。

下面是一些修订版本关键字的例子:

$ svn diff -r PREV:COMMITTED foo.c
# shows the last change committed to foo.c

$ svn log -r HEAD
# shows log message for the latest repository commit

$ svn diff -r HEAD
# compares your working copy (with all of its local changes) to the
# latest version of that tree in the repository

$ svn diff -r BASE:HEAD foo.c
# compares the unmodified version of foo.c with the latest version of
# foo.c in the repository

$ svn log -r BASE:HEAD
# shows all commit logs for the current versioned directory since you
# last updated

$ svn update -r PREV foo.c
# rewinds the last change on foo.c, decreasing foo.c's working revision

$ svn diff -r BASE:14 foo.c
# compares the unmodified version of foo.c with the way foo.c looked
# in revision 14

版本日期

Revision numbers reveal nothing about the world outside the version control system, but sometimes you need to correlate a moment in real time with a moment in version history. To facilitate this, the --revision (-r) option can also accept as input date specifiers wrapped in curly braces ({ and }). Subversion accepts the standard ISO-8601 date and time formats, plus a few others. Here are some examples. (Remember to use quotes around any date that contains spaces.)

$ svn checkout -r {2006-02-17}
$ svn checkout -r {15:30}
$ svn checkout -r {15:30:00.200000}
$ svn checkout -r {"2006-02-17 15:30"}
$ svn checkout -r {"2006-02-17 15:30 +0230"}
$ svn checkout -r {2006-02-17T15:30}
$ svn checkout -r {2006-02-17T15:30Z}
$ svn checkout -r {2006-02-17T15:30-04:00}
$ svn checkout -r {20060217T1530}
$ svn checkout -r {20060217T1530Z}
$ svn checkout -r {20060217T1530-0500}
…

当你指定一个日期,Subversion会在版本库找到接近这个日期的最近版本,并且对这个版本继续操作:

$ svn log -r {2006-11-28}
------------------------------------------------------------------------
r12 | ira | 2006-11-27 12:31:51 -0600 (Mon, 27 Nov 2006) | 6 lines
…

你可以使用时间段,Subversion会找到这段时间的所有版本:

$ svn log -r {2006-11-20}:{2006-11-29}
…

警告

Since the timestamp of a revision is stored as an unversioned, modifiable property of the revision (see “属性”一节), revision timestamps can be changed to represent complete falsifications of true chronology, or even removed altogether. Subversion's ability to correctly convert revision dates into real revision numbers depends on revision datestamps maintaining a sequential ordering—the younger the revision, the younger its timestamp. If this ordering isn't maintained, you will likely find that trying to use dates to specify revision ranges in your repository doesn't always return the data you might have expected.

属性

我们已经详细讲述了Subversion存储和检索版本库中不同版本的文件和目录的细节,并且用了好几个章节来论述这个工具的基本功能。如果对于版本化的支持到此为止,从版本控制的角度来看Subversion已经完整了。

但不仅仅如此。

作为目录和文件版本化的补充,Subversion提供了对每一个版本化的目录和文件添加、修改和删除版本化的元数据的接口,我们用属性来表示这些元数据。我们可以认为它们是一个两列的表,附加到你的工作拷贝的每个条目上,映射属性名到任意的值。一般来说,属性的名称和值可以是你希望的任何值,限制就是名称必须是可读的文本,并且最好的一点是这些属性也是版本化的,就像你的文本文件内容,你可以像提交文本修改一样修改、提交和恢复属性修改,当你更新时也会接收到别人的属性修改—你不必为适应属性改变你的工作流程。

注意

Subversion自己保留了一组名称以svn:开头的属性,现在已经有了一些在用的属性,所以在你根据需要创建自定义属性时,需要避免这些前缀开头的名称,否则,Subversion的新版本可能会采用同名的属性来满足新的特性,而其含义可能会完全不同。

Properties show up elsewhere in Subversion, too. Just as files and directories may have arbitrary property names and values attached to them, each revision as a whole may have arbitrary properties attached to it. The same constraints apply—human-readable names and anything-you-want binary values. The main difference is that revision properties are not versioned. In other words, if you change the value of, or delete, a revision property, there's no way, within the scope of Subversion's functionality, to recover the previous value.

Subversion has no particular policy regarding the use of properties. It asks only that you not use property names that begin with the prefix svn:. That's the namespace that it sets aside for its own use. And Subversion does, in fact, use properties—both the versioned and unversioned variety. Certain versioned properties have special meaning or effects when found on files and directories, or they house a particular bit of information about the revisions on which they are found. Certain revision properties are automatically attached to revisions by Subversion's commit process, and they carry information about the revision. Most of these properties are mentioned elsewhere in this or other chapters as part of the more general topics to which they are related. For an exhaustive list of Subversion's predefined properties, see “Subversion Properties”一节.

In this section, we will examine the utility—both to users of Subversion and to Subversion itself—of property support. You'll learn about the property-related svn subcommands and how property modifications affect your normal Subversion workflow.

为什么需要属性?

就像Subversion使用属性保存其包含的文件、目录和修订版本的附加信息,你也会发现属性有一些类似的使用,你会发现如果在数据附近有个地方保存自定义元数据会非常有用。

Say you wish to design a web site that houses many digital photos and displays them with captions and a datestamp. Now, your set of photos is constantly changing, so you'd like to have as much of this site automated as possible. These photos can be quite large, so as is common with sites of this nature, you want to provide smaller thumbnail images to your site visitors.

Now, you can get this functionality using traditional files. That is, you can have your image123.jpg and an image123-thumbnail.jpg side by side in a directory. Or if you want to keep the filenames the same, you might have your thumbnails in a different directory, such as thumbnails/image123.jpg. You can also store your captions and datestamps in a similar fashion, again separated from the original image file. But the problem here is that your collection of files multiplies with each new photo added to the site.

Now consider the same web site deployed in a way that makes use of Subversion's file properties. Imagine having a single image file, image123.jpg, with properties set on that file that are named caption, datestamp, and even thumbnail. Now your working copy directory looks much more manageable—in fact, it looks to the casual browser like there are nothing but image files in it. But your automation scripts know better. They know that they can use svn (or better yet, they can use the Subversion language bindings—see “使用API”一节) to dig out the extra information that your site needs to display without having to read an index file or play path manipulation games.

Custom revision properties are also frequently used. One common such use is a property whose value contains an issue tracker ID with which the revision is associated, perhaps because the change made in that revision fixes a bug filed in the tracker issue with that ID. Other uses include hanging more friendly names on the revision—it might be hard to remember that revision 1935 was a fully tested revision. But if there's, say, a test-results property on that revision with the value all passing, that's meaningful information to have.

操作属性

The svn command affords a few ways to add or modify file and directory properties. For properties with short, human-readable values, perhaps the simplest way to add a new property is to specify the property name and value on the command line of the propset subcommand:

$ svn propset copyright '(c) 2006 Red-Bean Software' calc/button.c
property 'copyright' set on 'calc/button.c'
$

But we've been touting the flexibility that Subversion offers for your property values. And if you are planning to have a multiline textual, or even binary, property value, you probably do not want to supply that value on the command line. So the propset subcommand takes a --file (-F) option for specifying the name of a file that contains the new property value.

$ svn propset license -F /path/to/LICENSE calc/button.c
property 'license' set on 'calc/button.c'
$

对于属性名称也有一些限制,属性名必须以一个字符、一个冒号(:)或下划线(_)开始,之后你可以使用数字,横线(-)和句号(.)。 [8]

In addition to the propset command, the svn program supplies the propedit command. This command uses the configured editor program (see “配置”一节) to add or modify properties. When you run the command, svn invokes your editor program on a temporary file that contains the current value of the property (or that is empty, if you are adding a new property). Then, you just modify that value in your editor program until it represents the new value you wish to store for the property, save the temporary file, and then exit the editor program. If Subversion detects that you've actually changed the existing value of the property, it will accept that as the new property value. If you exit your editor without making any changes, no property modification will occur:

$ svn propedit copyright calc/button.c  ### exit the editor without changes
No changes to property 'copyright' on 'calc/button.c'
$

We should note that, as with other svn subcommands, those related to properties can act on multiple paths at once. This enables you to modify properties on whole sets of files with a single command. For example, we could have done the following:

$ svn propset copyright '(c) 2006 Red-Bean Software' calc/*
property 'copyright' set on 'calc/Makefile'
property 'copyright' set on 'calc/button.c'
property 'copyright' set on 'calc/integer.c'
…
$

如果不能方便的得到存储的属性值,那么属性的添加和编辑操作也不会很容易,所以svn提供了两个子命令来显示文件和目录存储的属性名和值。svn proplist命令会列出路径上存在的所有属性名称,一旦你知道了某个节点的属性名称,你可以用svn propget获取它的值,这个命令获取给定的路径(或者是一组路径)和属性名称,打印这个属性的值到标准输出。

$ svn proplist calc/button.c
Properties on 'calc/button.c':
  copyright
  license
$ svn propget copyright calc/button.c
(c) 2006 Red-Bean Software

还有一个proplist变种命令会列出所有属性的名称和值,只需要设置--verbose(-v)选项。

$ svn proplist -v calc/button.c
Properties on 'calc/button.c':
  copyright : (c) 2006 Red-Bean Software
  license : ================================================================
Copyright (c) 2006 Red-Bean Software.  All rights reserved.

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions 
are met:

1. Redistributions of source code must retain the above copyright
notice, this list of conditions, and the recipe for Fitz's famous
red-beans-and-rice.
…

最后一个与属性相关的子命令是propdel,因为Subversion允许属性值为空,所有不能用propedit或者propset命令删除一个属性。例如,这个命令不会产生预期的效果:

$ svn propset license '' calc/button.c
property 'license' set on 'calc/button.c'
$ svn proplist -v calc/button.c
Properties on 'calc/button.c':
  copyright : (c) 2006 Red-Bean Software
  license : 
$

你需要用propdel来删除属性,语法与其它与属性命令相似:

$ svn propdel license calc/button.c
property 'license' deleted from 'calc/button.c'.
$ svn proplist -v calc/button.c
Properties on 'calc/button.c':
  copyright : (c) 2006 Red-Bean Software
$

Remember those unversioned revision properties? You can modify those, too, using the same svn subcommands that we just described. Simply add the --revprop command line parameter and specify the revision whose property you wish to modify. Since revisions are global, you don't need to specify a target path to these property-related commands so long as you are positioned in a working copy of the repository whose revision property you wish to modify. Otherwise, you can simply provide the URL of any path in the repository of interest (including the repository's root URL). For example, you might want to replace the commit log message of an existing revision. [9] If your current working directory is part of a working copy of your repository, you can simply run the svn propset command with no target path:

$ svn propset svn:log '* button.c: Fix a compiler warning.' -r11 --revprop
property 'svn:log' set on repository revision '11'
$

但是即使你没有从版本库检出一个工作拷贝,你仍然可以通过提供版本库根URL来影响属性修改。

$ svn propset svn:log '* button.c: Fix a compiler warning.' -r11 --revprop \
              http://svn.example.com/repos/project
property 'svn:log' set on repository revision '11'
$

注意,修改这些未版本化的属性的能力一定要明确的添加给版本库管理员(见“修正提交消息”一节)。因为属性没有版本化,如果编辑的时候不小心,就会冒丢失信息的风险,版本库管理员可以设置方法来防范这种意外,缺省情况下,修改未版本化的属性是禁止的。

提示

Users should, where possible, use svn propedit instead of svn propset. While the end result of the commands is identical, the former will allow them to see the current value of the property that they are about to change, which helps them to verify that they are, in fact, making the change they think they are making. This is especially true when modifying unversioned revision properties. Also, it is significantly easier to modify multiline property values in a text editor than at the command line.

属性和 Subversion 工作流程

现在你已经熟悉了所有与属性相关的svn子命令,让我们看看属性修改如何影响Subversion的工作流。我们前面提到过,文件和目录的属性是版本化的,这一点类似于版本化的文件内容。后果之一,就是Subversion具有了同样的机制来合并—用干净或者冲突的方式—其他人的修改应用到你的修改。

As with file contents, your property changes are local modifications, made permanent only when you commit them to the repository with svn commit. Your property changes can be easily unmade, too—the svn revert command will restore your files and directories to their unedited states—contents, properties, and all. Also, you can receive interesting information about the state of your file and directory properties by using the svn status and svn diff commands.

$ svn status calc/button.c
 M     calc/button.c
$ svn diff calc/button.c
Property changes on: calc/button.c
___________________________________________________________________
Name: copyright
   + (c) 2006 Red-Bean Software

$

注意status子命令显示的M在第二列而不是在第一列,这是因为我们修改了calc/button.c的属性,而不是它的文本内容,如果我们都修改了,我们也会看到M出现在第一列(见“查看你的修改概况”一节)。

You might also have noticed the nonstandard way that Subversion currently displays property differences. You can still run svn diff and redirect the output to create a usable patch file. The patch program will ignore property patches—as a rule, it ignores any noise it can't understand. This does, unfortunately, mean that to fully apply a patch generated by svn diff, any property modifications will need to be applied by hand.

自动设置属性

属性是Subversion一个强大的特性,成为本章和其它章讨论的许多Subversion特性的关键组成部分—文本区别和合并支持、关键字替换、新行的自动转换等等。但是为了从属性得到完全的利益,他们必须设置到正确的文件和目录。不幸的是,在日常工作中很容易忘记这一步工作,特别是当没有设置属性不会引起明显的错误时(至少相对与未能添加一个文件到版本控制这种操作),为了帮助你在需要添加属性的文件上添加属性,Subversion提供了一些简单但是有用的特性。

Whenever you introduce a file to version control using the svn add or svn import commands, Subversion tries to assist by setting some common file properties automatically. First, on operating systems whose filesystems support an execute permission bit, Subversion will automatically set the svn:executable property on newly added or imported files whose execute bit is enabled. (See “文件的可执行性”一节 later in this chapter for more about this property.)

Secondly, Subversion tries to determine the file's MIME type. If you've configured a mime-types-files runtime configuration parameter, Subversion will try to find a MIME type mapping in that file for your file's extension. If it finds such a mapping, it will set your file's svn:mime-type property to the MIME type it found. If no mapping file is configured, or no mapping for your file's extension could be found, Subversion runs a very basic heuristic to determine if the file contains nontextual content. If so, it automatically sets the svn:mime-type property on that file to application/octet-stream (the generic “this is a collection of bytes” MIME type). Of course, if Subversion guesses incorrectly, or if you wish to set the svn:mime-type property to something more precise—perhaps image/png or application/x-shockwave-flash—you can always remove or edit that property. (For more on Subversion's use of MIME types, see “文件内容类型”一节 later in this chapter.)

Subversion also provides, via its runtime configuration system (see “运行配置区”一节), a more flexible automatic property setting feature that allows you to create mappings of filename patterns to property names and values. Once again, these mappings affect adds and imports, and can not only override the default MIME type decision made by Subversion during those operations, but can also set additional Subversion or custom properties, too. For example, you might create a mapping that says that any time you add JPEG files—ones whose names match the pattern *.jpg—Subversion should automatically set the svn:mime-type property on those files to image/jpeg. Or perhaps any files that match *.cpp should have svn:eol-style set to native, and svn:keywords set to Id. Automatic property support is perhaps the handiest property related tool in the Subversion toolbox. See “配置”一节 for more about configuring that support.

文件移植性

幸运的是,对于许多在不同操作系统下工作的用户,Subversion命令行程序的行为方式几乎完全一致,如果你知道在一个平台上如何运行svn,你也就学会了在其他平台上运行。

However, the same is not always true of other general classes of software or of the actual files you keep in Subversion. For example, on a Windows machine, the definition of a “text file” would be similar to that used on a Linux box, but with a key difference—the character sequences used to mark the ends of the lines of those files. There are other differences, too. Unix platforms have (and Subversion supports) symbolic links; Windows does not. Unix platforms use filesystem permission to determine executability; Windows uses filename extensions.

因为Subversion不是要将世界上的所有此类事情统一起来,所以我们最好是尽可能让我们在多个计算机和操作系统上使用版本化文件和目录时能够更简单,本节描述了Subversion是如何做的。

文件内容类型

Subversion joins the ranks of the many applications that recognize and make use of Multipurpose Internet Mail Extensions (MIME) content types. Besides being a general-purpose storage location for a file's content type, the value of the svn:mime-type file property determines some behavioral characteristics of Subversion itself.

For example, one of the benefits that Subversion typically provides is contextual, line-based merging of changes received from the server during an update into your working file. But for files containing nontextual data, there is often no concept of a “line.” So, for versioned files whose svn:mime-type property is set to a nontextual MIME type (generally, something that doesn't begin with text/, though there are exceptions), Subversion does not attempt to perform contextual merges during updates. Instead, any time you have locally modified a binary working copy file that is also being updated, your file is left untouched and Subversion creates two new files. One file has a .oldrev extension and contains the BASE revision of the file. The other file has a .newrev extension and contains the contents of the updated revision of the file. This behavior is really for the protection of the user against failed attempts at performing contextual merges on files that simply cannot be contextually merged.

警告

The svn:mime-type property, when set to a value that does not indicate textual file contents, can cause some unexpected behaviors with respect to other properties. For example, since the idea of line endings (and therefore, line-ending conversion) makes no sense when applied to nontextual files, Subversion will prevent you from setting the svn:eol-style property on such files. This is obvious when attempted on a single file target—svn propset will error out. But it might not be as clear if you perform a recursive property set, where Subversion will silently skip over files that it deems unsuitable for a given property.

Beginning in Subversion 1.5, users can configure a new mime-types-file runtime configuration parameter, which identifies the location of a MIME types mapping file. Subversion will consult this mapping file to determine the MIME type of newly added and imported files.

另外,如果设置了svn:mime-type属性,Subversion的Apache模块会使用这个值来在HTTP头里输入Content-type:,这给了web浏览器如何显示版本库的一个文件提供了至关重要的线索。

文件的可执行性

On many operating systems, the ability to execute a file as a command is governed by the presence of an execute permission bit. This bit usually defaults to being disabled, and must be explicitly enabled by the user for each file that needs it. But it would be a monumental hassle to have to remember exactly which files in a freshly checked-out working copy were supposed to have their executable bits toggled on, and then to have to do that toggling. So, Subversion provides the svn:executable property as a way to specify that the executable bit for the file on which that property is set should be enabled, and Subversion honors that request when populating working copies with such files.

这个属性对于没有可执行权限位的文件系统无效,如FAT32和NTFS。 [11]也就是说,尽管它没有定义的值,在设置这个属性时,Subversion会强制它的值为*,最后,这个属性只对文件有效,目录无效。

行结束字符串

Unless otherwise noted using a versioned file's svn:mime-type property, Subversion assumes the file contains human-readable data. Generally speaking, Subversion uses this knowledge only to determine if contextual difference reports for that file are possible. Otherwise, to Subversion, bytes are bytes.

This means that by default, Subversion doesn't pay any attention to the type of end-of-line (EOL) markers used in your files. Unfortunately, different operating systems have different conventions about which character sequences represent the end of a line of text in a file. For example, the usual line-ending token used by software on the Windows platform is a pair of ASCII control characters—a carriage return (CR) followed by a line feed (LF). Unix software, however, just uses the LF character to denote the end of a line.

Not all of the various tools on these operating systems understand files that contain line endings in a format that differs from the native line-ending style of the operating system on which they are running. So, typically, Unix programs treat the CR character present in Windows files as a regular character (usually rendered as ^M), and Windows programs combine all of the lines of a Unix file into one giant line because no carriage return-linefeed (or CRLF) character combination was found to denote the ends of the lines.

This sensitivity to foreign EOL markers can be frustrating for folks who share a file across different operating systems. For example, consider a source code file, and developers that edit this file on both Windows and Unix systems. If all the developers always use tools that preserve the line-ending style of the file, no problems occur.

But in practice, many common tools either fail to properly read a file with foreign EOL markers, or they convert the file's line endings to the native style when the file is saved. If the former is true for a developer, he has to use an external conversion utility (such as dos2unix or its companion, unix2dos) to prepare the file for editing. The latter case requires no extra preparation. But both cases result in a file that differs from the original quite literally on every line! Prior to committing his changes, the user has two choices. Either he can use a conversion utility to restore the modified file to the same line-ending style that it was in before his edits were made. Or, he can simply commit the file—new EOL markers and all.

The result of scenarios like these include wasted time and unnecessary modifications to committed files. Wasted time is painful enough. But when commits change every line in a file, this complicates the job of determining which of those lines were changed in a nontrivial way. Where was that bug really fixed? On what line was a syntax error introduced?

The solution to this problem is the svn:eol-style property. When this property is set to a valid value, Subversion uses it to determine what special processing to perform on the file so that the file's line-ending style isn't flip-flopping with every commit that comes from a different operating system. The valid values are:

native

This causes the file to contain the EOL markers that are native to the operating system on which Subversion was run. In other words, if a user on a Windows machine checks out a working copy that contains a file with an svn:eol-style property set to native, that file will contain CRLF EOL markers. A Unix user checking out a working copy that contains the same file will see LF EOL markers in his copy of the file.

注意Subversion实际上使用LF的EOL标志,而不会考略操作系统,尽管这对用户来说是透明的。

CRLF

这会导致这个文件使用CRLF序列作为EOL标志,不管使用何种操作系统。

LF

这会导致文件使用LF字符作为EOL标志,不管使用何种操作系统。

CR

This causes the file to contain CR characters for EOL markers, regardless of the operating system in use. This line-ending style is not very common. It was used on older Macintosh platforms (on which Subversion doesn't even run).

忽略未版本控制的条目

In any given working copy, there is a good chance that alongside all those versioned files and directories are other files and directories that are neither versioned nor intended to be. Text editors litter directories with backup files. Software compilers generate intermediate—or even final—files that you typically wouldn't bother to version. And users themselves drop various other files and directories wherever they see fit, often in version control working copies.

It's ludicrous to expect Subversion working copies to be somehow impervious to this kind of clutter and impurity. In fact, Subversion counts it as a feature that its working copies are just typical directories, just like unversioned trees. But these not-to-be-versioned files and directories can cause some annoyance for Subversion users. For example, because the svn add and svn import commands act recursively by default and don't know which files in a given tree you do and don't wish to version, it's easy to accidentally add stuff to version control that you didn't mean to. And because svn status reports, by default, every item of interest in a working copy—including unversioned files and directories—its output can get quite noisy where many of these things exist.

So Subversion provides two ways for telling it which files you would prefer for it to simply disregard. One of the ways involves the use of Subversion's runtime configuration system (see “运行配置区”一节), and therefore applies to all the Subversion operations that make use of that runtime configuration—generally those performed on a particular computer or by a particular user of a computer. The other way makes use of Subversion's directory property support and is more tightly bound to the versioned tree itself, and therefore affects everyone who has a working copy of that tree. Both of the mechanisms use file patterns (strings of literal and special wildcard characters used to match against filenames) to decide which files to ignore.

The Subversion runtime configuration system provides an option, global-ignores, whose value is a whitespace-delimited collection of file patterns. The Subversion client checks these patterns against the names of the files that are candidates for addition to version control, as well as to unversioned files that the svn status command notices. If any file's name matches one of the patterns, Subversion will basically act as if the file didn't exist at all. This is really useful for the kinds of files that you almost never want to version, such as editor backup files such as Emacs' *~ and .*~ files.

When found on a versioned directory, the svn:ignore property is expected to contain a list of newline-delimited file patterns that Subversion should use to determine ignorable objects in that same directory. These patterns do not override those found in the global-ignores runtime configuration option, but are instead appended to that list. And it's worth noting again that, unlike the global-ignores option, the patterns found in the svn:ignore property apply only to the directory on which that property is set, and not to any of its subdirectories. The svn:ignore property is a good way to tell Subversion to ignore files that are likely to be present in every user's working copy of that directory, such as compiler output or—to use an example more appropriate to this book—the HTML, PDF, or PostScript files generated as the result of a conversion of some source DocBook XML files to a more legible output format.

注意

Subversion对于忽略文件模式的支持仅限于将未版本化文件和目录添加到版本控制时,如果一个文件已经在Subversion控制下,忽略模式机制不会再有效果,不要期望Subversion会阻止你提交一个符合忽略条件的修改—Subversion一直认为它是版本化的对象。

The global list of ignore patterns tends to be more a matter of personal taste and ties more closely to a user's particular tool chain than to the details of any particular working copy's needs. So, the rest of this section will focus on the svn:ignore property and its uses.

假定你的svn status有如下输出:

$ svn status calc
 M     calc/button.c
?      calc/calculator
?      calc/data.c
?      calc/debug_log
?      calc/debug_log.1
?      calc/debug_log.2.gz
?      calc/debug_log.3.gz

In this example, you have made some property modifications to button.c, but in your working copy, you also have some unversioned files: the latest calculator program that you've compiled from your source code, a source file named data.c, and a set of debugging output log files. Now, you know that your build system always results in the calculator program being generated. [12] And you know that your test suite always leaves those debugging log files lying around. These facts are true for all working copies of this project, not just your own. And you know that you aren't interested in seeing those things every time you run svn status, and you are pretty sure that nobody else is interested in them either. So you use svn propedit svn:ignore calc to add some ignore patterns to the calc directory. For example, you might add this as the new value of the svn:ignore property:

calculator
debug_log*

当你添加完这些属性,你会在calc目录有一个本地修改,但是注意你的svn status输出有什么其他的不同:

$ svn status
 M     calc
 M     calc/button.c
?      calc/data.c

Now, all that cruft is missing from the output! Your calculator compiled program and all those logfiles are still in your working copy; Subversion just isn't constantly reminding you that they are present and unversioned. And now with all the uninteresting noise removed from the display, you are left with more intriguing items—such as that source code file data.c that you probably forgot to add to version control.

当然,不仅仅只有这种简略的工作拷贝状态输出,如果想查看被忽略的文件,可以使用Subversion的--no-ignore选项:

$ svn status --no-ignore
 M     calc
 M     calc/button.c
I      calc/calculator
?      calc/data.c
I      calc/debug_log
I      calc/debug_log.1
I      calc/debug_log.2.gz
I      calc/debug_log.3.gz

我们在前面提到过,svn addsvn import也会使用这个忽略模式列表,这两个操作都包括了询问Subversion来开始管理一组文件和目录。比强制用户挑拣目录树中那个文件要纳入版本控制的方式更好,Subversion使用忽略模式来检测那个文件不应该在大的迭代添加和导入操作中进入版本控制系统。再次说明,操作Subversion文件和目录时你可以使用--no-ignore选项忽略这个忽略列表。

提示

Even if svn:ignore is set, you may run into problems if you use shell wildcards in a command. Shell wildcards are expanded into an explicit list of targets before Subversion operates on them, so running svn SUBCOMMAND * is just like running svn SUBCOMMAND file1 file2 file3 …. In the case of the svn add command, this has an effect similar to passing the --no-ignore option. So instead of using a wildcard, use svn add --force . to do a bulk scheduling of unversioned things for addition. The explicit target will ensure that the current directory isn't overlooked because of being already under version control, and the --force option will cause Subversion to crawl through that directory, adding unversioned files while still honoring the svn:ignore property and global-ignores runtime configuration variable. Be sure to also provide the --depth files option to the svn add command if you don't want a fully recursive crawl for things to add.

关键字替换

Subversion具备添加关键字的能力—一些有用的,关于版本化的文件动态信息的片断—不必直接添加到文件本身。关键字通常会用来描述文件最后一次修改的一些信息,因为这些信息每次都有改变,更重要的一点,这是在文件修改之后,除了版本控制系统,对于任何企图保持数据最新的过程都是一场混乱,作为人类作者,信息变得陈旧是不可避免的。

举个例子,你有一个文档希望显示最后修改的日期,你需要麻烦每个作者提交之前做这件事情,也要修改文档的一部分来描述何时作的修改,但是迟早会有人忘记做这件事,不选择简单的告诉Subversion来执行替换LastChangedDate关键字的操作,你通过在目标位置放置一个keyword anchor来控制关键字插入的位置,这个anchor只是一个格式为$KeywordName$字符串。

所有作为anchor出现在文件里的关键字是大小写敏感的:为了关键字的扩展,你必须使用正确的大写,你必须考虑svn:keywords的属性值也是大小写敏感—特定的关键字名会忽略大小写,但是这个特性已经被废弃了。

Subversion定义了用来替换的关键字列表,这个列表保存了如下五个关键字,有一些也包括了可用的别名:

Date

This keyword describes the last time the file was known to have been changed in the repository, and is of the form $Date: 2006-07-22 21:42:37 -0700 (Sat, 22 Jul 2006) $. It may also be specified as LastChangedDate. Unlike the Id keyword, which uses UTC, the Date keyword displays dates using the local time zone.

Revision

这个关键字描述了这个文件最后一次修改的修订版本,看起来像$Revision: 144 $,也可以通过LastChangedRevision或者Rev引用。

Author

这个关键字描述了最后一个修改这个文件的用户,看起来类似$Author: harry $,也可以用LastChangedBy来指定。

HeadURL

这个关键字描述了这个文件在版本库最新版本的完全URL,看起来类似$HeadURL: http://svn.collab.net/repos/trunk/README $,可以缩写为URL

Id

This keyword is a compressed combination of the other keywords. Its substitution looks something like $Id: calc.c 148 2006-07-28 21:30:43Z sally $, and is interpreted to mean that the file calc.c was last changed in revision 148 on the evening of July 28, 2006 by the user sally. The date displayed by this keyword is in UTC, unlike that of the Date keyword (which uses the local time zone).

Several of the previous descriptions use the phrase “last known” or similar wording. Keep in mind that keyword expansion is a client-side operation, and your client “knows” only about changes that have occurred in the repository when you update your working copy to include those changes. If you never update your working copy, your keywords will never expand to different values even if those versioned files are being changed regularly in the repository.

Simply adding keyword anchor text to your file does nothing special. Subversion will never attempt to perform textual substitutions on your file contents unless explicitly asked to do so. After all, you might be writing a document [13] about how to use keywords, and you don't want Subversion to substitute your beautiful examples of unsubstituted keyword anchors!

To tell Subversion whether to substitute keywords on a particular file, we again turn to the property-related subcommands. The svn:keywords property, when set on a versioned file, controls which keywords will be substituted on that file. The value is a space-delimited list of the keyword names or aliases found in the previous table.

举个例子,假定你有一个版本化的文件weather.txt,内容如下:

Here is the latest report from the front lines.
$LastChangedDate$
$Rev$
Cumulus clouds are appearing more frequently as summer approaches.

当没有svn:keywords属性设置到这个文件,Subversion不会有任何特别操作,现在让我们允许LastChangedDate关键字的替换。

$ svn propset svn:keywords "Date Author" weather.txt
property 'svn:keywords' set on 'weather.txt'
$

Now you have made a local property modification on the weather.txt file. You will see no changes to the file's contents (unless you made some of your own prior to setting the property). Notice that the file contained a keyword anchor for the Rev keyword, yet we did not include that keyword in the property value we set. Subversion will happily ignore requests to substitute keywords that are not present in the file and will not substitute keywords that are not present in the svn:keywords property value.

Immediately after you commit this property change, Subversion will update your working file with the new substitute text. Instead of seeing your keyword anchor $LastChangedDate$, you'll see its substituted result. That result also contains the name of the keyword and continues to be bounded by the dollar sign ($) characters. And as we predicted, the Rev keyword was not substituted because we didn't ask for it to be.

Note also that we set the svn:keywords property to Date Author, yet the keyword anchor used the alias $LastChangedDate$ and still expanded correctly:

Here is the latest report from the front lines.
$LastChangedDate: 2006-07-22 21:42:37 -0700 (Sat, 22 Jul 2006) $
$Rev$
Cumulus clouds are appearing more frequently as summer approaches.

If someone else now commits a change to weather.txt, your copy of that file will continue to display the same substituted keyword value as before—until you update your working copy. At that time, the keywords in your weather.txt file will be resubstituted with information that reflects the most recent known commit to that file.

Subversion 1.2 introduced a new variant of the keyword syntax, which brought additional, useful—though perhaps atypical—functionality. You can now tell Subversion to maintain a fixed length (in terms of the number of bytes consumed) for the substituted keyword. By using a double colon (::) after the keyword name, followed by a number of space characters, you define that fixed width. When Subversion goes to substitute your keyword for the keyword and its value, it will essentially replace only those space characters, leaving the overall width of the keyword field unchanged. If the substituted value is shorter than the defined field width, there will be extra padding characters (spaces) at the end of the substituted field; if it is too long, it is truncated with a special hash (#) character just before the final dollar sign terminator.

例如,你有一篇文档,其中一段是一些反映Subversion关键字的表格数据,使用原始的Subversion关键字替换语法,你的文件或许像这样:

$Rev$:     Revision of last commit
$Author$:  Author of last commit
$Date$:    Date of last commit

现在,表格看起来佷漂亮,但是当你提交文件(当然,关键字替换功能已打开),你会看到:

$Rev: 12 $:     Revision of last commit
$Author: harry $:  Author of last commit
$Date: 2006-03-15 02:33:03 -0500 (Wed, 15 Mar 2006) $:    Date of last commit

The result is not so beautiful. And you might be tempted to then adjust the file after the substitution so that it again looks tabular. But that holds only as long as the keyword values are the same width. If the last committed revision rolls into a new place value (say, from 99 to 100), or if another person with a longer username commits the file, stuff gets all crooked again. However, if you are using Subversion 1.2 or better, you can use the new fixed-length keyword syntax and define some field widths that seem sane so your file might look like this:

$Rev::               $:  Revision of last commit
$Author::            $:  Author of last commit
$Date::              $:  Date of last commit

You commit this change to your file. This time, Subversion notices the new fixed-length keyword syntax and maintains the width of the fields as defined by the padding you placed between the double colon and the trailing dollar sign. After substitution, the width of the fields is completely unchanged—the short values for Rev and Author are padded with spaces, and the long Date field is truncated by a hash character:

$Rev:: 13            $:  Revision of last commit
$Author:: harry      $:  Author of last commit
$Date:: 2006-03-15 0#$:  Date of last commit

固定长度关键字在执行复杂文件格式的替换中非常易用,也可以处理那些很难通过其他程序(例如Microsoft Office文档)进行修改的文件。

警告

Be aware that because the width of a keyword field is measured in bytes, the potential for corruption of multibyte values exists. For example, a username that contains some multibyte UTF-8 characters might suffer truncation in the middle of the string of bytes that make up one of those characters. The result will be a mere truncation when viewed at the byte level, but will likely appear as a string with an incorrect or garbled final character when viewed as UTF-8 text. It is conceivable that certain applications, when asked to load the file, would notice the broken UTF-8 text and deem the entire file corrupt, refusing to operate on the file altogether. So, when limiting keywords to a fixed size, choose a size that allows for this type of byte-wise expansion.

Sparse Directories

By default, most Subversion operations on directories act in a recursive manner. For example, svn checkout creates a working copy with every file and directory in the specified area of the repository, descending recursively through the repository tree until the entire structure is copied to your local disk. Subversion 1.5 introduces a feature called sparse directories (or shallow checkouts) that allows you to easily check out a working copy—or a portion of a working copy—more shallowly than full recursion, with the freedom to bring in previously ignored files and subdirectories at a later time.

For example, say we have a repository with a tree of files and directories with names of the members of a human family with pets. (It's an odd example, to be sure, but bear with us.) A regular svn checkout operation will give us a working copy of the whole tree:

$ svn checkout file:///var/svn/repos mom
A    mom/son
A    mom/son/grandson
A    mom/daughter
A    mom/daughter/granddaughter1
A    mom/daughter/granddaughter1/bunny1.txt
A    mom/daughter/granddaughter1/bunny2.txt
A    mom/daughter/granddaughter2
A    mom/daughter/fishie.txt
A    mom/kitty1.txt
A    mom/doggie1.txt
Checked out revision 1.
$

Now, let's check out the same tree again, but this time, we'll ask Subversion to give us only the top-most directory with none of its children at all:

$ svn checkout file:///var/svn/repos mom-empty --depth empty
Checked out revision 1
$

Notice that we added to our original svn checkout command line a new --depth option. This option is present on many of Subversion's subcommands and is similar to the --non-recursive (-N) and --recursive (-R) options. In fact, it combines, improves upon, supercedes, and ultimately obsoletes these two older options. For starters, it expands the supported degrees of depth specification available to users, adding some previously unsupported (or inconsistently supported) depths. Here are the depth values that you can request for a given Subversion operation:

--depth empty

Include only the immediate target of the operation, not any of its file or directory children.

--depth files

Include the immediate target of the operation and any of its immediate file children.

--depth immediates

Include the immediate target of the operation and any of its immediate file or directory children. The directory children will themselves be empty.

--depth infinity

Include the immediate target, its file and directory children, its children's children, and so on to full recursion.

Of course, merely combining two existing options into one hardly constitutes a new feature worthy of a whole section in our book. Fortunately, there is more to this story. This idea of depth extends not just to the operations you perform with your Subversion client, but also as a description of a working copy citizen's ambient depth, which is the depth persistently recorded by the working copy for that item. Its key strength is this very persistence—the fact that it is sticky. The working copy remembers the depth you've selected for each item in it until you later change that depth selection; by default, Subversion commands operate on the working copy citizens present, regardless of their selected depth settings.

提示

You can check the recorded ambient depth of a working copy using the svn info command. If the ambient depth is anything other than infinite recursion, svn info will display a line describing that depth value:

$ svn info mom-immediates | grep '^Depth:'
Depth: immediates
$

Our previous examples demonstrated checkouts of infinite depth (the default for svn checkout) and empty depth. Let's look now at examples of the other depth values:

$ svn checkout file:///var/svn/repos mom-files --depth files
A    mom-files/kitty1.txt
A    mom-files/doggie1.txt
Checked out revision 1.
$ svn checkout file:///var/svn/repos mom-immediates --depth immediates
A    mom-immediates/son
A    mom-immediates/daughter
A    mom-immediates/kitty1.txt
A    mom-immediates/doggie1.txt
Checked out revision 1.
$

As described, each of these depths is something more than only-the-target, but something less than full recursion.

We've used svn checkout as an example here, but you'll find the --depth option present on many other Subversion commands, too. In those other commands, depth specification is a way to limit the scope of an operation to some depth, much like the way the older --non-recursive (-N) and --recursive (-R) options behave. This means that when operating on a working copy of some depth, while requesting an operation of a shallower depth, the operation is limited to that shallower depth. In fact, we can make an even more general statement: given a working copy of any arbitrary—even mixed—ambient depth, and a Subversion command with some requested operational depth, the command will maintain the ambient depth of the working copy members while still limiting the scope of the operation to the requested (or default) operational depth.

In addition to the --depth option, the svn update and svn switch subcommands also accept a second depth-related option: --set-depth. It is with this option that you can change the sticky depth of a working copy item. Watch what happens as we take our empty-depth checkout and gradually telescope it deeper using svn update --set-depth:

$ svn update --set-depth files mom-empty
A    mom-empty/kittie1.txt
A    mom-empty/doggie1.txt
Updated to revision 1.
$ svn update --set-depth immediates mom-empty
A    mom-empty/son
A    mom-empty/daughter
Updated to revision 1.
$ svn update --set-depth infinity mom-empty
A    mom-empty/son/grandson
A    mom-empty/daughter/granddaughter1
A    mom-empty/daughter/granddaughter1/bunny1.txt
A    mom-empty/daughter/granddaughter1/bunny2.txt
A    mom-empty/daughter/granddaughter2
A    mom-empty/daughter/fishie1.txt
Updated to revision 1.
$

As we gradually increased our depth selection, the repository gave us more pieces of our tree.

In our example, we operated only on the root of our working copy, changing its ambient depth value. But we can independently change the ambient depth value of any subdirectory inside the working copy, too. Careful use of this ability allows us to flesh out only certain portions of the working copy tree, leaving other portions absent altogether (hence the “sparse” bit of the feature's name). Here's an example of how we might build out a portion of one branch of our family's tree, enable full recursion on another branch, and keep still other pieces pruned (absent from disk).

$ rm -rf mom-empty
$ svn checkout file:///var/svn/repos mom-empty --depth empty
Checked out revision 1.
$ svn update --set-depth empty mom-empty/son
A    mom-empty/son
Updated to revision 1.
$ svn update --set-depth empty mom-empty/daughter
A    mom-empty/daughter
Updated to revision 1.
$ svn update --set-depth infinity mom-empty/daughter/granddaughter1
A    mom-empty/daughter/granddaughter1
A    mom-empty/daughter/granddaughter1/bunny1.txt
A    mom-empty/daughter/granddaughter1/bunny2.txt
Updated to revision 1.
$

Fortunately, having a complex collection of ambient depths in a single working copy doesn't complicate the way you interact with that working copy. You can still make, revert, display, and commit local modifications in your working copy without providing any new options (including --depth or --set-depth) to the relevant subcommands. Even svn update works as it does elsewhere when no specific depth is provided—it updates the working copy targets that are present while honoring their sticky depths.

You might at this point be wondering, “So what? When would I use this?” One scenario where this feature finds utility is tied to a particular repository layout, specifically where you have many related or codependent projects or software modules living as siblings in a single repository location (trunk/project1, trunk/project2, trunk/project3, and so on). In such scenarios, it might be the case that you personally care only about a handful of those projects—maybe some primary project and a few other modules on which it depends. You can check out individual working copies of all of these things, but those working copies are disjoint and, as a result, it can be cumbersome to perform operations across several or all of them at the same time. The alternative is to use the sparse directories feature, building out a single working copy that contains only the modules you care about. You'd start with an empty-depth checkout of the common parent directory of the projects, and then update with infinite depth only the items you wish to have, like we demonstrated in the previous example. Think of it like an opt-in system for working copy citizens.

Subversion 1.5's implementation of shallow checkouts is good but does not support a couple of interesting behaviors. First, you cannot de-telescope a working copy item. Running svn update --set-depth empty on an infinite-depth working copy will not have the effect of discarding everything but the top-most directory—it will simply error out. Secondly, there is no depth value to indicate that you wish an item to be explicitly excluded. You have to do implicit exclusion of an item by including everything else.

锁定

Subversion's copy-modify-merge version control model lives and dies on its data merging algorithms—specifically on how well those algorithms perform when trying to resolve conflicts caused by multiple users modifying the same file concurrently. Subversion itself provides only one such algorithm: a three-way differencing algorithm that is smart enough to handle data at a granularity of a single line of text. Subversion also allows you to supplement its content merge processing with external differencing utilities (as described in “外置 diff3”一节), some of which may do an even better job, perhaps providing granularity of a word or a single character of text. But common among those algorithms is that they generally work only on text files. The landscape starts to look pretty grim when you start talking about content merges of non-textual file formats. And when you can't find a tool that can handle that type of merging, you begin to run into problems with the copy-modify-merge model.

让我们看一个使用这个模型的真实例子,Harry和Sally是同一个项目的图形设计师,汽车技工的间接营销。海报的设计一个小车,需要一些主要部分的工作,使用PNG文件格式。海报的布局几乎完成,Harry和Sally都看上了一个从损坏小车得到的特别照片—一个1967的淡蓝色的Ford Mustang,挡泥板有一些溅迹。

Now, as is common in graphic design work, there's a change in plans, which causes the car's color to be a concern. So Sally updates her working copy to HEAD, fires up her photo-editing software, and sets about tweaking the image so that the car is now cherry red. Meanwhile, Harry, feeling particularly inspired that day, decides that the image would have greater impact if the car also appears to have suffered greater impact. He, too, updates to HEAD, and then draws some cracks on the vehicle's windshield. He manages to finish his work before Sally finishes hers, and after admiring the fruits of his undeniable talent, commits the modified image. Shortly thereafter, Sally is finished with the car's new finish and tries to commit her changes. But, as expected, Subversion fails the commit, informing Sally that her version of the image is now out of date.

Here's where the difficulty sets in. If Harry and Sally were making changes to a text file, Sally would simply update her working copy, receiving Harry's changes in the process. In the worst possible case, they would have modified the same region of the file, and Sally would have to work out by hand the proper resolution to the conflict. But these aren't text files—they are binary images. And while it's a simple matter to describe what one would expect the results of this content merge to be, there is precious little chance that any software exists that is smart enough to examine the common baseline image that each of these graphic artists worked against, the changes that Harry made, and the changes that Sally made, and then spit out an image of a busted-up red Mustang with a cracked windshield!

Of course, things would have gone more smoothly if Harry and Sally had serialized their modifications to the image—if, say, Harry had waited to draw his windshield cracks on Sally's now-red car, or if Sally had tweaked the color of a car whose windshield was already cracked. As is discussed in “拷贝-修改-合并 方案”一节, most of these types of problems go away entirely where perfect communication between Harry and Sally exists. [14] But as one's version control system is, in fact, one form of communication, it follows that having that software facilitate the serialization of nonparallelizable editing efforts is no bad thing. This is where Subversion's implementation of the lock-modify-unlock model steps into the spotlight. This is where we talk about Subversion's locking feature, which is similar to the “reserved checkouts” mechanisms of other version control systems.

Subversion 的锁定特性为两个主要目的服务:

  1. 顺序访问资源。允许用户得到一个排他的修改文件权,这个用户可以确定不可合并的修改不会被浪费—他对这个修改的提交会成功。

  2. 辅助交流。通过要求用户对某个版本化对象串行工作,用户可以知道对象正在被别人修改,这样可以防止浪费精力和时间去修改一个不可合并和提交的对象。

When referring to Subversion's locking feature, one is actually talking about a fairly diverse collection of behaviors, which include the ability to lock a versioned file [15] (claiming the exclusive right to modify the file), to unlock that file (yielding that exclusive right to modify), to see reports about which files are locked and by whom, to annotate files for which locking before editing is strongly advised, and so on. In this section, we'll cover all of these facets of the larger locking feature.

Creating Locks

In the Subversion repository, a lock is a piece of metadata that grants exclusive access to one user to change a file. This user is said to be the lock owner. Each lock also has a unique identifier, typically a long string of characters, known as the lock token. The repository manages locks, ultimately handling their creation, enforcement, and removal. If any commit transaction attempts to modify or delete a locked file (or delete one of the parent directories of the file), the repository will demand two pieces of information—that the client performing the commit be authenticated as the lock owner, and that the lock token has been provided as part of the commit process as a sort of proof that client knows which lock it is using.

To demonstrate lock creation, let's refer back to our example of multiple graphic designers working on the same binary image files. Harry has decided to change a JPEG image. To prevent other people from committing changes to the file while he is modifying it (as well as alerting them that he is about to change it), he locks the file in the repository using the svn lock command.

$ svn lock banana.jpg -m "Editing file for tomorrow's release."
'banana.jpg' locked by user 'harry'.
$

There are a number of new things demonstrated in the previous example. First, notice that Harry passed the --message (-m) option to svn lock. Similar to svn commit, the svn lock command can take comments (either via --message (-m) or --file (-F) to describe the reason for locking the file. Unlike svn commit, however, svn lock will not demand a message by launching your preferred text editor. Lock comments are optional, but still recommended to aid communication.

Secondly, the lock attempt succeeded. This means that the file wasn't already locked, and that Harry had the latest version of the file. If Harry's working copy of the file had been out of date, the repository would have rejected the request, forcing Harry to svn update and reattempt the locking command. The locking command would also have failed if the file had already been locked by someone else.

就像你看到的,svn lock打印了锁定成功的确认信息。此时,通过svn statussvn info的输出我们可以看到文件已经锁定。

$ svn status
     K banana.jpg

$ svn info banana.jpg
Path: banana.jpg
Name: banana.jpg
URL: http://svn.example.com/repos/project/banana.jpg
Repository UUID: edb2f264-5ef2-0310-a47a-87b0ce17a8ec
Revision: 2198
Node Kind: file
Schedule: normal
Last Changed Author: frank
Last Changed Rev: 1950
Last Changed Date: 2006-03-15 12:43:04 -0600 (Wed, 15 Mar 2006)
Text Last Updated: 2006-06-08 19:23:07 -0500 (Thu, 08 Jun 2006)
Properties Last Updated: 2006-06-08 19:23:07 -0500 (Thu, 08 Jun 2006)
Checksum: 3b110d3b10638f5d1f4fe0f436a5a2a5
Lock Token: opaquelocktoken:0c0f600b-88f9-0310-9e48-355b44d4a58e
Lock Owner: harry
Lock Created: 2006-06-14 17:20:31 -0500 (Wed, 14 Jun 2006)
Lock Comment (1 line):
Editing file for tomorrow's release.

$

The fact that the svn info command, which does not contact the repository when run against working copy paths, can display the lock token reveals an important piece of information about those tokens: they are cached in the working copy. The presence of the lock token is critical. It gives the working copy authorization to make use of the lock later on. Also, the svn status command shows a K next to the file (short for locKed), indicating that the lock token is present.

现在Harry已经锁定了banana.jpg,Sally不能修改或删除这个文件:

$ svn delete banana.jpg
D         banana.jpg
$ svn commit -m "Delete useless file."
Deleting       banana.jpg
svn: Commit failed (details follow):
svn: Server sent unexpected return value (423 Locked) in response to DELETE\
 request for '/repos/project/!svn/wrk/64bad3a9-96f9-0310-818a-df4224ddc35d/\
banana.jpg'
$

But Harry, after touching up the banana's shade of yellow, is able to commit his changes to the file. That's because he authenticates as the lock owner and also because his working copy holds the correct lock token:

$ svn status
M    K banana.jpg
$ svn commit -m "Make banana more yellow"
Sending        banana.jpg
Transmitting file data .
Committed revision 2201.
$ svn status
$

Notice that after the commit is finished, svn status shows that the lock token is no longer present in working copy. This is the standard behavior of svn commit—it searches the working copy (or list of targets, if you provide such a list) for local modifications and sends all the lock tokens it encounters during this walk to the server as part of the commit transaction. After the commit completes successfully, all of the repository locks that were mentioned are released—even on files that weren't committed. This is meant to discourage users from being sloppy about locking or from holding locks for too long. If Harry haphazardly locks 30 files in a directory named images because he's unsure of which files he needs to change, yet changes only 4 of those files, when he runs svn commit images, the process will still release all 30 locks.

自动释放锁定的特性可以通过svn commit--no-unlock选项关闭,当你要提交文件,同时期望继续修改而必须保留锁定时非常有用。这个特性也可以半永久性的设定,方法是设置运行中config文件(见“运行配置区”一节)的no-unlock = yes

当然,锁定一个文件不会强制一个人要提交修改,任何时候都可以通过运行svn unlock命令释放锁定:

$ svn unlock banana.c
'banana.c' unlocked.

Discovering Locks

最明显的方式就是因为锁定而不能提交一个文件,最简单的方式是svn status --show-updates

$ svn status -u
M              23   bar.c
M    O         32   raisin.jpg
       *       72   foo.h
Status against revision:     105
$

In this example, Sally can see not only that her copy of foo.h is out of date, but that one of the two modified files she plans to commit is locked in the repository. The O symbol stands for “Other,” meaning that a lock exists on the file and was created by somebody else. If she were to attempt a commit, the lock on raisin.jpg would prevent it. Sally is left wondering who made the lock, when, and why. Once again, svn info has the answers:

$ svn info http://svn.example.com/repos/project/raisin.jpg
Path: raisin.jpg
Name: raisin.jpg
URL: http://svn.example.com/repos/project/raisin.jpg
Repository UUID: edb2f264-5ef2-0310-a47a-87b0ce17a8ec
Revision: 105
Node Kind: file
Last Changed Author: sally
Last Changed Rev: 32
Last Changed Date: 2006-01-25 12:43:04 -0600 (Sun, 25 Jan 2006)
Lock Token: opaquelocktoken:fc2b4dee-98f9-0310-abf3-653ff3226e6b
Lock Owner: harry
Lock Created: 2006-02-16 13:29:18 -0500 (Thu, 16 Feb 2006)
Lock Comment (1 line):
Need to make a quick tweak to this image.
$

就像svn info可以检验工作拷贝的对象,它也可以检验版本库的对象,如果svn info的主要参数是工作拷贝路径,所有工作拷贝的缓存信息都会显示,发现了锁定就意味着工作拷贝拥有锁定令牌(如果一个文件被另一个用户在另一个工作拷贝锁定,工作拷贝路径上运行svn info不会显示锁定信息)。如果svn info的主参数是URL,就会反映版本库中最新版本的对象信息,任何对锁定的提及描述了当前对象的锁定。

So in this particular example, Sally can see that Harry locked the file on February 16th to “make a quick tweak.” It being June, she suspects that he probably forgot all about the lock. She might phone Harry to complain and ask him to release the lock. If he's unavailable, she might try to forcibly break the lock herself or ask an administrator to do so.

Breaking and Stealing Locks

A repository lock isn't sacred—in Subversion's default configuration state, locks can be released not only by the person who created them, but by anyone. When somebody other than the original lock creator destroys a lock, we refer to this as breaking the lock.

从管理员的位子上很容易打破锁定,svnlooksvnadmin程序都有能力从版本库直接显示和删除锁定。(关于这些工具的信息可以看“管理员的工具箱”一节。)

$ svnadmin lslocks /var/svn/repos
Path: /project2/images/banana.jpg
UUID Token: opaquelocktoken:c32b4d88-e8fb-2310-abb3-153ff1236923
Owner: frank
Created: 2006-06-15 13:29:18 -0500 (Thu, 15 Jun 2006)
Expires: 
Comment (1 line):
Still improving the yellow color.

Path: /project/raisin.jpg
UUID Token: opaquelocktoken:fc2b4dee-98f9-0310-abf3-653ff3226e6b
Owner: harry
Created: 2006-02-16 13:29:18 -0500 (Thu, 16 Feb 2006)
Expires: 
Comment (1 line):
Need to make a quick tweak to this image.

$ svnadmin rmlocks /var/svn/repos /project/raisin.jpg
Removed lock on '/project/raisin.jpg'.
$

The more interesting option is allowing users to break each other's locks over the network. To do this, Sally simply needs to pass the --force to the svn unlock command:

$ svn status -u
M              23   bar.c
M    O         32   raisin.jpg
       *       72   foo.h
Status against revision:     105
$ svn unlock raisin.jpg
svn: 'raisin.jpg' is not locked in this working copy
$ svn info raisin.jpg | grep URL
URL: http://svn.example.com/repos/project/raisin.jpg
$ svn unlock http://svn.example.com/repos/project/raisin.jpg
svn: Unlock request failed: 403 Forbidden (http://svn.example.com)
$ svn unlock --force http://svn.example.com/repos/project/raisin.jpg
'raisin.jpg' unlocked.
$

Sally初始的unlock命令失败了,因为她直接在自己的工作拷贝上运行了svn unlock,而这里没有锁定令牌。为了直接从版本库删除锁定,她需要给svn unlock传递URL参数,她的这一次尝试又失败了,因为她不是锁定的拥有者(也没有锁定令牌)。当她使用了--force选项后,认证和授权的要求被忽略了,远程的锁定被打破了。

当然,简单的打破锁定也许还不够,在这个例子里,Sally不仅想要打破Harry遗忘的锁定,她也希望自己重新锁定。她可以通过运行svn unlock --force紧接着svn lock,但是有可能有人在这两次命令之间锁定了文件,最简单的方式是窃取这个锁定,将打破和重新锁定变成一种原子操作,为此需要运行svn lock--force选项:

$ svn lock raisin.jpg
svn: Lock request failed: 423 Locked (http://svn.example.com)
$ svn lock --force raisin.jpg
'raisin.jpg' locked by user 'sally'.
$

在任何情况下,无论锁定被打破还是窃取,Harry都会感到惊讶。Harry的工作拷贝还保留有原来的锁定令牌,但是锁定已经不存在了,锁定令牌可以说已经死掉了。锁定令牌指代的锁定被打破(版本库中不再存在)或者是窃取了(被另一个锁定代替了),任何一种情况下,Harry都可以使用svn status询问版本库:

$ svn status
     K raisin.jpg
$ svn status -u
     B         32   raisin.jpg
$ svn update
  B  raisin.jpg
$ svn status
$

如果版本库锁定被打破了,svn status --show-updates会在文件旁边显示一个B (Broken)。如果有一个新的锁,就会显示一个T (sTolen)符号。最终,svn update会注意到所有死掉的锁定并且把它们从工作拷贝中删除掉。

锁定交流

我们已经见到了如何利用svn locksvn unlock来创建、释放、打破和窃取锁定,这就满足了顺序访问文件的要求,但是浪费时间这个大问题该如何呢?

For example, suppose Harry locks an image file and then begins editing it. Meanwhile, miles away, Sally wants to do the same thing. She doesn't think to run svn status --show-updates, so she has no idea that Harry has already locked the file. She spends hours editing the file, and when she tries to commit her change, she discovers that either the file is locked or that she's out of date. Regardless, her changes aren't mergeable with Harry's. One of these two people has to throw away their work, and a lot of time has been wasted.

Subversion's solution to this problem is to provide a mechanism to remind users that a file ought to be locked before the editing begins. The mechanism is a special property: svn:needs-lock. If that property is attached to a file (regardless of its value, which is irrelevant), then Subversion will try to use filesystem-level permissions to make the file read-only—unless, of course, the user has explicitly locked the file. When a lock token is present (as a result of running svn lock), the file becomes read-write. When the lock is released, the file becomes read-only again.

根据这个原理,如果一个图像文件有这个属性,Sally打开编辑文件就会立刻注意到有些特别,大多数程序会在打开只读文件时立刻警告,至少所有的程序会防止她保存修改,这提醒了她编辑之前需要锁定文件,这样她就发现了原来存在的锁定:

$ /usr/local/bin/gimp raisin.jpg
gimp: error: file is read-only!
$ ls -l raisin.jpg
-r--r--r--   1 sally   sally   215589 Jun  8 19:23 raisin.jpg
$ svn lock raisin.jpg
svn: Lock request failed: 423 Locked (http://svn.example.com)
$ svn info http://svn.example.com/repos/project/raisin.jpg | grep Lock
Lock Token: opaquelocktoken:fc2b4dee-98f9-0310-abf3-653ff3226e6b
Lock Owner: harry
Lock Created: 2006-06-08 07:29:18 -0500 (Thu, 08 June 2006)
Lock Comment (1 line):
Making some tweaks.  Locking for the next two hours.
$

提示

Users and administrators alike are encouraged to attach the svn:needs-lock property to any file that cannot be contextually merged. This is the primary technique for encouraging good locking habits and preventing wasted effort.

Note that this property is a communication tool that works independently from the locking system. In other words, any file can be locked, whether or not this property is present. And conversely, the presence of this property doesn't make the repository require a lock when committing.

这个系统并不是毫无瑕疵,即使有这个属性,只读提醒也有可能失效。有些程序“偷偷的篡改了”文件的只读属性,悄无声息的允许用户编辑和保存文件,不幸的是,Subversion对此无能为力—即使到了现今,还是没有任何工具能够代替人与人的良好交流。[16]

外部定义

Sometimes it is useful to construct a working copy that is made out of a number of different checkouts. For example, you may want different subdirectories to come from different locations in a repository or perhaps from different repositories altogether. You could certainly set up such a scenario by hand—using svn checkout to create the sort of nested working copy structure you are trying to achieve. But if this layout is important for everyone who uses your repository, every other user will need to perform the same checkout operations that you did.

Fortunately, Subversion provides support for externals definitions. An externals definition is a mapping of a local directory to the URL—and ideally a particular revision—of a versioned directory. In Subversion, you declare externals definitions in groups using the svn:externals property. You can create or modify this property using svn propset or svn propedit (see “操作属性”一节). It can be set on any versioned directory, and its value describes both the external repository location and the client-side directory to which that location should be checked out.

The convenience of the svn:externals property is that once it is set on a versioned directory, everyone who checks out a working copy with that directory also gets the benefit of the externals definition. In other words, once one person has made the effort to define the nested working copy structure, no one else has to bother—Subversion will, after checking out the original working copy, automatically also check out the external working copies.

警告

外部定义的相对目标子目录不需要存在于你的或其它用户的系统中—Subversion会在检出工作拷贝时创建这些文件。实际上,你一定不要使用外部定义来产生已经在版本控制的路径。

You also get in the externals definition design all the regular benefits of Subversion properties. The definitions are versioned. If you need to change an externals definition, you can do so using the regular property modification subcommands. When you commit a change to the svn:externals property, Subversion will synchronize the checked-out items against the changed externals definition when you next run svn update. The same thing will happen when others update their working copies and receive your changes to the externals definition.

提示

因为svn:externals的值是多行的,所以我们强烈建议使用svn propedit,而不是使用svn propset

Subversion releases prior to 1.5 honor an externals definition format that is a multiline table of subdirectories (relative to the versioned directory on which the property is set), optional revision flags, and fully qualified, absolute Subversion repository URLs. An example of this might looks as follows:

$ svn propget svn:externals calc
third-party/sounds             http://svn.example.com/repos/sounds
third-party/skins -r148        http://svn.example.com/skinproj
third-party/skins/toolkit -r21 http://svn.example.com/skin-maker

When someone checks out a working copy of the calc directory referred to in the previous example, Subversion also continues to check out the items found in its externals definition.

$ svn checkout http://svn.example.com/repos/calc
A  calc
A  calc/Makefile
A  calc/integer.c
A  calc/button.c
Checked out revision 148.

Fetching external item into calc/third-party/sounds
A  calc/third-party/sounds/ding.ogg
A  calc/third-party/sounds/dong.ogg
A  calc/third-party/sounds/clang.ogg
…
A  calc/third-party/sounds/bang.ogg
A  calc/third-party/sounds/twang.ogg
Checked out revision 14.

Fetching external item into calc/third-party/skins
…

As of Subversion 1.5, though, a new format of the svn:externals property is supported. Externals definitions are still multiline, but the order and format of the various pieces of information has changed. The new syntax more closely mimics the order of arguments you might pass to svn checkout: the optional revision flags come first, then the external Subversion repository URL, and finally the relative local subdirectory. Notice, though, that this time we didn't say “fully qualified, absolute Subversion repository URLs.” That's because the new format supports relative URLs and URLs that carry peg revisions. The previous example of an externals definition might, in Subversion 1.5, look like the following:

$ svn propget svn:externals calc
      http://svn.example.com/repos/sounds third-party/sounds
-r148 http://svn.example.com/skinproj third-party/skins
-r21  http://svn.example.com/skin-maker third-party/skins/toolkit

Or, making use of the peg revision syntax (which we describe in detail in “Peg和实施修订版本”一节), it might appear as:

$ svn propget svn:externals calc
http://svn.example.com/repos/sounds third-party/sounds
http://svn.example.com/skinproj@148 third-party/skins
http://svn.example.com/skin-maker@21 third-party/skins/toolkit

提示

You should seriously consider using explicit revision numbers in all of your externals definitions. Doing so means that you get to decide when to pull down a different snapshot of external information, and exactly which snapshot to pull. Besides avoiding the surprise of getting changes to third-party repositories that you might not have any control over, using explicit revision numbers also means that as you backdate your working copy to a previous revision, your externals definitions will also revert to the way they looked in that previous revision, which in turn means that the external working copies will be updated to match the way they looked back when your repository was at that previous revision. For software projects, this could be the difference between a successful and a failed build of an older snapshot of your complex codebase.

For most repositories, these three ways of formatting the external definitions have the same ultimate effect. They all bring the same benefits. Unfortunately, they all bring the same annoyances, too. Since the definitions shown use absolute URLs, moving or copying a directory to which they are attached will not affect what gets checked out as an external (though the relative local target subdirectory will, of course, move with renamed directory). This can be confusing—even frustrating—in certain situations. For example, say you have a top-level directory named my-project, and you've created an externals definition on one of its subdirectories (my-project/some-dir) that tracks the latest revision of another of its subdirectories (my-project/external-dir).

$ svn checkout http://svn.example.com/projects .
A    my-project
A    my-project/some-dir
A    my-project/external-dir
…
Fetching external item into 'my-project/some-dir/subdir'
Checked out external at revision 11.

Checked out revision 11.
$ svn propget svn:externals my-project/some-dir
subdir http://svn.example.com/projects/my-project/external-dir

$

现在你使用svn move将目录my-project改名,此刻,你的外部定义还是指向my-project目录,即使这个目录已经不存在了。

$ svn move -q my-project renamed-project
$ svn commit -m "Rename my-project to renamed-project."
Deleting       my-project
Adding         my-renamed-project

Committed revision 12.
$ svn update

Fetching external item into 'renamed-project/some-dir/subdir'
svn: Target path does not exist
$

Also, absolute URLs can cause problems with repositories that are available via multiple URL schemes. For example, if your Subversion server is configured to allow everyone to check out the repository over http:// or https://, but only allow commits to come in via https://, you have an interesting problem on your hands. If your externals definitions use the http:// form of the repository URLs, you won't be able to commit anything from the working copies created by those externals. On the other hand, if they use the https:// form of the URLs, anyone who might be checking out via http:// because their client doesn't support https:// will be unable to fetch the external items. Be aware, too, that if you need to re-parent your working copy (using svn switch --relocate), externals definitions will not also be re-parented.

Subversion 1.5 takes a huge step in relieving these frustrations. As mentioned earlier, the URLs used in the new externals definition format can be relative, and Subversion provides syntax magic for specifying multiple flavors of URL relativity.

../

Relative to the URL of the directory on which the svn:externals property is set.

^/

Relative to the root of the repository in which the svn:externals property is versioned.

//

Relative to the scheme of the URL of the directory on which the svn:externals property is set.

/

Relative to the root URL of the server on which the svn:externals property is versioned.

So, looking a fourth time at our previous externals definition example, and making use of the new absolute URL syntax in various ways, we might now see:

$ svn propget svn:externals calc
^/sounds third-party/sounds
/skinproj@148 third-party/skins
//svn.example.com/skin-maker@21 third-party/skins/toolkit

The support that exists for externals definitions in Subversion remains less than ideal, though. An externals definition can only point to directories, not files. Also, the local subdirectory part of the definition cannot contain .. parent directory indicators (such as ../../skins/myskin). Perhap most disappointinly, the working copies created via the externals definition support are still disconnected from the primary working copy (on whose versioned directories the svn:externals property was actually set). And Subversion still only truly operates on nondisjoint working copies. So, for example, if you want to commit changes that you've made in one or more of those external working copies, you must run svn commit explicitly on those working copies—committing on the primary working copy will not recurse into any external ones.

We've already mentioned some of the additional shortcomings of the old svn:externals format and how the new Subversion 1.5 format improves upon it. But be careful when making use of the new format that you don't inadvertantly cause problems for other folks accessing your repository who are using older Subversion clients. While Subversion 1.5 clients will continue to recognize and support the original externals definition format, older clients will not be able to correctly parse the new format.

Besides the svn checkout, svn update, svn switch, and svn export commands which actually manage the disjoint (or disconnected) subdirectories into which externals are checked out, the svn status command also recognizes externals definitions. It displays a status code of X for the disjoint external subdirectories, and then recurses into those subdirectories to display the status of the external items themselves. You can pass the --ignore-externals option to any of these subcommands to disable externals definition processing.

Peg和实施修订版本

We copy, move, rename, and completely replace files and directories on our computers all the time. And your version control system shouldn't get in the way of your doing these things with your version-controlled files and directories, either. Subversion's file management support is quite liberating, affording almost as much flexibility for versioned files as you'd expect when manipulating your unversioned ones. But that flexibility means that across the lifetime of your repository, a given versioned object might have many paths, and a given path might represent several entirely different versioned objects. This introduces a certain level of complexity to your interactions with those paths and objects.

Subversion is pretty smart about noticing when an object's version history includes such “changes of address.” For example, if you ask for the revision history log of a particular file that was renamed last week, Subversion happily provides all those logs—the revision in which the rename itself happened, plus the logs of relevant revisions both before and after that rename. So, most of the time, you don't even have to think about such things. But occasionally, Subversion needs your help to clear up ambiguities.

The simplest example of this occurs when a directory or file is deleted from version control, and then a new directory or file is created with the same name and added to version control. The thing you deleted and the thing you later added aren't the same thing. They merely happen to have had the same path—/trunk/object, for example. What, then, does it mean to ask Subversion about the history of /trunk/object? Are you asking about the thing currently at that location, or the old thing you deleted from that location? Are you asking about the operations that have happened to all the objects that have ever lived at that path? Subversion needs a hint about what you really want.

And thanks to moves, versioned object history can get far more twisted than that, even. For example, you might have a directory named concept, containing some nascent software project you've been toying with. Eventually, though, that project matures to the point that the idea seems to actually have some wings, so you do the unthinkable and decide to give the project a name. [17] Let's say you called your software Frabnaggilywort. At this point, it makes sense to rename the directory to reflect the project's new name, so concept is renamed to frabnaggilywort. Life goes on, Frabnaggilywort releases a 1.0 version and is downloaded and used daily by hordes of people aiming to improve their lives.

这是一个美好的故事,但是没有在这里结束,作为主办人,你一定想到了另一件事,所以你创建了一个目录叫做concept,周期重新开始。实际上,这个循环在几年里开始了多次,每一个想法从使用旧的concept目录开始,然后有时在想法成熟之后重新命名,有时你放弃了这个注意而删除了这个目录。或者更加变态一点,或许你把concept改成其他名字之后又因为一些原因重新改回concept

In scenarios like these, attempting to instruct Subversion to work with these re-used paths can be a little like instructing a motorist in Chicago's West Suburbs to drive east down Roosevelt Road and turn left onto Main Street. In a mere 20 minutes, you can cross “Main Street” in Wheaton, Glen Ellyn, and Lombard. And no, they aren't the same street. Our motorist—and our Subversion—need a little more detail in order to do the right thing.

In version 1.1, Subversion introduced a way for you to tell it exactly which Main Street you meant. It's called the peg revision, and it is provided to Subversion for the sole purpose of identifying a unique line of history. Because at most, one versioned object may occupy a path at any given time—or, more precisely, in any one revision—the combination of a path and a peg revision is all that is needed to refer to a specific line of history. Peg revisions are specified to the Subversion command-line client using at syntax, so called because the syntax involves appending an “at sign” (@) and the peg revision to the end of the path with which the revision is associated.

But what of the --revision (-r) of which we've spoken so much in this book? That revision (or set of revisions) is called the operative revision (or operative revision range). Once a particular line of history has been identified using a path and peg revision, Subversion performs the requested operation using the operative revision(s). To map this to our Chicagoland streets analogy, if we are told to go to 606 N. Main Street in Wheaton, [18] we can think of “Main Street” as our path and “Wheaton” as our peg revision. These two pieces of information identify a unique path that can be travelled (north or south on Main Street), and they keep us from travelling up and down the wrong Main Street in search of our destination. Now we throw in “606 N.” as our operative revision of sorts, and we know exactly where to go.

Say that long ago we created our repository, and in revision 1 added our first concept directory, plus an IDEA file in that directory talking about the concept. After several revisions in which real code was added and tweaked, we, in revision 20, renamed this directory to frabnaggilywort. By revision 27, we had a new concept, a new concept directory to hold it, and a new IDEA file to describe it. And then 5 years and 20 thousand revisions flew by, just like they would in any good romance story.

Now, years later, we wonder what the IDEA file looked like back in revision 1. But Subversion needs to know if we are asking about how the current file looked back in revision 1, or if we are asking for the contents of whatever file lived at concepts/IDEA in revision 1. Certainly those questions have different answers, and because of peg revisions, you can ask question. To find out how the current IDEA file looked in that old revision, you run:

$ svn cat -r 1 concept/IDEA 
svn: Unable to find repository location for 'concept/IDEA' in revision 1

Of course, in this example, the current IDEA file didn't exist yet in revision 1, so Subversion gives an error. The previous command is shorthand for a longer notation which explicitly lists a peg revision. The expanded notation is:

$ svn cat -r 1 concept/IDEA@BASE
svn: Unable to find repository location for 'concept/IDEA' in revision 1

当执行时,它包含期望的结果。

The perceptive reader is probably wondering at this point if the peg revision syntax causes problems for working copy paths or URLs that actually have at signs in them. After all, how does svn know whether news@11 is the name of a directory in my tree or just a syntax for “revision 11 of news”? Thankfully, while svn will always assume the latter, there is a trivial workaround. You need only append an at sign to the end of the path, such as news@11@. svn cares only about the last at sign in the argument, and it is not considered illegal to omit a literal peg revision specifier after that at sign. This workaround even applies to paths that end in an at sign—you would use filename@@ to talk about a file named filename@.

然后让我们询问另一个问题—在修订版本1 ,占据concepts/IDEA路径的文件的内容到底是什么?我们会使用一个明确的peg修订版本来帮助我们完成。

$ svn cat concept/IDEA@1
The idea behind this project is to come up with a piece of software
that can frab a naggily wort.  Frabbing naggily worts is tricky
business, and doing it incorrectly can have serious ramifications, so
we need to employ over-the-top input validation and data verification
mechanisms.

注意我们这一次没有提供操作修订版本,那是因为如果没有指定操作修订版本,Subversion假定缺省的操作修订版本是peg修订版本。

As you can see, the output from our operation appears to be correct. The text even mentions frabbing naggily worts, so this is almost certainly the file that describes the software now called Frabnaggilywort. In fact, we can verify this using the combination of an explicit peg revision and explicit operative revision. We know that in HEAD, the Frabnaggilywort project is located in the frabnaggilywort directory. So we specify that we want to see how the line of history identified in HEAD as the path frabnaggilywort/IDEA looked in revision 1.

$ svn cat -r 1 frabnaggilywort/IDEA@HEAD
The idea behind this project is to come up with a piece of software
that can frab a naggily wort.  Frabbing naggily worts is tricky
business, and doing it incorrectly can have serious ramifications, so
we need to employ over-the-top input validation and data verification
mechanisms.

而且peg修订版本和实施修订版本也不需要这样琐碎,举个例子,我们的frabnaggilywort已经在HEAD删除,但我们知道在修订版本20它是存在的,我们希望知道IDEA从修订版本4到10的区别,我们可以使用peg修订版本20和IDEA文件的修订版本20的URL的组合,然后使用4到10作为我们的实施修订版本范围。

$ svn diff -r 4:10 http://svn.red-bean.com/projects/frabnaggilywort/IDEA@20
Index: frabnaggilywort/IDEA
===================================================================
--- frabnaggilywort/IDEA	(revision 4)
+++ frabnaggilywort/IDEA	(revision 10)
@@ -1,5 +1,5 @@
-The idea behind this project is to come up with a piece of software
-that can frab a naggily wort.  Frabbing naggily worts is tricky
-business, and doing it incorrectly can have serious ramifications, so
-we need to employ over-the-top input validation and data verification
-mechanisms.
+The idea behind this project is to come up with a piece of
+client-server software that can remotely frab a naggily wort.
+Frabbing naggily worts is tricky business, and doing it incorrectly
+can have serious ramifications, so we need to employ over-the-top
+input validation and data verification mechanisms.

幸运的是,几乎所有的人不会面临如此复杂的情形,但是如果是,记住peg修订版本是帮助Subversion清除混淆的额外提示。

Changelists

It is commonplace for a developer to find himself working at any given time on multiple different, distinct changes to a particular bit of source code. This isn't necessarily due to poor planning or some form of digital masochism. A software engineer often spots bugs in his peripheral vision while working on some nearby chunk of source code. Or perhaps he's halfway through some large change when he realizes the solution he's working on is best committed as several smaller logical units. Often, these logical units aren't nicely contained in some module, safely separated from other changes. The units might overlap, modifying different files in the same module, or even modifying different lines in the same file.

There are various work methodologies that developers can employ to keep these logical changes organized. Some use separate working copies of the same repository to hold each individual change in progress. Others might choose to create short-lived feature branches in the repository and use a single working copy that is constantly switched to point to one such branch or another. Still others use diff and patch tools to back up and restore uncommitted changes to and from patchfiles associated with each change. Each of these methods has its pros and cons, and to a large degree, the details of the changes being made heavily influence the methodology used to distinguish them.

Subversion 1.5 brings a new changelists feature that adds yet another method to the mix. Changelists are basically arbitrary labels applied to working copy files for the express purpose of associating multiple files together. Users of many of Google's software offerings are familiar with this concept already. For example, Gmail doesn't provide the traditional folders-based email organization mechanism. In Gmail, you apply arbitrary labels to emails, and multiple emails can be said to be part of the same group if they happen to share a particular label. Viewing only a group of similarly labeled emails then becomes a simple user interface trick. Many other Web 2.0 sites have similar mechanisms—consider the “tags” used by sites such as YouTube and Flickr, “categories” applied to blog posts, and so on. Folks understand today that organization of data is critical, but that how that data is organized needs to be a flexible concept. The old files-and-folders paradigm is too rigid for some applications.

Subversion's changelist support allows you to create changelists by applying labels to files you want to be associated with that changelist, remove those labels, and limit the scope of the files on which its subcommands operate to only those bearing a particular label. In this section, we'll look in detail at how to do these things.

Creating and Modifying Changelists

You can create, modify, and delete changelists using the svn changelist command. More accurately, you use this command to set or unset the changelist association of a particular working copy file. A changelist is effectively created the first time you label a file with that changelist; it is deleted when you remove that label from the last file that had it. Let's examine a usage scenario that demonstrates these concepts.

Harry is fixing some bugs in the calculator application's mathematics logic. His work leads him to change a couple of files:

$ svn status
M      integer.c
M      mathops.c
$

While testing his bug fix, Harry notices that his changes bring to light a tangentially related bug in the user interface logic found in button.c. Harry decides that he'll go ahead and fix that bug, too, as a separate commit from his math fixes. Now, in a small working copy with only a handful of files and few logical changes, Harry can probably keep his two logical change groupings mentally organized without any problem. But today he's going to use Subversion's changelists feature as a special favor to the authors of this book.

Harry first creates a changelist and associates with it the two files he's already changed. He does this by using the svn changelist command to assign the same arbitrary changelist name to those files:

$ svn changelist math-fixes integer.c mathops.c
Path 'integer.c' is now a member of changelist 'math-fixes'.
Path 'mathops.c' is now a member of changelist 'math-fixes'.
$ svn status

--- Changelist 'math-fixes':
M      integer.c
M      mathops.c
$

As you can see, the output of svn status reflects this new grouping.

Harry now sets off to fix the secondary UI problem. Since he knows which file he'll be changing, he assigns that path to a changelist, too. Unfortunately, Harry careless assigns this third file to the same changelist as the previous two files:

$ svn changelist math-fix button.c
Path 'button.c' is now a member of changelist 'math-fixes'.
$ svn status

--- Changelist 'math-fixes':
       button.c
M      integer.c
M      mathops.c
$

Fortunately, Harry catches his mistake. At this point, he has two options. He can remove the changelist association from button.c, and then assign a different changelist name:

$ svn changelist --remove button.c
Path 'button.c' is no longer a member of a changelist.
$ svn changelist ui-fix button.c
Path 'button.c' is now a member of changelist 'ui-fix'.
$

Or, he can skip the removal and just assign a new changelist name. In this case, Subversion will first warn Harry that button.c is being removed from the first changelist:

$ svn changelist ui-fix button.c
svn: warning: Removing 'button.c' from changelist 'math-fixes'.
Path 'button.c' is now a member of changelist 'ui-fix'.
$ svn status

--- Changelist 'ui-fix':
       button.c

--- Changelist 'math-fixes':
M      integer.c
M      mathops.c
$

Harry now has two distinct changelists present in his working copy, and svn status will group its output according to these changelist determinations. Notice that even though Harry hasn't yet modified button.c, it still shows up in the output of svn status as interesting because it has a changelist assignment. Changelists can be added to and removed from files at any time, regardless of whether they contain local modifications.

Harry now fixes the user interface problem in button.c.

$ svn status

--- Changelist 'ui-fix':
M      button.c

--- Changelist 'math-fixes':
M      integer.c
M      mathops.c
$

Changelists as Operation Filters

The visual grouping that Harry sees in the output of svn status as shown in our previous section is nice, but not entirely useful. The status command is but one of many operations that he might wish to perform on his working copy. Fortunately, many of Subversion's other operations understand how to operate on changelists via the use of the --changelist option.

When provided with a --changelist option, Subversion commands will limit the scope of their operation to only those files to which a particular changelist name is assigned. If Harry now wants to see the actual changes he's made to the files in his math-fixes changelist, he could explicitly list only the files that make up that changelist on the svn diff command line.

$ svn diff integer.c mathops.c
Index: integer.c
===================================================================
--- integer.c	(revision 1157)
+++ integer.c	(working copy)
…
Index: mathops.c
===================================================================
--- mathops.c	(revision 1157)
+++ mathops.c	(working copy)
…
$

That works okay for a few files, but what if Harry's change touched 20 or 30 files? That would be an annoyingly long list of explicitly named files. Now that he's using changelists, though, Harry can avoid explicitly listing the set of files in his changelist from now on, and instead provide just the changelist name:

$ svn diff --changelist math-fixes
Index: integer.c
===================================================================
--- integer.c	(revision 1157)
+++ integer.c	(working copy)
…
Index: mathops.c
===================================================================
--- mathops.c	(revision 1157)
+++ mathops.c	(working copy)
…
$

And when it's time to commit, Harry can again use the --changelist option to limit the scope of the commit to files in a certain changelist. He might commit his user interface fix by doing the following:

$ svn ci -m "Fix a UI bug found while working on math logic." \
      --changelist ui-fix
Sending        button.c
Transmitting file data .
Committed revision 1158.
$

In fact, the svn commit command provides a second changelists-related option: --keep-changelists. Normally, changelist assignments are removed from files after they are committed. But if --keep-changelists is provided, Subversion will leave the changelist assignment on the committed (and now unmodified) files. In any case, committing files assigned to one changelist leaves other changelists undisturbed.

$ svn status

--- Changelist 'math-fixes':
M      integer.c
M      mathops.c
$

注意

The --changelist option acts only as a filter for Subversion command targets, and will not add targets to an operation. For example, on a commit operation specified as svn commit /path/to/dir, the target is the directory /path/to/dir and its children (to infinite depth). If you then add a changelist specifier to that command, only those files in and under /path/to/dir that are assigned that changelist name will be considered as targets of the commit—the commit will not include files located elsewhere (such is in /path/to/another-dir), regardless of their changelist assignment, even if they are part of the same working copy as the operation's target(s).

Even the svn changelist command accepts the --changelist option. This allows you to quickly and easily rename or remove a changelist:

$ svn changelist math-bugs --changelist math-fixes --depth infinity .
svn: warning: Removing 'integer.c' from changelist 'math-fixes'.
Path 'integer.c' is now a member of changelist 'math-bugs'.
svn: warning: Removing 'mathops.c' from changelist 'math-fixes'.
Path 'mathops.c' is now a member of changelist 'math-bugs'.
$ svn changelist --remove --changelist math-bugs --depth infinity .
Path 'integer.c' is no longer a member of a changelist.
Path 'mathops.c' is no longer a member of a changelist.
$

Finally, you can specify multiple instances of the --changelist option on a single command line. Doing so limits the operation you are performing to files found in any of the specified changesets.

Changelist Limitations

Subversion's changelist feature is a handy tool for grouping working copy files, but it does have a few limitations. Changelists are artifacts of a particular working copy, which means that changelist assignments cannot be propagated to the repository or otherwise shared with other users. Changelists can only be assigned to files—Subversion doesn't currently support the use of changelists with directories. Finally, you can have at most one changelist assignment on a given working copy file. Here is where the blog post category and photo service tag analogies break down—if you find yourself needing to assign a file to multiple changelists, you're out of luck.

网络模型

在某些情况下,你需要理解Subversion客户端如何与服务器通讯。Subversion网络层是抽象的,意味着Subversion客户端不管其操作的对象都会使用相同的行为方式,不管是使用HTTP协议(http://)与Apache HTTP服务器通讯或是使用自定义Subversion协议(svn://)与svnserve通讯,基本的网络模型是相同的。在本小节,我们要解释网络模型基础,包括Subversion如何管理认证和授权信息。

请求和响应

Subversion客户端花费大量的时间来管理工作拷贝,当它需要远程版本库的信息,它会做一个网络请求,然后服务器给一个恰当的回答,具体的网络协议细节对用户不可见,客户端尝试去访问一个URL,根据URL模式的不同,会使用特定的协议与服务器联系(见版本库的URL)。

提示

用户可以运行svn --version来查看客户端可以使用的URL模式和协议。

When the server process receives a client request, it often demands that the client identify itself. It issues an authentication challenge to the client, and the client responds by providing credentials back to the server. Once authentication is complete, the server responds with the original information that the client asked for. Notice that this system is different from systems such as CVS, where the client pre-emptively offers credentials (“logs in”) to the server before ever making a request. In Subversion, the server “pulls” credentials by challenging the client at the appropriate moment, rather than the client “pushing” them. This makes certain operations more elegant. For example, if a server is configured to allow anyone in the world to read a repository, then the server will never issue an authentication challenge when a client attempts to svn checkout.

If the particular network requests issued by the client result in a new revision being created in the repository, (e.g., svn commit), then Subversion uses the authenticated username associated with those requests as the author of the revision. That is, the authenticated user's name is stored as the value of the svn:author property on the new revision (see “Subversion Properties”一节). If the client was not authenticated (in other words, if the server never issued an authentication challenge), then the revision's svn:author property is empty.

客户端凭证缓存

Many servers are configured to require authentication on every request. This would be a big annoyance to users if they were forced to type their passwords over and over again. Fortunately, the Subversion client has a remedy for this—a built-in system for caching authentication credentials on disk. By default, whenever the command-line client successfully responds to a server's authentication challenge, it saves the credentials in the user's private runtime configuration area (~/.subversion/auth/ on Unix-like systems or %APPDATA%/Subversion/auth/ on Windows; see “运行配置区”一节 for more details about the runtime configuration system). Successful credentials are cached on disk and keyed on a combination of the server's hostname, port, and authentication realm.

当客户端接收到一个认证请求,它会首先查找用户磁盘中的认证凭证缓存,如果没有发现,或者是缓存的凭证认证失败,客户端会提示用户提供需要的信息。

十分关心安全的人们一定会想“把密码缓存在磁盘?太可怕了,永远不要这样做!

Subversion开发者认识到这种关注的正确性,所以Subversion使用操作系统和环境提供的机制来减少泄露这些信息的风险,下面是在大多数平台上这种含义的列表:

  • 在Windows 2000或更新的系统上,Subversion客户端使用标准Windows加密服务来加密磁盘上的密码。因为加密密钥是Windows管理的,与用户的登陆凭证相关,只有用户可以解密密码。(注意:如果用户的Windows账户密码被管理员重置,所有的缓存密码就不可以解密了,此时Subversion客户端就会当它们根本不存在,在需要时继续询问密码。)

  • 类似的,在Mac OS X,Subversion客户端在登陆keyring(使用Keychain管理)保存了所有的版本库密码,使用户用帐号密码保护。用户选择的设置可以强加额外的政策,例如在需要用户密码时要求输入用户帐号密码。

  • 对于其他类Unix系统,没有标准的加密服务。然而auth/缓存区只有用户(拥有者)可以访问,而不是全世界都可以,操作系统的访问许可可以保护密码文件。

当然,对于真正的妄想狂,没有任何机制是完美的。这类人希望用无限的安全来牺牲便利性,Subversion提供了各种方法来完全关闭凭证缓存。

你可以关闭凭证缓存,只需要一个简单的命令,使用参数--no-auth-cache

$ svn commit -F log_msg.txt --no-auth-cache
Authentication realm: <svn://host.example.com:3690> example realm
Username:  joe
Password for 'joe':

Adding         newfile
Transmitting file data .
Committed revision 2324.

# password was not cached, so a second commit still prompts us

$ svn delete newfile
$ svn commit -F new_msg.txt
Authentication realm: <svn://host.example.com:3690> example realm
Username:  joe
…

Or, if you want to disable credential caching permanently, you can edit the config file in your runtime configuration area and set the store-auth-creds option to no. This will prevent the storing of credentials used in any Subversion interactions you perform on the affected computer. This can be extended to cover all users on the computer, too, by modifying the system-wide runtime configuration area (described in “配置区布局”一节).

[auth]
store-auth-creds = no

有时候,用户希望从磁盘缓存删除特定的凭证,为此你可以浏览到auth/区域,删除特定的缓存文件,凭证都是作为一个单独的文件缓存,如果你打开每一个文件,你会看到键和值,svn:realmstring描述了这个文件关联的特定服务器的域:

$ ls ~/.subversion/auth/svn.simple/
5671adf2865e267db74f09ba6f872c28
3893ed123b39500bca8a0b382839198e
5c3c22968347b390f349ff340196ed39

$ cat ~/.subversion/auth/svn.simple/5671adf2865e267db74f09ba6f872c28

K 8
username
V 3
joe
K 8
password
V 4
blah
K 15
svn:realmstring
V 45
<https://svn.domain.com:443> Joe's repository
END

一旦你定位了正确的缓存文件,只需要删除它。

One last word about svn's authentication behavior, specifically regarding the --username and --password options. Many client subcommands accept these options, but it is important to understand using these options does not automatically send credentials to the server. As discussed earlier, the server “pulls” credentials from the client when it deems necessary; the client cannot “push” them at will. If a username and/or password are passed as options, they will only be presented to the server if the server requests them. These options are typically used to authenticate as a different user than Subversion would have chosen by default (such as your system login name) or when trying to avoid interactive prompting (such as when calling svn from a script).

注意

A common mistake is to misconfigure a server so that it never issues an authentication challenge. When users pass --username and --password options to the client, they're surprised to see that they're never used; i.e., new revisions still appear to have been committed anonymously!

这里是Subversion客户端在收到认证请求的时候的行为方式最终总结:

  1. First, the client checks whether the user specified any credentials as command-line options (--username and/or --password). If so, the client will try to use those credentials to authenticate against the server.

  2. If no command-line credentials were provided, or the provided ones were invalid, the client looks up the server's hostname, port, and realm in the runtime configuration's auth/ area, to see if appropriate credentials are cached there. If so, it attempts to use those credentials to authenticate.

  3. Finally, if the previous mechanisms failed to successfully authenticate the user against the server, the client resorts to interactively prompting the user for valid credentials (unless instructed not to do so via the --non-interactive option or its client-specific equivalents).

If the client successfully authenticates by any of these methods, it will attempt to cache the credentials on disk (unless the user has disabled this behavior, as mentioned earlier).



[8] 如果你熟悉XML,其实这就是XML的"Name"语法的ASCII子集。

[9] 修正提交日志信息的拼写错误,文法错误和“简单的错误”是--revprop选项最常见用例。

[10] 你认为那样过于粗狂?在同一个时代里,WordPerfect也使用.DOC作为它们私有文件格式的扩展名!

[11] The Windows filesystems use file extensions (such as .EXE, .BAT, and .COM) to denote executable files.

[12] 这不是编译系统的基本功能吗?

[13] … 或者可能是一本书的一个小节 …

[14] Communication wouldn't have been such bad medicine for Harry and Sally's Hollywood namesakes, either, for that matter.

[15] Subversion目前不允许锁定目录。

[16] 除非是,或许一个经典的火神精神融合。

[17] 你不是被期望去命名它,一旦你取了名字,你开始与之联系在一起。” — Mike Wazowski

[18] 606 N. Main Street, Wheaton, Illinois, is the home of the Wheaton History Center. It seemed appropriate….

分支与合并