jerojasro@343: \chapter{Una gira de Mercurial: lo básico} jerojasro@343: \label{chap:tour-basic} jerojasro@343: jerojasro@343: \section{Instalar Mercurial en su sistema} jerojasro@343: \label{sec:tour:install} jerojasro@343: Hay paquetes binarios precompilados de Mercurial disponibles para cada jerojasro@343: sistema operativo popular. Esto hace fácil empezar a usar Mercurial jerojasro@343: en su computador inmediatamente. jerojasro@343: jerojasro@343: \subsection{Linux} jerojasro@343: jerojasro@343: Dado que cada distribución de Linux tiene sus propias herramientas de jerojasro@343: manejo de paquetes, políticas, y ritmos de desarrollo, es difícil dar jerojasro@343: un conjunto exhaustivo de instrucciones sobre cómo instalar el paquete jerojasro@343: de Mercurial. La versión de Mercurial que usted tenga a disposición jerojasro@343: puede variar dependiendo de qué tan activa sea la persona que mantiene jerojasro@343: el paquete para su distribución. jerojasro@343: jerojasro@343: Para mantener las cosas simples, me enfocaré en instalar Mercurial jerojasro@343: desde la línea de comandos en las distribuciones de Linux más jerojasro@343: populares. La mayoría de estas distribuciones proveen administradores jerojasro@343: de paquetes gráficos que le permitirán instalar Mercurial con un solo jerojasro@343: clic; el nombre de paquete a buscar es \texttt{mercurial}. jerojasro@343: jerojasro@343: \begin{itemize} jerojasro@343: \item[Debian] jerojasro@343: \begin{codesample4} jerojasro@343: apt-get install mercurial jerojasro@343: \end{codesample4} jerojasro@343: jerojasro@343: \item[Fedora Core] jerojasro@343: \begin{codesample4} jerojasro@343: yum install mercurial jerojasro@343: \end{codesample4} jerojasro@343: jerojasro@343: \item[Gentoo] jerojasro@343: \begin{codesample4} jerojasro@343: emerge mercurial jerojasro@343: \end{codesample4} jerojasro@343: jerojasro@343: \item[OpenSUSE] jerojasro@343: \begin{codesample4} jerojasro@343: yum install mercurial jerojasro@343: \end{codesample4} jerojasro@343: jerojasro@343: \item[Ubuntu] El paquete de Mercurial de Ubuntu está basado en el de jerojasro@343: Debian. Para instalarlo, ejecute el siguiente comando. jerojasro@343: \begin{codesample4} jerojasro@343: apt-get install mercurial jerojasro@343: \end{codesample4} jerojasro@343: El paquete de Mercurial para Ubuntu tiende a atrasarse con respecto jerojasro@343: a la versión de Debian por un margen de tiempo considerable jerojasro@343: (al momento de escribir esto, 7 meses), lo que en algunos casos jerojasro@343: significará que usted puede encontrarse con problemas que ya habrán jerojasro@343: sido resueltos en el paquete de Debian. jerojasro@343: \end{itemize} jerojasro@343: jerojasro@343: \subsection{Solaris} jerojasro@343: jerojasro@343: SunFreeWare, en \url{http://www.sunfreeware.com}, es una buena fuente jerojasro@343: para un gran número de paquetes compilados para Solaris para las jerojasro@343: arquitecturas Intel y Sparc de 32 y 64 bits, incluyendo versiones jerojasro@343: actuales de Mercurial. jerojasro@343: jerojasro@343: \subsection{Mac OS X} jerojasro@343: jerojasro@343: Lee Cantey publica un instalador de Mercurial para Mac OS~X en jerojasro@343: \url{http://mercurial.berkwood.com}. Este paquete funciona en tanto jerojasro@343: en Macs basados en Intel como basados en PowerPC. Antes de que pueda jerojasro@343: usarlo, usted debe instalar una versión compatible de Universal jerojasro@343: MacPython~\cite{web:macpython}. Esto es fácil de hacer; simplemente jerojasro@343: siga las instrucciones de el sitio de Lee. jerojasro@343: jerojasro@343: También es posible instalar Mercurial usando Fink o MacPorts, dos jerojasro@343: administradores de paquetes gratuitos y populares para Mac OS X. Si jerojasro@343: usted tiene Fink, use \command{sudo apt-get install mercurial-py25}. jerojasro@343: Si usa MacPorts, \command{sudo port install mercurial}. jerojasro@343: jerojasro@343: \subsection{Windows} jerojasro@343: jerojasro@343: Lee Cantey publica un instalador de Mercurial para Windows en jerojasro@343: \url{http://mercurial.berkwood.com}. Este paquete no tiene jerojasro@343: % TODO traducción de it just works. Agreed? jerojasro@343: dependencias externas; ``simplemente funciona''. jerojasro@343: jerojasro@343: \begin{note} jerojasro@343: La versión de Windows de Mercurial no convierte automáticamente jerojasro@343: los fines de línea entre estilos Windows y Unix. Si usted desea jerojasro@343: compartir trabajo con usuarios de Unix, deberá hacer un trabajo jerojasro@343: adicional de configuración. XXX Terminar esto. jerojasro@343: \end{note} jerojasro@343: jerojasro@343: \section{Arrancando} jerojasro@343: jerojasro@343: Para empezar, usaremos el comando \hgcmd{version} para revisar si jerojasro@343: Mercurial está instalado adecuadamente. La información de la versión jerojasro@343: que es impresa no es tan importante; lo que nos importa es si imprime jerojasro@343: algo en absoluto. jerojasro@343: jerojasro@343: \interaction{tour.version} jerojasro@343: jerojasro@343: % TODO builtin-> integrado? jerojasro@343: \subsection{Ayuda integrada} jerojasro@343: jerojasro@343: Mercurial provee un sistema de ayuda integrada. Esto es invaluable jerojasro@343: para ésas ocasiones en la que usted está atorado tratando de recordar jerojasro@343: cómo ejecutar un comando. Si está completamente atorado, simplemente jerojasro@343: ejecute \hgcmd{help}; esto imprimirá una breve lista de comandos, jerojasro@343: junto con una descripción de qué hace cada uno. Si usted solicita jerojasro@343: ayuda sobre un comando específico (como abajo), se imprime información jerojasro@343: más detallada. jerojasro@343: \interaction{tour.help} jerojasro@343: Para un nivel más impresionante de detalle (que usted no va a jerojasro@343: necesitar usualmente) ejecute \hgcmdargs{help}{\hggopt{-v}}. La opción jerojasro@343: \hggopt{-v} es la abreviación para \hggopt{--verbose}, y le indica a jerojasro@343: Mercurial que imprima más información de lo que haría usualmente. jerojasro@343: jerojasro@343: \section{Working with a repository} jerojasro@343: jerojasro@343: In Mercurial, everything happens inside a \emph{repository}. The jerojasro@343: repository for a project contains all of the files that ``belong to'' jerojasro@343: that project, along with a historical record of the project's files. jerojasro@343: jerojasro@343: There's nothing particularly magical about a repository; it is simply jerojasro@343: a directory tree in your filesystem that Mercurial treats as special. jerojasro@343: You can rename or delete a repository any time you like, using either the jerojasro@343: command line or your file browser. jerojasro@343: jerojasro@343: \subsection{Making a local copy of a repository} jerojasro@343: jerojasro@343: \emph{Copying} a repository is just a little bit special. While you jerojasro@343: could use a normal file copying command to make a copy of a jerojasro@343: repository, it's best to use a built-in command that Mercurial jerojasro@343: provides. This command is called \hgcmd{clone}, because it creates an jerojasro@343: identical copy of an existing repository. jerojasro@343: \interaction{tour.clone} jerojasro@343: If our clone succeeded, we should now have a local directory called jerojasro@343: \dirname{hello}. This directory will contain some files. jerojasro@343: \interaction{tour.ls} jerojasro@343: These files have the same contents and history in our repository as jerojasro@343: they do in the repository we cloned. jerojasro@343: jerojasro@343: Every Mercurial repository is complete, self-contained, and jerojasro@343: independent. It contains its own private copy of a project's files jerojasro@343: and history. A cloned repository remembers the location of the jerojasro@343: repository it was cloned from, but it does not communicate with that jerojasro@343: repository, or any other, unless you tell it to. jerojasro@343: jerojasro@343: What this means for now is that we're free to experiment with our jerojasro@343: repository, safe in the knowledge that it's a private ``sandbox'' that jerojasro@343: won't affect anyone else. jerojasro@343: jerojasro@343: \subsection{What's in a repository?} jerojasro@343: jerojasro@343: When we take a more detailed look inside a repository, we can see that jerojasro@343: it contains a directory named \dirname{.hg}. This is where Mercurial jerojasro@343: keeps all of its metadata for the repository. jerojasro@343: \interaction{tour.ls-a} jerojasro@343: jerojasro@343: The contents of the \dirname{.hg} directory and its subdirectories are jerojasro@343: private to Mercurial. Every other file and directory in the jerojasro@343: repository is yours to do with as you please. jerojasro@343: jerojasro@343: To introduce a little terminology, the \dirname{.hg} directory is the jerojasro@343: ``real'' repository, and all of the files and directories that coexist jerojasro@343: with it are said to live in the \emph{working directory}. An easy way jerojasro@343: to remember the distinction is that the \emph{repository} contains the jerojasro@343: \emph{history} of your project, while the \emph{working directory} jerojasro@343: contains a \emph{snapshot} of your project at a particular point in jerojasro@343: history. jerojasro@343: jerojasro@343: \section{A tour through history} jerojasro@343: jerojasro@343: One of the first things we might want to do with a new, unfamiliar jerojasro@343: repository is understand its history. The \hgcmd{log} command gives jerojasro@343: us a view of history. jerojasro@343: \interaction{tour.log} jerojasro@343: By default, this command prints a brief paragraph of output for each jerojasro@343: change to the project that was recorded. In Mercurial terminology, we jerojasro@343: call each of these recorded events a \emph{changeset}, because it can jerojasro@343: contain a record of changes to several files. jerojasro@343: jerojasro@343: The fields in a record of output from \hgcmd{log} are as follows. jerojasro@343: \begin{itemize} jerojasro@343: \item[\texttt{changeset}] This field has the format of a number, jerojasro@343: followed by a colon, followed by a hexadecimal string. These are jerojasro@343: \emph{identifiers} for the changeset. There are two identifiers jerojasro@343: because the number is shorter and easier to type than the hex jerojasro@343: string. jerojasro@343: \item[\texttt{user}] The identity of the person who created the jerojasro@343: changeset. This is a free-form field, but it most often contains a jerojasro@343: person's name and email address. jerojasro@343: \item[\texttt{date}] The date and time on which the changeset was jerojasro@343: created, and the timezone in which it was created. (The date and jerojasro@343: time are local to that timezone; they display what time and date it jerojasro@343: was for the person who created the changeset.) jerojasro@343: \item[\texttt{summary}] The first line of the text message that the jerojasro@343: creator of the changeset entered to describe the changeset. jerojasro@343: \end{itemize} jerojasro@343: The default output printed by \hgcmd{log} is purely a summary; it is jerojasro@343: missing a lot of detail. jerojasro@343: jerojasro@343: Figure~\ref{fig:tour-basic:history} provides a graphical representation of jerojasro@343: the history of the \dirname{hello} repository, to make it a little jerojasro@343: easier to see which direction history is ``flowing'' in. We'll be jerojasro@343: returning to this figure several times in this chapter and the chapter jerojasro@343: that follows. jerojasro@343: jerojasro@343: \begin{figure}[ht] jerojasro@343: \centering jerojasro@343: \grafix{tour-history} jerojasro@343: \caption{Graphical history of the \dirname{hello} repository} jerojasro@343: \label{fig:tour-basic:history} jerojasro@343: \end{figure} jerojasro@343: jerojasro@343: \subsection{Changesets, revisions, and talking to other jerojasro@343: people} jerojasro@343: jerojasro@343: As English is a notoriously sloppy language, and computer science has jerojasro@343: a hallowed history of terminological confusion (why use one term when jerojasro@343: four will do?), revision control has a variety of words and phrases jerojasro@343: that mean the same thing. If you are talking about Mercurial history jerojasro@343: with other people, you will find that the word ``changeset'' is often jerojasro@343: compressed to ``change'' or (when written) ``cset'', and sometimes a jerojasro@343: changeset is referred to as a ``revision'' or a ``rev''. jerojasro@343: jerojasro@343: While it doesn't matter what \emph{word} you use to refer to the jerojasro@343: concept of ``a~changeset'', the \emph{identifier} that you use to jerojasro@343: refer to ``a~\emph{specific} changeset'' is of great importance. jerojasro@343: Recall that the \texttt{changeset} field in the output from jerojasro@343: \hgcmd{log} identifies a changeset using both a number and a jerojasro@343: hexadecimal string. jerojasro@343: \begin{itemize} jerojasro@343: \item The revision number is \emph{only valid in that repository}, jerojasro@343: \item while the hex string is the \emph{permanent, unchanging jerojasro@343: identifier} that will always identify that exact changeset in jerojasro@343: \emph{every} copy of the repository. jerojasro@343: \end{itemize} jerojasro@343: This distinction is important. If you send someone an email talking jerojasro@343: about ``revision~33'', there's a high likelihood that their jerojasro@343: revision~33 will \emph{not be the same} as yours. The reason for this jerojasro@343: is that a revision number depends on the order in which changes jerojasro@343: arrived in a repository, and there is no guarantee that the same jerojasro@343: changes will happen in the same order in different repositories. jerojasro@343: Three changes $a,b,c$ can easily appear in one repository as $0,1,2$, jerojasro@343: while in another as $1,0,2$. jerojasro@343: jerojasro@343: Mercurial uses revision numbers purely as a convenient shorthand. If jerojasro@343: you need to discuss a changeset with someone, or make a record of a jerojasro@343: changeset for some other reason (for example, in a bug report), use jerojasro@343: the hexadecimal identifier. jerojasro@343: jerojasro@343: \subsection{Viewing specific revisions} jerojasro@343: jerojasro@343: To narrow the output of \hgcmd{log} down to a single revision, use the jerojasro@343: \hgopt{log}{-r} (or \hgopt{log}{--rev}) option. You can use either a jerojasro@343: revision number or a long-form changeset identifier, and you can jerojasro@343: provide as many revisions as you want. \interaction{tour.log-r} jerojasro@343: jerojasro@343: If you want to see the history of several revisions without having to jerojasro@343: list each one, you can use \emph{range notation}; this lets you jerojasro@343: express the idea ``I want all revisions between $a$ and $b$, jerojasro@343: inclusive''. jerojasro@343: \interaction{tour.log.range} jerojasro@343: Mercurial also honours the order in which you specify revisions, so jerojasro@343: \hgcmdargs{log}{-r 2:4} prints $2,3,4$ while \hgcmdargs{log}{-r 4:2} jerojasro@343: prints $4,3,2$. jerojasro@343: jerojasro@343: \subsection{More detailed information} jerojasro@343: jerojasro@343: While the summary information printed by \hgcmd{log} is useful if you jerojasro@343: already know what you're looking for, you may need to see a complete jerojasro@343: description of the change, or a list of the files changed, if you're jerojasro@343: trying to decide whether a changeset is the one you're looking for. jerojasro@343: The \hgcmd{log} command's \hggopt{-v} (or \hggopt{--verbose}) jerojasro@343: option gives you this extra detail. jerojasro@343: \interaction{tour.log-v} jerojasro@343: jerojasro@343: If you want to see both the description and content of a change, add jerojasro@343: the \hgopt{log}{-p} (or \hgopt{log}{--patch}) option. This displays jerojasro@343: the content of a change as a \emph{unified diff} (if you've never seen jerojasro@343: a unified diff before, see section~\ref{sec:mq:patch} for an overview). jerojasro@343: \interaction{tour.log-vp} jerojasro@343: jerojasro@343: \section{All about command options} jerojasro@343: jerojasro@343: Let's take a brief break from exploring Mercurial commands to discuss jerojasro@343: a pattern in the way that they work; you may find this useful to keep jerojasro@343: in mind as we continue our tour. jerojasro@343: jerojasro@343: Mercurial has a consistent and straightforward approach to dealing jerojasro@343: with the options that you can pass to commands. It follows the jerojasro@343: conventions for options that are common to modern Linux and Unix jerojasro@343: systems. jerojasro@343: \begin{itemize} jerojasro@343: \item Every option has a long name. For example, as we've already jerojasro@343: seen, the \hgcmd{log} command accepts a \hgopt{log}{--rev} option. jerojasro@343: \item Most options have short names, too. Instead of jerojasro@343: \hgopt{log}{--rev}, we can use \hgopt{log}{-r}. (The reason that jerojasro@343: some options don't have short names is that the options in question jerojasro@343: are rarely used.) jerojasro@343: \item Long options start with two dashes (e.g.~\hgopt{log}{--rev}), jerojasro@343: while short options start with one (e.g.~\hgopt{log}{-r}). jerojasro@343: \item Option naming and usage is consistent across commands. For jerojasro@343: example, every command that lets you specify a changeset~ID or jerojasro@343: revision number accepts both \hgopt{log}{-r} and \hgopt{log}{--rev} jerojasro@343: arguments. jerojasro@343: \end{itemize} jerojasro@343: In the examples throughout this book, I use short options instead of jerojasro@343: long. This just reflects my own preference, so don't read anything jerojasro@343: significant into it. jerojasro@343: jerojasro@343: Most commands that print output of some kind will print more output jerojasro@343: when passed a \hggopt{-v} (or \hggopt{--verbose}) option, and less jerojasro@343: when passed \hggopt{-q} (or \hggopt{--quiet}). jerojasro@343: jerojasro@343: \section{Making and reviewing changes} jerojasro@343: jerojasro@343: Now that we have a grasp of viewing history in Mercurial, let's take a jerojasro@343: look at making some changes and examining them. jerojasro@343: jerojasro@343: The first thing we'll do is isolate our experiment in a repository of jerojasro@343: its own. We use the \hgcmd{clone} command, but we don't need to jerojasro@343: clone a copy of the remote repository. Since we already have a copy jerojasro@343: of it locally, we can just clone that instead. This is much faster jerojasro@343: than cloning over the network, and cloning a local repository uses jerojasro@343: less disk space in most cases, too. jerojasro@343: \interaction{tour.reclone} jerojasro@343: As an aside, it's often good practice to keep a ``pristine'' copy of a jerojasro@343: remote repository around, which you can then make temporary clones of jerojasro@343: to create sandboxes for each task you want to work on. This lets you jerojasro@343: work on multiple tasks in parallel, each isolated from the others jerojasro@343: until it's complete and you're ready to integrate it back. Because jerojasro@343: local clones are so cheap, there's almost no overhead to cloning and jerojasro@343: destroying repositories whenever you want. jerojasro@343: jerojasro@343: In our \dirname{my-hello} repository, we have a file jerojasro@343: \filename{hello.c} that contains the classic ``hello, world'' program. jerojasro@343: Let's use the ancient and venerable \command{sed} command to edit this jerojasro@343: file so that it prints a second line of output. (I'm only using jerojasro@343: \command{sed} to do this because it's easy to write a scripted example jerojasro@343: this way. Since you're not under the same constraint, you probably jerojasro@343: won't want to use \command{sed}; simply use your preferred text editor to jerojasro@343: do the same thing.) jerojasro@343: \interaction{tour.sed} jerojasro@343: jerojasro@343: Mercurial's \hgcmd{status} command will tell us what Mercurial knows jerojasro@343: about the files in the repository. jerojasro@343: \interaction{tour.status} jerojasro@343: The \hgcmd{status} command prints no output for some files, but a line jerojasro@343: starting with ``\texttt{M}'' for \filename{hello.c}. Unless you tell jerojasro@343: it to, \hgcmd{status} will not print any output for files that have jerojasro@343: not been modified. jerojasro@343: jerojasro@343: The ``\texttt{M}'' indicates that Mercurial has noticed that we jerojasro@343: modified \filename{hello.c}. We didn't need to \emph{inform} jerojasro@343: Mercurial that we were going to modify the file before we started, or jerojasro@343: that we had modified the file after we were done; it was able to jerojasro@343: figure this out itself. jerojasro@343: jerojasro@343: It's a little bit helpful to know that we've modified jerojasro@343: \filename{hello.c}, but we might prefer to know exactly \emph{what} jerojasro@343: changes we've made to it. To do this, we use the \hgcmd{diff} jerojasro@343: command. jerojasro@343: \interaction{tour.diff} jerojasro@343: jerojasro@343: \section{Recording changes in a new changeset} jerojasro@343: jerojasro@343: We can modify files, build and test our changes, and use jerojasro@343: \hgcmd{status} and \hgcmd{diff} to review our changes, until we're jerojasro@343: satisfied with what we've done and arrive at a natural stopping point jerojasro@343: where we want to record our work in a new changeset. jerojasro@343: jerojasro@343: The \hgcmd{commit} command lets us create a new changeset; we'll jerojasro@343: usually refer to this as ``making a commit'' or ``committing''. jerojasro@343: jerojasro@343: \subsection{Setting up a username} jerojasro@343: jerojasro@343: When you try to run \hgcmd{commit} for the first time, it is not jerojasro@343: guaranteed to succeed. Mercurial records your name and address with jerojasro@343: each change that you commit, so that you and others will later be able jerojasro@343: to tell who made each change. Mercurial tries to automatically figure jerojasro@343: out a sensible username to commit the change with. It will attempt jerojasro@343: each of the following methods, in order: jerojasro@343: \begin{enumerate} jerojasro@343: \item If you specify a \hgopt{commit}{-u} option to the \hgcmd{commit} jerojasro@343: command on the command line, followed by a username, this is always jerojasro@343: given the highest precedence. jerojasro@343: \item If you have set the \envar{HGUSER} environment variable, this is jerojasro@343: checked next. jerojasro@343: \item If you create a file in your home directory called jerojasro@343: \sfilename{.hgrc}, with a \rcitem{ui}{username} entry, that will be jerojasro@343: used next. To see what the contents of this file should look like, jerojasro@343: refer to section~\ref{sec:tour-basic:username} below. jerojasro@343: \item If you have set the \envar{EMAIL} environment variable, this jerojasro@343: will be used next. jerojasro@343: \item Mercurial will query your system to find out your local user jerojasro@343: name and host name, and construct a username from these components. jerojasro@343: Since this often results in a username that is not very useful, it jerojasro@343: will print a warning if it has to do this. jerojasro@343: \end{enumerate} jerojasro@343: If all of these mechanisms fail, Mercurial will fail, printing an jerojasro@343: error message. In this case, it will not let you commit until you set jerojasro@343: up a username. jerojasro@343: jerojasro@343: You should think of the \envar{HGUSER} environment variable and the jerojasro@343: \hgopt{commit}{-u} option to the \hgcmd{commit} command as ways to jerojasro@343: \emph{override} Mercurial's default selection of username. For normal jerojasro@343: use, the simplest and most robust way to set a username for yourself jerojasro@343: is by creating a \sfilename{.hgrc} file; see below for details. jerojasro@343: jerojasro@343: \subsubsection{Creating a Mercurial configuration file} jerojasro@343: \label{sec:tour-basic:username} jerojasro@343: jerojasro@343: To set a user name, use your favourite editor to create a file called jerojasro@343: \sfilename{.hgrc} in your home directory. Mercurial will use this jerojasro@343: file to look up your personalised configuration settings. The initial jerojasro@343: contents of your \sfilename{.hgrc} should look like this. jerojasro@343: \begin{codesample2} jerojasro@343: # This is a Mercurial configuration file. jerojasro@343: [ui] jerojasro@343: username = Firstname Lastname jerojasro@343: \end{codesample2} jerojasro@343: The ``\texttt{[ui]}'' line begins a \emph{section} of the config file, jerojasro@343: so you can read the ``\texttt{username = ...}'' line as meaning ``set jerojasro@343: the value of the \texttt{username} item in the \texttt{ui} section''. jerojasro@343: A section continues until a new section begins, or the end of the jerojasro@343: file. Mercurial ignores empty lines and treats any text from jerojasro@343: ``\texttt{\#}'' to the end of a line as a comment. jerojasro@343: jerojasro@343: \subsubsection{Choosing a user name} jerojasro@343: jerojasro@343: You can use any text you like as the value of the \texttt{username} jerojasro@343: config item, since this information is for reading by other people, jerojasro@343: but for interpreting by Mercurial. The convention that most people jerojasro@343: follow is to use their name and email address, as in the example jerojasro@343: above. jerojasro@343: jerojasro@343: \begin{note} jerojasro@343: Mercurial's built-in web server obfuscates email addresses, to make jerojasro@343: it more difficult for the email harvesting tools that spammers use. jerojasro@343: This reduces the likelihood that you'll start receiving more junk jerojasro@343: email if you publish a Mercurial repository on the web. jerojasro@343: \end{note} jerojasro@343: jerojasro@343: \subsection{Writing a commit message} jerojasro@343: jerojasro@343: When we commit a change, Mercurial drops us into a text editor, to jerojasro@343: enter a message that will describe the modifications we've made in jerojasro@343: this changeset. This is called the \emph{commit message}. It will be jerojasro@343: a record for readers of what we did and why, and it will be printed by jerojasro@343: \hgcmd{log} after we've finished committing. jerojasro@343: \interaction{tour.commit} jerojasro@343: jerojasro@343: The editor that the \hgcmd{commit} command drops us into will contain jerojasro@343: an empty line, followed by a number of lines starting with jerojasro@343: ``\texttt{HG:}''. jerojasro@343: \begin{codesample2} jerojasro@343: \emph{empty line} jerojasro@343: HG: changed hello.c jerojasro@343: \end{codesample2} jerojasro@343: Mercurial ignores the lines that start with ``\texttt{HG:}''; it uses jerojasro@343: them only to tell us which files it's recording changes to. Modifying jerojasro@343: or deleting these lines has no effect. jerojasro@343: jerojasro@343: \subsection{Writing a good commit message} jerojasro@343: jerojasro@343: Since \hgcmd{log} only prints the first line of a commit message by jerojasro@343: default, it's best to write a commit message whose first line stands jerojasro@343: alone. Here's a real example of a commit message that \emph{doesn't} jerojasro@343: follow this guideline, and hence has a summary that is not readable. jerojasro@343: \begin{codesample2} jerojasro@343: changeset: 73:584af0e231be jerojasro@343: user: Censored Person jerojasro@343: date: Tue Sep 26 21:37:07 2006 -0700 jerojasro@343: summary: include buildmeister/commondefs. Add an exports and install jerojasro@343: \end{codesample2} jerojasro@343: jerojasro@343: As far as the remainder of the contents of the commit message are jerojasro@343: concerned, there are no hard-and-fast rules. Mercurial itself doesn't jerojasro@343: interpret or care about the contents of the commit message, though jerojasro@343: your project may have policies that dictate a certain kind of jerojasro@343: formatting. jerojasro@343: jerojasro@343: My personal preference is for short, but informative, commit messages jerojasro@343: that tell me something that I can't figure out with a quick glance at jerojasro@343: the output of \hgcmdargs{log}{--patch}. jerojasro@343: jerojasro@343: \subsection{Aborting a commit} jerojasro@343: jerojasro@343: If you decide that you don't want to commit while in the middle of jerojasro@343: editing a commit message, simply exit from your editor without saving jerojasro@343: the file that it's editing. This will cause nothing to happen to jerojasro@343: either the repository or the working directory. jerojasro@343: jerojasro@343: If we run the \hgcmd{commit} command without any arguments, it records jerojasro@343: all of the changes we've made, as reported by \hgcmd{status} and jerojasro@343: \hgcmd{diff}. jerojasro@343: jerojasro@343: \subsection{Admiring our new handiwork} jerojasro@343: jerojasro@343: Once we've finished the commit, we can use the \hgcmd{tip} command to jerojasro@343: display the changeset we just created. This command produces output jerojasro@343: that is identical to \hgcmd{log}, but it only displays the newest jerojasro@343: revision in the repository. jerojasro@343: \interaction{tour.tip} jerojasro@343: We refer to the newest revision in the repository as the tip revision, jerojasro@343: or simply the tip. jerojasro@343: jerojasro@343: \section{Sharing changes} jerojasro@343: jerojasro@343: We mentioned earlier that repositories in Mercurial are jerojasro@343: self-contained. This means that the changeset we just created exists jerojasro@343: only in our \dirname{my-hello} repository. Let's look at a few ways jerojasro@343: that we can propagate this change into other repositories. jerojasro@343: jerojasro@343: \subsection{Pulling changes from another repository} jerojasro@343: \label{sec:tour:pull} jerojasro@343: jerojasro@343: To get started, let's clone our original \dirname{hello} repository, jerojasro@343: which does not contain the change we just committed. We'll call our jerojasro@343: temporary repository \dirname{hello-pull}. jerojasro@343: \interaction{tour.clone-pull} jerojasro@343: jerojasro@343: We'll use the \hgcmd{pull} command to bring changes from jerojasro@343: \dirname{my-hello} into \dirname{hello-pull}. However, blindly jerojasro@343: pulling unknown changes into a repository is a somewhat scary jerojasro@343: prospect. Mercurial provides the \hgcmd{incoming} command to tell us jerojasro@343: what changes the \hgcmd{pull} command \emph{would} pull into the jerojasro@343: repository, without actually pulling the changes in. jerojasro@343: \interaction{tour.incoming} jerojasro@343: (Of course, someone could cause more changesets to appear in the jerojasro@343: repository that we ran \hgcmd{incoming} in, before we get a chance to jerojasro@343: \hgcmd{pull} the changes, so that we could end up pulling changes that we jerojasro@343: didn't expect.) jerojasro@343: jerojasro@343: Bringing changes into a repository is a simple matter of running the jerojasro@343: \hgcmd{pull} command, and telling it which repository to pull from. jerojasro@343: \interaction{tour.pull} jerojasro@343: As you can see from the before-and-after output of \hgcmd{tip}, we jerojasro@343: have successfully pulled changes into our repository. There remains jerojasro@343: one step before we can see these changes in the working directory. jerojasro@343: jerojasro@343: \subsection{Updating the working directory} jerojasro@343: jerojasro@343: We have so far glossed over the relationship between a repository and jerojasro@343: its working directory. The \hgcmd{pull} command that we ran in jerojasro@343: section~\ref{sec:tour:pull} brought changes into the repository, but jerojasro@343: if we check, there's no sign of those changes in the working jerojasro@343: directory. This is because \hgcmd{pull} does not (by default) touch jerojasro@343: the working directory. Instead, we use the \hgcmd{update} command to jerojasro@343: do this. jerojasro@343: \interaction{tour.update} jerojasro@343: jerojasro@343: It might seem a bit strange that \hgcmd{pull} doesn't update the jerojasro@343: working directory automatically. There's actually a good reason for jerojasro@343: this: you can use \hgcmd{update} to update the working directory to jerojasro@343: the state it was in at \emph{any revision} in the history of the jerojasro@343: repository. If you had the working directory updated to an old jerojasro@343: revision---to hunt down the origin of a bug, say---and ran a jerojasro@343: \hgcmd{pull} which automatically updated the working directory to a jerojasro@343: new revision, you might not be terribly happy. jerojasro@343: jerojasro@343: However, since pull-then-update is such a common thing to do, jerojasro@343: Mercurial lets you combine the two by passing the \hgopt{pull}{-u} jerojasro@343: option to \hgcmd{pull}. jerojasro@343: \begin{codesample2} jerojasro@343: hg pull -u jerojasro@343: \end{codesample2} jerojasro@343: If you look back at the output of \hgcmd{pull} in jerojasro@343: section~\ref{sec:tour:pull} when we ran it without \hgopt{pull}{-u}, jerojasro@343: you can see that it printed a helpful reminder that we'd have to take jerojasro@343: an explicit step to update the working directory: jerojasro@343: \begin{codesample2} jerojasro@343: (run 'hg update' to get a working copy) jerojasro@343: \end{codesample2} jerojasro@343: jerojasro@343: To find out what revision the working directory is at, use the jerojasro@343: \hgcmd{parents} command. jerojasro@343: \interaction{tour.parents} jerojasro@343: If you look back at figure~\ref{fig:tour-basic:history}, you'll see jerojasro@343: arrows connecting each changeset. The node that the arrow leads jerojasro@343: \emph{from} in each case is a parent, and the node that the arrow jerojasro@343: leads \emph{to} is its child. The working directory has a parent in jerojasro@343: just the same way; this is the changeset that the working directory jerojasro@343: currently contains. jerojasro@343: jerojasro@343: To update the working directory to a particular revision, give a jerojasro@343: revision number or changeset~ID to the \hgcmd{update} command. jerojasro@343: \interaction{tour.older} jerojasro@343: If you omit an explicit revision, \hgcmd{update} will update to the jerojasro@343: tip revision, as shown by the second call to \hgcmd{update} in the jerojasro@343: example above. jerojasro@343: jerojasro@343: \subsection{Pushing changes to another repository} jerojasro@343: jerojasro@343: Mercurial lets us push changes to another repository, from the jerojasro@343: repository we're currently visiting. As with the example of jerojasro@343: \hgcmd{pull} above, we'll create a temporary repository to push our jerojasro@343: changes into. jerojasro@343: \interaction{tour.clone-push} jerojasro@343: The \hgcmd{outgoing} command tells us what changes would be pushed jerojasro@343: into another repository. jerojasro@343: \interaction{tour.outgoing} jerojasro@343: And the \hgcmd{push} command does the actual push. jerojasro@343: \interaction{tour.push} jerojasro@343: As with \hgcmd{pull}, the \hgcmd{push} command does not update the jerojasro@343: working directory in the repository that it's pushing changes into. jerojasro@343: (Unlike \hgcmd{pull}, \hgcmd{push} does not provide a \texttt{-u} jerojasro@343: option that updates the other repository's working directory.) jerojasro@343: jerojasro@343: What happens if we try to pull or push changes and the receiving jerojasro@343: repository already has those changes? Nothing too exciting. jerojasro@343: \interaction{tour.push.nothing} jerojasro@343: jerojasro@343: \subsection{Sharing changes over a network} jerojasro@343: jerojasro@343: The commands we have covered in the previous few sections are not jerojasro@343: limited to working with local repositories. Each works in exactly the jerojasro@343: same fashion over a network connection; simply pass in a URL instead jerojasro@343: of a local path. jerojasro@343: \interaction{tour.outgoing.net} jerojasro@343: In this example, we can see what changes we could push to the remote jerojasro@343: repository, but the repository is understandably not set up to let jerojasro@343: anonymous users push to it. jerojasro@343: \interaction{tour.push.net} jerojasro@343: jerojasro@343: %%% Local Variables: jerojasro@343: %%% mode: latex jerojasro@343: %%% TeX-master: "00book" jerojasro@343: %%% End: