Working with check-in, checkout, revert, and tracking history

All right, we've got a working HG repository. It's time to add some files and check it in. Mercurial is different than some of the other systems because we have to do one piece of initialization globally before we can start checking things in. What we have to do is to create a file in the user profile called Mercurial.ini, and inside of it we had to put in our username and email address. So we'll say Notepad, %userprofile%\Mercurial.ini. It doesn't exist, we do want to create it.

I copied the text that we need onto the clipboard, so you will type in ui, then username = and here you will put in your username, and over here you'll put your email address. Just remember, this is not a real email address. Don't try sending email there. Okay, now we have got Mercurial fully initialized. We can start putting in some code. So let's open up Notepad and create our file. The code we are going to put in has some content from the new project's haiku error messages page.

We'll pick one that's happen to all of us before we start using Version Control "With searching comes loss\n", "and the presence of absence:/n", and then the kicker "My Novel\" not found.\n". Well, at least this one will never happen to you again, now that you are going to be a Version Control expert. All right. We'll save it. Now if we do hg status, you can see that HG sees it but doesn't know what to do with the file.

So we need to tell it that this is a new file for the repository, hg add f1.c. Okay, now we can ask Hg the status again and see that it's got the file ready to add. So finally, remembering that we set up our Mercurial.ini file at the top here, we'll check it in, and everything would be fine, hg commit -m "Initial Checkin". All right. Now if we ask Hg status again, you will see there is nothing going on, and then we see hg log, and you can see that it puts out five lines for each changeset.

If you look at the first line, you can see the changeset identifier. Now in Mercurial there is two numbers, one-- 0 in this case--is the local identifier that you can use for doing things like reverting and versions. We'll see that later on in this movie. The second is a hash that's used to uniquely identify this changeset so that when this changeset is for example exported or pushed to a remote repository, they can make sure that they can uniquely identify that from other changeset zeroes that might come in from other users.

You can see our user as we set it up in the Mercurial.ini file, the date, the summary and so forth where you see tag: tip--that's the same as Git Head--and it essentially says this changeset was added to the most recent version of repository. So this is the top or the tip of the iceberg here that has everything in it, that's the very latest stuff. So, now let's make another change so we can see Hg's versioning in action. We'll open up our file, and we'll change My Novel to Website and then we'll save it.

This time we are going to do hg commit, but we are not going to put a message audit on the command line here, we are going to let it open up Notepad so that we can put in a multi-line message. So we'll say Change My Novel to Website, This is a better change for modern times. We'll save it. Now if say hg log, you will see that you only see the summary.

If you want to see the full commit messages, you have to say hg log -v. And there you can see in our changeset, number 1 there. It's--instead of saying summary, it now says description. All right. So now let's make another change and then examine how to revert or rollback. Once again, we'll open up f1.c, and we'll change Website to Webpage, and we'll save it. If we run hg diff, you can see that if we committed this change. It would change Website to Webpage, and you can see it shows you the difference between what's in your working set and what's in the repository.

Note in our case, we only have one file. If you had 25 files, when you said hg diff, it would show the difference in all of the files. You can ask it to show you the differences in only a single file by typing hg diff f1.c. Now in our case, that's exactly the same since we only have one file, but you understand now how to see the difference for an individual file. There is a lot of additional options on the diff command so that you can ask it to show you the difference between different times, and you can also ask it to show you the difference between the working set and a specific changeset.

If we do hg log, we can see that we have changeset 1 and changeset 0. So if we say hg diff -r 0, it will show us the difference between our current working set which has web page and the initial Checkin which had My Novel. All right. So now let's roll back to our most recent commit, and we do that by saying hg revert in our file name, we type out f1.c, and you can see now we are back to website. Finally, you can also revert to any revision, not just the topmost one by using the revert command and specifying a changeset so that we'll do hg revert -r 0 f1.c, and again we type out f1.c. You can see we are back to My Novel.

Now, note none of your changes in the repository have been lost, you are just updating the working set. So if we just say hg revert f1.c, you can see we are back to website. So that's the basics of checking in and checking out and reverting with Hg. Now let's move on to talking about tagging.