Easy-to-follow video tutorials help you learn software, creative, and business skills.Become a member
Regular expressions provide a very powerful pattern-matching mechanism, much more powerful than the simple like patterns and also much more complex. Regular expressions are actually a pattern-matching language. It's an incredibly useful feature and it's also a very large subject in itself. In this movie I'll give you a brief overview of the power of regular expressions, but I'm really just scratching the surface. For a more complete discussion of regular expressions. The lynda.com Online Training Library has an excellent course called Using Regular Expressions with Kevin Skoglund.
If you watched the lesson on like expressions, you may remember this statement. I'm going to select the world database here. And. Select name, continent, population, from country where name like underscore a percent order by name. And we'll press go and here we have a list of all of the countries with an a in the second letter of their name in order.
Now suppose instead of just. The letter A in the second position, you wanted all the names with an A, or a B, or a C, or a D in the second position. To do this with like, you would need four like expressions bound together with or operators. And it would end up looking something like this. I'm going to break this into. Multiple lines to make it a little easier. This is actually looking pretty complex isn't it? And, if I run that, we now have.
Oh, there's a C. There just aren't a lot of. B's or D's in the second position are there. But there you have it, it works. Now with a regular expression, this is actually very easy. Instead of all of these or's you would have one where and let me just put this on the same line like that and instead of like. And this pattern here you would say REGEXP which is short for regular expression and a pattern like this.
I'm going to go a through e so we get a few extras here like that. And that's our regular expression. And when I select go, and you see we have some e's in here as well. So you can see this is very powerful. You can do all of this in one line. And it's also pretty cryptic looking. Regular expressions are very powerful. They can be a little complex, but the basics are actually pretty easy to grasp. So in this query, the caret at the beginning. That's an anchor for the beginning of the line.
And so, this anchors the pattern to the beginning of the line so that this dot, which represents one character, it's kind of like the underscore in the like expressions, that represents the very first character in the string. And so the anchor anchors at the beginning and then you have the first character, which can be any character. And then the second character is represented by this bracketed expression that says a-e. And that means any character between a and e. So that's a, b, c, d, and e.
And then we have another dot. Which, as you know, means any character. because we learned that in the first character. Followed by an asterisk, which means, 0 or more of those. And in this case, it's the any character. And then you can actually anchor it at the end. With a dollar sign which anchors the end of the string so I press go, of course this will do the same thing because we didn't really need that anchor because we're matching any character to the end of the string. On the other hand if I just wanted to match anything that ends in a G I can just do this and that matches anything that ends in a G.
Because that G is anchored at the end of the string. And so, when I press Go I just get the names of countries that end in G. So, a period matches any character in that position. For example, if I just do this, that'll match anything. With an O and any character and another O after it. And if I press Go here we get Cocos, Columbia, Comoros, Morocco because of the oro, Solomon, Togo. So, that matches any string with an o followed by one other character followed by another o.
The plus sign matches one or more of a character. So for example, if I change this dot to a C with a plus sign, this'll match anything with one or more C's between two O's. And if I press go here, we get Cocos because it has OCO. We also get Morocco because it has an OCCO. So it matches one or more Cs between the Os. The asterisk matches zero or more of a character. So, if I use an asterisk here instead I'll get any two Os next to each other or any two Os with one or more Cs in between them.
And if I press go here, so we have Cameroon with two O's. We have Cocos because it has O-C-O. And cook, and Morocco. Cook just has the two O's next to each other, and Morocco has two C's in between the O's. So that's the asterisk. It matches zero or more of a particular character. The question mark matches zero or one of a character. So if I change this to a question mark, we won't get Morocco anymore. I'll press go.
We don't get Morocco because it has more than one C. But we do get the ones with no Cs. So we have zero or one. Regular expressions can also specify character classes. A character class is a set of characters that can match a position in a string. So, again, if we want to match the second letter in a string, we can anchor to the beginning and match any character for one position. And then, give it a character class. And then .asterisk in the end of the string.
So inside this character class, I'm just going to select vowels, A, E, I, O, and U. And this will find anything with a vowel in the second position. So we get 165 of those. We have Australia, Bahamas, Belarus. So we have E's, O's, U's, A's. All the vowels in the second position. So this character class allows us to say match any character that's in this list, but just one of them; or if I put a plus sign there it's one or more, and a question mark is zero or one, or an asterisk is zero or more.
You can also specify of course a range. Of letters and we did this before when we said a through e, the dash, so that's a range of letters, so a, b, c, d and e. So, if I press go we just get those. There are also a number of predefined classes, for example if I want to match. A space character. These have to be in an extra set of brackets and between colons. And I put in the word space and that actually matches any white space.
A space character, a tab character, a new line character, and a few others. So, if I take out all of this other stuff because we're not probably going to find a space as the second position. But this'll give us any strings with a space in them anywhere. And so this has just the ones with more than one word. Other predefined classes include these classes here. You'll find this list and other information in the MySQL documentation at this URL.
R like is a synonym for Reg X. And it looks like this. And it works exactly the same. You'll see this occasionally. It's included mostly for legacy purposes. The R like keyword was originally included for compatibility with an older database system called MSQL. For now its best to use the Regexp keyword
Get unlimited access to all courses for just $25/month.Become a member