Viewers: in countries Watching now:
Learn how to find and manipulate text quickly and easily using regular expressions. Author Kevin Skoglund covers the basic syntax of regular expressions, shows how to create flexible matching patterns, and demonstrates how the regular expression engine parses text to find matches. The course also covers referring back to previous matches with backreferences and creating complex matching patterns with lookaround assertions, and explores the most common applications of regular expressions.
Now that we have the basics of working with alternations, I want us to dive a little bit deeper and make sure that we write logical and efficient alternations. First, keep in mind the basic principals of regular expressions that we've seen so far, that they're eager and they're greedy. Those have an impact on the way that it process alternations. Let's go into regexpal, and let's start by just putting in peanut butter, and then up here let's put in either peanut|peanutbutter. Notice that it matched peanut, not peanut butter.
That's because it's eager. It's eager to return a result, and as we saw before, the leftmost item gets priority. Therefore, it's going to prefer to match the first item and never even attempt the second expression at all. Now if you wanted it to return peanut butter, we saw how to do that before by using our option. Let's put in this. There it is. Now it uses peanut and then butter is optional, but butter is preferred because, remember, it's greedy. So by default it's going to prefer that it has it over not having it.
See the difference? Let's try a slightly more complicated example. Let's say that here for a string we have a file name that's 2003_report.xls, and to match that, I'm just going to paste in a regular expression here. What I've got is an alternation. The first alternation is right here. It's just a word character one or more times, and then I've got a second choice which is that it's FY four digits_report.xls. Actually, let's escape this. There it is. That's correct. So now the second one is clearly a better match.
We look at the two and you think, oh well, the second one matches it really well. But which one is it using to match it? It's using the first one. That's why it didn't highlight all of the .xls that's using this expression right here. It was eager to return us a match, so we never tried the second one. It has no concept that that's a more appropriate match or anything like that. It just uses the left one. It says, I've got a match, here is your result, and just returns it back to you. Now let's take a little bit closer look at the way that the parser actually does parse things, because I want you to make sure that you understand this. Let's go back to our example where we had abc|def|ghi|jkl.
So it matched all of those. Turn off Global so you can see that it's matching just the abc. Eager to return a result to you. Now I try putting xyz at the beginning. That's the first option, xyz. Now what did it return to you? It did not jump down here and find xyz. It returned the first option. It returned the abc. That's because it doesn't scan the whole string looking for the first option and then come back and then scan the whole string looking for the second option. Instead, it starts at the beginning of the string and starts going through all those options.
Let's watch how it happens, and I think that will make more sense. So in my example here I have "I think those are thin trees," and I have four possible options in my alteration: three|see|thee|tree. So what happens is the parser starts at the beginning, and of course I doesn't match any of those, so we really don't need to worry about that. It finally though gets to the T and then it says, all right, do I have a match for this first option here? It says, yes, the first one does match. I see a T. So option one is a possibility. Let's check the next character. It's an H. Ah! Option one is still a possibility.
It goes to the I and it says option one is no longer a possibility, because it's not an R. So at that point it jumps back to the T again, and says all right, let's check option two now. Option two, well, that's an S. No, it can't be option two. Let's try option three. Option three has a T in it, so it says that's a possibility. Let's check the next character. It's an H, still a possibility. It gets to the I. No longer a match. Option three has now been ruled out. It jumps back to the T again. Now it says, let's try option four.
Option four, the T works. Try the second character. Nope, option four has been ruled out. All four options have been ruled out; therefore the T is not the beginning of a match. So we moved to the next character an H. Does H match the first one? No. The second one, the third one, the fourth one? None of those match. So it keeps working its way along until it gets to the T and the H of those. So again, option one matches both of those. It gets to the O. It doesn't match, so it rewinds back to the T. Option two doesn't match.
Option three is a possibility. Let's check its next character. Still a possibility. It goes to the third character. No longer possibility. It jumps back to the T. So you can see it does the same thing. It moves along and you get the same thing on thin as well. So when we finally get to the word trees, then it says all right, the first one, that's a possibility. It's a T. So it goes to the second character. That's not a match. Rewind. Now it goes to the second option. That's not a possibility. We rule it out. We moved to third one. That has the T, great! Let's check the next character.
It gets H, it's no longer possibility. So it rewinds back to the T and then it says, all right, the fourth option t-r-e-e, and then it knows it's got its match. So you see the little dance that it does with moving forwards and backwards as it tries each of these options. It does not take three and somehow scan the entire string looking for three and then back up and scan the whole string looking for C and then back up and scan the whole string looking for thee and then finally tree.
It doesn't do it that way; it does it by position as it moves through. So it moves through and does its little backtracking dance each time as it tries each of those four options, and that's what I want you to understand. You'll see why it does not highlight the xyz here at the end. It finds the abc instead, because it starts at the beginning and it says, no, first option is not a possibility. Then it moves to the second option and it say, here is a possibility. It moves along to the A and the B and the C and so it says, oh, got a match. I'm all done.
It can very eagerly return that match to you. One last point I want to make about writing efficient alternations is perhaps common sense, but it is just to put the simplest or the most efficient expression first. Let's imagine that we have three alternations like this. So the first one says, look for any word character one or more, underscore, and then either two to four digits, or look for four digits, underscore, two digits, underscore, and some unknown quantity of word characters. Or, that's the third choice, look for the literal characters, export, and then two digits after it.
So that's what we're looking for. It's much more efficient if we flip it around and write it in the opposite way, because this way the regular expression engine can check a character and say, is this character an E? It is not an E; therefore option one has been ruled out right away. We're ready to move on to option two. Option two says, is it a digit? If it is a digit then we can go ahead and check it, but if it's not, we've ruled out two of our three options now, very quickly. And now we can check the third one, which is going to take a lot longer, because it's much more permissive. That first character can be a lot of things.
It can be any letter, any number, or an underscore, and the character after that can be any of those things, and so on. So it can check a lot of things and do a lot of backtracking as it tries to figure it out in that third option. But the first two have been quickly ruled out. And if we have a line where one of those first two items matches, then we save ourselves the trouble of doing that third step at all. We find the match and we move on. So hopefully now you have a good understanding of the way that alternations work, the way that the regular expression engine moves through them, and how you can write good, logical, and efficient alternations.
There are currently no FAQs about Using Regular Expressions.
Access exercise files from a button right under the course name.
Search within course videos and transcripts, and jump right to the results.
Remove icons showing you already watched videos if you want to start over.
Make the video wide, narrow, full-screen, or pop the player out of the page into its own window.
Click on text in the transcript to jump to that spot in the video. As the video plays, the relevant spot in the transcript will be highlighted.