Characters And Metacharacters
Literal Characters: Letters, digits and unicode
All letters, digits and most unicode characters in a regex pattern are literal, so the regex engine will search for exactly that pattern, without any other processing.
So if you search for at, your pattern will match these strings: "cat", "bat", "You were late, you need to be at home at 10".
Remember: Regex patterns are case sensitive, although there are ways to specify a case insensitive search (depending on the language used).
Reserved Character List (Metacharacters)
Almost any other characters (both ASCII and Unicode) will be treated as literals too.
But there is a list of reserved, special characters called metacharacters.
Click here to show the table of reserved characters
These are the most important (we will see all of them in more detail in the subsequent lessons):
| Chars | Description |
|---|---|
. | Any single character. |
( ) | Grouping delimitators. |
[ ] | Character set delimiters. OR at character level |
| | OR expression for patterns (one or another). |
{ } | Repetitions delimiters. |
* | Zero or more repetitions of the previous character. |
+ | One or more repetitions. |
? | Zero or one repetition. Also it's used for lazy matches. |
^ | Start of the string. You can use it to force a pattern to match only at the start. Also it's used as a NOT inside of character sets. |
$ | End of the string. |
/ | Separator. In many regex engines, regex patterns must be enclosed in /s |
- | Range definition. Used to define a range of consecutive characters, like A-Z |
\ | Escape character for all reserved characters, so \? will search for a literal ?. It's also used for other special search patterns (see below). |
Anchors:
^and$are also called anchors. Anchors match zero characters.
Note: In other search engines, the
*is a wildcard that matches everything. However in regex, the wildcard is created with the union of two metacharacters:.*, which means "any single character zero or more times". Similarly, in other search engines?is equivalent to.in regex (any single character, once).
Other metacharacters
Click here to show the table of additional metacharacters
| Pattern | Description |
|---|---|
\w | Any letter (a to z, both lowercase and uppercase), digit (0 to 9), or the underscore char _. |
\W | Just the opposite of the previous metacharacter. |
\d | Any digit, 0 to 9. |
\D | Just the opposite of the previous metacharacter. |
\b | Word boundaries. It's an anchor used to find the start or the end of a word (defined as any number of consecutive \w characters). \bat will match " attack " and "atlantis" but not "bat" because the word doesn't start with at. |
\B | Just the opposite of the previous metacharacter. |
\t | Tabulation. |
\r | Carriage return. |
\n | New line. |
\s | Means either: space or \t or \r or \n. That is, any kind of white space. |
\S | Just the opposite of the previous metacharacter. |
Note: There are many other patterns using
\, but these are the most important.
Backslash
Using backslash, outside of the previously defined metacharacters, will escape the following character. The regex engine will then consider this character as a simple literal. According to the table above, * is a quantifier, but if you need to use the character * in a string, \* means "the character *".
Furthermore, the \ can escape itself if you need to match a backslash character. You can use \\ to match backslash.
Search flags and modifiers
Most regex engines have some flags to change the search behaviour. Check your language to know how to apply them.
Click here to show search flags
| Modifier | Description |
|---|---|
| i | Case insensitive search |
| g | Global search. Some regex engines stop on the first match, the g flag forces the search to return all possible matches. |
| s | Single line, . will match \n so the whole text is considered as a single line. |
| m | Multiline, ^ and $ will match on each line, and not only at start and end of string. |
In the next lesson, we will learn about Character Sets.
