Regular expression \b

Quote a Passage online:

\b is a special code prescribed by regular expressions (well, some people call it metacharacters, metacharacter), which represents the beginning or end of a word, that is, the boundary of a word. Although English words are usually delimited by spaces, punctuation marks, or line breaks, \b does not match any of these word-delimited characters, it only matches one position.

If a more precise argument is needed, \b matches such a position: its previous character and the next one are not all (one is, one is not or does not exist) \w.

Many people do not understand the meaning of the \b in the regular, after seeing the above paragraph, many people still do not understand exactly what a "position" of \b.

Today is to say what I understand.

What is location

It ' s a nice day today.

' I ' occupies a position, ' t ' occupies a position, and all individual characters (including invisible white space characters) will occupy a position where I give it a name called "explicit position".

Note: There is also a position between the character and the character, for example ' I ' and ' t ' have a position (nothing), so I give it a name called "Implicit position".

"Implicit location" is the key to \b! Popular understanding, \b is "implicit position".

At this point, let's try to understand this sentence:

If a more precise argument is needed, \b matches such a position: its previous character and the next one are not all (one is, one is not or does not exist) \w.

I use my words to translate this sentence:

The "implicit location" \b, which matches this position: its previous "explicit position" character and the subsequent "explicit position" characters are not all \w.

Now, is there a sense of enlightened feeling? Do you have one? Do you have one? Do you have one?

Example Explanation

Just use "It's a nice day today." For example:

Correct regularization: \bnice\b

Analysis: The first \b in front of a character is a space, followed by a character is ' n ', not all \w, so you can match the ' n ' is the beginning of a word. The second \b before a character is ' E ', the following character is a space, not all \w, can match the ' e ' is the end of a word. So, together, you can match a word that ends with ' e ' starting with ' n ', which matches the word "nice".

Incorrect regex: A\bnice

Analysis: I have seen someone like this to write the regular, want to achieve the purpose is to match the last word with ' a ' end, the next word with ' n ' part, here want to match "a nice." But this is not the purpose of the regular expression, \b Front is the character ' A ', followed by the character ' N ', two are "explicit characters", obviously contrary to the meaning of \b, so this is a wrong expression, can not match anything. To match "A nice", the correct regular notation is: A\b.\bnice (cannot wrap)

Http://www.cnblogs.com/litmmp/p/4925374.html

Regular expression \b (rpm)