I created a regexp to match the following scenerios: SomethingCool, HelloWorld, MyNameIsDonato, etc. However, it does not account for spaces:
> 'Something Cooler' =~ /([A-Z][a-z0-9]+)+/
=> 0
That passes and it should not pass. A space is not an alphanumeric character. So why does this pass and how can I fix it?
You need to anchor the regex to the beginning and end of the string, or it will just match one of the words:
^([A-Z][a-z0-9]+)+$
^ and $ anchor the beginnings and ends of lines, respectively. To anchor to the beginning and end of the string, use \A and \Z.
It's worth noting that this is useless if you're trying to find camelcase names within a larger string. For that, use your original regex.
Related
I have a string, which contains a bunch of HTML documents, tagged with #name:
string = "#one\n\n<html>\n</html>\n\n#two\n<html>\n</html>\n\n\n"
I want to get an array of two-element arrays, each of which with a tag as the first element and the HTML document as the second:
[ ["#one", "<html>\n</html>"], ["#two", "<html>\n</html>"] ]
In order to solve the problem, I crafted the following regular expression:
regex = /(#.+)\n+([^#]+)\n+/
and applied it in string.scan regex.
However, instead of the desired output, I get the following:
[ ["#one", "<html>\n</html>\n"], ["#two", "<html>\n</html>\n\n"] ]
There are trailing newline characters at the end of each document. It appears that only one newline character was removed from the documents, but others stayed at the place.
How can the aforementioned regular expression be changed in order to remove all the trailing characters from the resulting documents?
The reason only the last \n was thrown away is because the two relevant capturing parts in your regex: .+ and [^#]+ capture everything up to the last \n (in order to make matching possible at all). It does not matter that they are followed by \n+. Remember that regex works from the left to the right. If some substring (sequences of \n in this case) can fit in either the preceding part of the following part of a regex, it actually fits in the preceding part.
With generality, I would suggest doing this:
string.split(/\s+(?=#)/).map{|s| s.strip.split(/\s+/, 2)}
# => [["#one", "<html>\n</html>"], ["#two", "<html>\n</html>"]]
You can remove duplicated newlines first:
string.gsub(/\n+/, "\n").scan(regex)
=> [["#one", "<html>\n</html>"], ["#two", "<html>\n</html>"]]
I want to match the path "/". I've tried the following alternatives, and the first two do match, but I don't know why the third doesn't:
/\A\/\z/.match("/") # <MatchData "/">
"/\A\/\z/".match("/") # <MatchData "/">
Regexp.new("/\A\/\z/").match("/") # nil
What's going on here? Why are they different?
The first snippet is the only correct one.
The second example is... misleading. That string literal "/\A\/\z/" is, obviously, not a regex. It's a string. Strings have #match method which converts its argument to a regexp (if not already one) and match against it. So, in this example, it's '/' that is the regular expression, and it matches a forward slash found in the other string.
The third line is completely broken: don't need the surrounding slashes there, they are part of regex literal, which you didn't use. Also use single quoted strings, not double quoted (which try to interpret escape sequences like \A)
Regexp.new('\A/\z').match("/") # => #<MatchData "/">
And, of course, none of the above is needed if you just want to check if a string consists of only one forward slash. Just use the equality check in this case.
s == '/'
So I have a string that looks like this:
#jackie#test.com, #mike#test.com
What I want to do is before any email in this comma separated list, I want to remove the #. The issue I keep running into is that if I try to do a regular \A flag like so /[\A#]+/, it finds all the instances of # in that string...including the middle crucial #.
The same thing happens if I do /[\s#]+/. I can't figure out how to just look at the beginning of each string, where each string is a complete email address.
Edit 1
Note that all I need is the regex, I already have the rest of the stuff I need to do what I want. Specifically, I am achieving everything else like this:
str.gsub(/#/, '').split(',').map(&:strip)
Where str is my string.
All I am looking for is the regex portion for my gsub.
You may use the below negative lookbehind based regex.
str.gsub(/(?<!\S)#/, '').split(',').map(&:strip)
(?<!\S) Negative lookbehind asserts that the character or substring we are going to match would be preceeded by any but not of a non-space character. So this matches the # which exists at the start or the # which exists next to a space character.
Difference between my answer and hwnd's str.gsub(/\B#/, '') is, mine won't match the # which exists in :# but hwnd's answer does. \B matches between two word characters or two non-word characters.
Here is one solution
str = "#jackie#test.com, #mike#test.com"
p str.split(/,[ ]+/).map{ |i| i.gsub(/^#/, '')}
Output
["jackie#test.com", "mike#test.com"]
Is there anyway to scan only if there is nothing before what I am scanning for.
For example I have a post and I am scanning for a forward slash and what follows it but I do not want to scan for a forward slash if it is not the beginning character.
I want to scan for /this but I do not want to scan for this/this or http://this.com.
The regular expression I am currently using is..
/\/(\w+)/
I am using this with gsub to link each /forwardslash.
I think what you are asking for is to only match words that begin with '/', not strings or lines beginning with '/'. If that is true, I believe the following regex will work: %r{(?:^|\s+)/(\w+)}:
For example:
"/foo /this this/that http://this".scan %r{(?:^|\s+)/(\w+)} # => [["foo"], ["this"]]
The caret (^) character means "beginning of string" -- a dollar sign ($) means "end of string."
So
/^\/(\w+)/
...will get you what you want -- only matching at the beginning of the string.
First thing, since you're using a regex with slashes change the delimiter to something else, then you won't have to escape the backslashes and it will be easier to read.
Secondly, if you want to replace the slash as well then include it in the capture.
On to the regex.
...if it is not the beginning
character...
...of a line:
!^(/\w+)!
if it is not the beginning
character...
...of a word:
!\s(/\w+)!
but that won't match if it's at the very beginning of a line. For that you'll need something a lot more complex, so I'd just run both the regexes here instead of creating that monster.
I work in Ruby, and have to create a single regexp for the following task, as I'm working with someone else's gem that uses this regexp to match fields to be worked on in a text file. I need to match beginning of string, any set of characters, and underscore, then any multi-digit integer that is not 1,2, 9, or 10, and end of string.
I.e., I want the following to match:
foo_4
bar_8
baz_120
BUT NOT:
foo_1
bar_9
baz_10
I tried
/^.+_(^(1|2|9|10))$/
but it did not work as apparently ^ only "negates" characters in brackets, not submatches.
Outside of a character class the ^ symbol means start of line. I think you want a negative lookahead instead:
/^.+_(?!(?:1|2|9|10)$)\d+$/
See it in action on rubular.