Positive Lookahead and Non-capturing group difference - ruby

When you want to match either of two patterns but not capture it, you would use a noncapturing group ?::
/(?:https?|ftp)://(.+)/
But what if I want to capture '_1' in the string 'john_1'. It could be '2' or '' followed by anything else. First I tried a non-capturing group:
'john_1'.gsub(/(?:.+)(_.+)/, "")
=> ""
It does not work. I am telling it to not capture one or more characters but to capture _ and all characters after it.
Instead the following works:
'john_1'.gsub(/(?=.+)(_.+)/, "")
=> "john"
I used a positive lookahead. The definition I found for positive lookahead was as follows:
q(?=u) matches a q that is
followed by a u, without making the u part of the match. The positive
lookahead construct is a pair of parentheses, with the opening
parenthesis followed by a question mark and an equals sign.
But that definition doesn't really fit my example. What makes the Positive Lookahead work but not the Non-capturing group work in the example I provide?

Capturing and matching are two different things. (?:expr) doesn't capture expr, but it's still included in the matched string. Zero-width assertions, e.g. (?=expr), don't capture or include expr in the matched string.
Perhaps some examples will help illustrate the difference:
> "abcdef"[/abc(def)/] # => abcdef
> $1 # => def
> "abcdef"[/abc(?:def)/] # => abcdef
> $1 # => nil
> "abcdef"[/abc(?=def)/] # => abc
> $1 # => nil
When you use a non-capturing group in your String#gsub call, it's still part of the match, and gets replaced by the replacement string.

Your first example doesn't work because a non-capturing group is still part of the overall capture, whereas the lookbehind is only used for matching but isn't part of the overall capture.
This is easier to understand if you get the actual match data:
# Non-capturing group
/(?:.+)(_.+)/.match 'john_1'
=> #<MatchData "john_1" 1:"_1">
# Positive Lookbehind
/(?=.+)(_.+)/.match 'john_1'
=> #<MatchData "_1" 1:"_1">
EDIT: I should also mention that sub and gsub work on the entire capture, not individual capture groups (although those can be used in the replacement).
'john_1'.gsub(/(?:.+)(_.+)/, 'phil\1')
=> "phil_1"

Let's consider a couple of situations.
The string preceding the underscore must be "john" and the underscore is followed by one or more characters
str = "john_1"
You have two choices.
Use a positive lookbehind
str[/(?<=john)_.+/]
#=> "_1"
The positive lookbehind requires that "john" must appear immediately before the underscore, but it is not part of the match that is returned.
Use a capture group:
str[/john(_.+)/, 1]
#=> "_1"
This regular expression matches "john_1", but "_.+" is captured in capture group 1. By examining the doc for the method String#[] you will see that one form of the method is str[regexp, capture], which returns the contents of the capture group capture. Here capture equals 1, meaning the first capture group.
Note that the string following the underscore may contain underscores: "john_1_a"[/(?<=john)_.+/] #=> "_1_a".
If the underscore can be at the end of the string replace + with * in the above regular expressions (meaning match zero or more characters after the underscore).
The string preceding the underscore can be anything and and the underscore is followed by one or more characters
str = "john_mary_tom_julie"
We may consider two cases.
The string returned is to begin with the first underscore
In this case we could write:
str[/_.+/]
#=> "_mary_tom_julie"
This works because the regex is by default greedy, meaning it will begin at the first underscore encountered.
The string returned is to begin with the last underscore
Here we could write:
str[/_[^_]+\z/]
#=> "_julie"
This regex matches an underscore followed by one or more characters that are not underscores, followed by the end-of-string anchor (\z).
Aside: the method String#[]
[] may seem an odd name for a method but it is a method nevertheless, so it can be invoked in the conventional way:
str.[](/john(_.+)/, 1)
#=> "_1"
The expression str[/john(_.+)/, 1] is an example (of which there are many in Ruby) of syntactic sugar. When written str[...] Ruby converts it to the conventional expression for methods before evaluating it.

Related

Convert string to camel case in Ruby

Working on a Ruby challenge to convert dash/underscore delimited words into camel casing. The first word within the output should be capitalized only if the original word was capitalized (known as Upper Camel Case).
My solution so far..:
def to_camel_case(str)
str.split('_,-').collect.camelize(:lower).join
end
However .camelize(:lower) is a rails method I believe and doesn't work with Ruby. Is there an alternative method, equally as simplistic? I can't seem to find one. Or do I need to approach the challenge from a completely different angle?
main.rb:4:in `to_camel_case': undefined method `camelize' for #<Enumerator: []:collect> (NoMethodError)
from main.rb:7:in `<main>'
I assume that:
Each "word" is made up of one or more "parts".
Each part is made of up characters other than spaces, hypens and underscores.
The first character of each part is a letter.
Each successive pair of parts is separated by a hyphen or underscore.
It is desired to return a string obtained by modifying each part and removing the hypen or underscore that separates each successive pair of parts.
For each part all letters but the first are to be converted to lowercase.
All characters in each part of a word that are not letters are to remain unchanged.
The first letter of the first part is to remain unchanged.
The first letter of each part other than the first is to be capitalized (if not already capitalized).
Words are separated by spaces.
It this describes the problem correctly the following method could be used.
R = /(?:(?<=^| )|[_-])[A-Za-z][^ _-]*/
def to_camel_case(str)
str.gsub(R) do |s|
c1 = s[0]
case c1
when /[A-Za-z]/
c1 + s[1..-1].downcase
else
s[1].upcase + s[2..-1].downcase
end
end
end
to_camel_case "Little Miss-muffet sat_on_HE$R Tuffett eating-her_cURDS And_whey"
# => "Little MissMuffet satOnHe$r Tuffett eatingHerCurds AndWhey"
The regular expression is can be written in free-spacing mode to make it self-documenting.
R = /
(?: # begin non-capture group
(?<=^| ) # use a positive lookbehind to assert that the next character
# is preceded by the beginning of the string or a space
| # or
[_-] # match '_' or '-'
) # end non-capture group
[A-Za-z] # match a letter
[^ _-]* # match 0+ characters other than ' ', '_' and '-'
/x # free-spacing regex definition mode
Most Rails methods can be added into basic Ruby projects without having to pull in the whole Rails source.
The trick is to figure out the minimum amount of files to require in order to define the method you need. If we go to APIDock, we can see that camelize is defined in active_support/inflector/methods.rb.
Therefore active_support/inflector seems like a good candidate to try. Let's test it:
irb(main)> require 'active_support/inflector'
=> true
irb(main)> 'foo_bar'.camelize
=> "FooBar"
Seems to work. Note that this assumes you already ran gem install activesupport earlier. If not, then do it first (or add it to your Gemfile).
In pure Ruby, no Rails, given str = 'my-var_name' you could do:
delimiters = Regexp.union(['-', '_'])
str.split(delimiters).then { |first, *rest| [first, rest.map(&:capitalize)].join }
#=> "myVarName"
Where str = 'My-var_name' the result is "MyVarName", since the first element of the splitting result is untouched, while the rest is mapped to be capitalized.
It works only with "dash/underscore delimited words", no spaces, or you need to split by spaces, then map with the presented method.
This method is using string splitting by delimiters, as explained here Split string by multiple delimiters,
chained with Object#then.

How do I specify in Ruby that I want to match a character provided that a sequence following that character does not match a pattern?

I'm using Ruby on Rails 5.1. In Ruby, how do I say taht I want to match a string if the first character matches something but the sequence that follows does NOT match a pattern? That is, I want to match a number provided that the sequence taht follows is not a character from an array I have followed by two other numbers. Here's my character array ...
2.4.0 :010 > TOKENS
=> [":", ".", "'"]
So this string would NOT match
3:00
since ":00" matches the pattern of a character from my array followed by two numbers. But this string
3400
would match. This string would also match
3:0
and this would match
3
since nothing follows the above. How do I write the appropriate regex in Ruby?
string =~ /\A\d+(?!:\d{2})/
This regular expression means:
\A anchors the match to the start of the string.
\d+ means "one or more digits".
(?!...) is a negative look-ahead. It checks that the pattern contained in the brackets does not match., looking ahead from the current position.
:\d{2} means : followed by two digits.
Consideration should be given to testing the first character and the remaining characters separately.
def match_it?(str, first_char_regex, no_match_regex)
str[0].match?(first_char_regex) && !str[1..-1].match?(no_match_regex)
end
match_it?("0:00", /0/, /\A[:. ]cat\z/) #=> true
match_it?("0:00", /\d/, /\A[:. ]\d+\z/) #=> false
match_it?("0:00", /[[:alpha:]]/, /\A[:. ]\d+\z/) #=> false
I believe this reads well and it simplifies testing when compared to methods that employ a single regular expression.

Regex: Match all hyphens or underscores not at the beginning or the end of the string

I am writing some code that needs to convert a string to camel case. However, I want to allow any _ or - at the beginning of the code.
I have had success matching up an _ character using the regex here:
^(?!_)(\w+)_(\w+)(?<!_)$
when the inputs are:
pro_gamer #matched
#ignored
_proto
proto_
__proto
proto__
__proto__
#matched as nerd_godess_of, skyrim
nerd_godess_of_skyrim
I recursively apply my method on the first match if it looks like nerd_godess_of.
I am having troubled adding - matches to the same, I assumed that just adding a - to the mix like this would work:
^(?![_-])(\w+)[_-](\w+)(?<![_-])$
and it matches like this:
super-mario #matched
eslint-path #matched
eslint-global-path #NOT MATCHED.
I would like to understand why the regex fails to match the last case given that it worked correctly for the _.
The (almost) full set of test inputs can be found here
The fact that
^(?![_-])(\w+)[_-](\w+)(?<![_-])$
does not match the second hyphen in "eslint-global-path" is because of the anchor ^ which limits the match to be on the first hyphen only. This regex reads, "Match the beginning of the line, not followed by a hyphen or underscore, then match one or more words characters (including underscores), a hyphen or underscore, and then one or more word characters in a capture group. Lastly, do not match a hyphen or underscore at the end of the line."
The fact that an underscore (but not a hyphen) is a word (\w) character completely messes up the regex. In general, rather than using \w, you might want to use \p{Alpha} or \p{Alnum} (or POSIX [[:alpha:]] or [[:alnum:]]).
Try this.
r = /
(?<= # begin a positive lookbehind
[^_-] # match a character other than an underscore or hyphen
) # end positive lookbehind
( # begin capture group 1
(?: # begin a non-capture group
-+ # match one or more hyphens
| # or
_+ # match one or more underscores
) # end non-capture group
[^_-] # match any character other than an underscore or hyphen
) # end capture group 1
/x # free-spacing regex definition mode
'_cats_have--nine_lives--'.gsub(r) { |s| s[-1].upcase }
#=> "_catsHaveNineLives--"
This regex is conventionally written as follows.
r = /(?<=[^_-])((?:-+|_+)[^_-])/
If all the letters are lower case one could alternatively write
'_cats_have--nine_lives--'.split(/(?<=[^_-])(?:_+|-+)(?=[^_-])/).
map(&:capitalize).join
#=> "_catsHaveNineLives--"
where
'_cats_have--nine_lives--'.split(/(?<=[^_-])(?:_+|-+)(?=[^_-])/)
#=> ["_cats", "have", "nine", "lives--"]
(?=[^_-]) is a positive lookahead that requires the characters on which the split is made to be followed by a character other than an underscore or hyphen
you can try the regex
^(?=[^-_])(\w+[-_]\w*)+(?=[^-_])\w$
see the demo here.
Switch _- to -_ so that - is not treated as a range op, as in a-z.

Ruby regular expression non capture group

I'm trying to grab id number from the string, say
id/number/2000GXZ2/ref=sr
using
(?:id\/number\/)([a-zA-Z0-9]{8})
for some reason non capture group is not worked, giving me:
id/number/2000GXZ2
As mentioned by others, non-capturing groups still count towards the overall match. If you don't want that part in your match use a lookbehind.
Rubular example
(?<=id\/number\/)([a-zA-Z0-9]{8})
(?<=pat) - Positive lookbehind assertion: ensures that the preceding characters match pat, but doesn't include those characters in the matched text
Ruby Doc Regexp
Also, the capture group around the id number is unnecessary in this case.
You have:
str = "id/number/2000GXZ2/ref=sr"
r = /
(?:id\/number\/) # match string in a non-capture group
([a-zA-Z0-9]{8}) # match character in character class 8 times, in capture group 1
/x # extended/free-spacing regex definition mode
Then (using String#[]):
str[r]
#=> "id/number/2000GXZ2"
returns the entire match, as it should, not just the contents of capture group 1. There are a few ways to remedy this. Consider first ones that do not use a capture group.
#jacob.m suggested putting the first part in a positive lookbehind (modified slightly from his code):
r = /
(?<=id\/number\/) # match string in positive lookbehind
[[:alnum:]]{8} # match >= 1 alphameric characters
/x
str[r]
#=> "2000GXZ2"
An alternative is:
r = /
id\/number\/ # match string
\K # forget everything matched so far
[[:alnum:]]{8} # match 8 alphanumeric characters
/x
str[r]
#=> "2000GXZ2"
\K is especially useful when the match to forget is variable-length, as (in Ruby) positive lookbehinds do not work with variable-length matches.
With both of these approaches, if the part to be matched contains only numbers and capital letters, you may want to use [A-Z0-9]+ instead of [[:alnum:]] (though the latter includes Unicode letters, not just those from the English alphabet). In fact, if all the entries have the form of your example, you might be able to use:
r = /
\d # match a digit
[A-Z0-9]{7} # match >= 0 capital letters or digits
/x
str[r]
#=> "2000GXZ2"
The other line of approach is to keep your capture group. One simple way is:
r = /
id\/number\/ # match string
([[:alnum:]]{8}) # match >= 1 alphameric characters in capture group 1
/x
str =~ r
str[r, 1] #=> "2000GXZ2"
Alternatively, you could use String#sub to replace the entire string with the contents of the capture group:
r = /
id\/number\/ # match string
([[:alnum:]]{8}) # match >= 1 alphameric characters in capture group 1
.* # match the remainder of the string
/x
str.sub(r, '\1') #=> "2000GXZ2"
str.sub(r, "\\1") #=> "2000GXZ2"
str.sub(r) { $1 } #=> "2000GXZ2"
This is Ruby Regexp expected match consistency evilness. Some Regexp-style methods will return the global-match while others will return specified matches.
In this case, one method we can use to get the behavior you're looking for is scan.
I don't think anyone here actually mentions how to get your Regexp working as you originally intended, which was to get the capture-only match. To do that, you would use the scan method like so with your original pattern:
test_me.rb
test_string="id/number/2000GXZ2/ref=sr"
result = test_string.scan(/(?:id\/number\/)([a-zA-Z0-9]{8})/)
puts result
2000GXZ2
That said, replacing (?:) with (?<=) for non-capture groups for look-behinds will benefit you both when you use scan as well as other parts of ruby that use Regexps.

Use regular expression to fetch 3 groups from string

This is my expected result.
Input a string and get three returned string.
I have no idea how to finish it with Regex in Ruby.
this is my roughly idea.
match(/(.*?)(_)(.*?)(\d+)/)
Input and expected output
# "R224_OO2003" => R224, OO, 2003
# "R2241_OOP2003" => R2244, OOP, 2003
If the example description I gave in my comment on the question is correct, you need a very straightforward regex:
r = /(.+)_(.+)(\d{4})/
Then:
"R224_OO2003".scan(r).flatten #=> ["R224", "OO", "2003"]
"R2241_OOP2003".scan(r).flatten #=> ["R2241", "OOP", "2003"]
Assuming that your three parts consist of (R and one or more digits), then an underbar, then (one or more non-whitespace characters), before finally (a 4-digit numeric date), then your regex could be something like this:
^(R\d+)_(\S+)(\d{4})$
The ^ indicates start of string, and the $ indicates end of string. \d+ indicates one or more digits, while \S+ says one or more non-whitespace characters. The \d{4} says exactly four digits.
To recover data from the matches, you could either use the pre-defined globals that line up with your groups, or you could could use named captures.
To use the match globals just use $1, $2, and $3. In general, you can figure out the number to use by counting the left parentheses of the specific group.
To use the named captures, include ? right after the left paren of a particular group. For example:
x = "R2241_OOP2003"
match_data = /^(?<first>R\d+)_(?<second>\S+)(?<third>\d{4})$/.match(x)
puts match_data['first'], match_data['second'], match_data['third']
yields
R2241
OOP
2003
as expected.
As long as your pattern covers all possibilities, then you just need to use the match object to return the 3 strings:
my_match = "R224_OO2003".match(/(.*?)(_)(.*?)(\d+)/)
#=> #<MatchData "R224_OO2003" 1:"R224" 2:"_" 3:"OO" 4:"2003">
puts my_match[0] #=> "R224_OO2003"
puts my_match[1] #=> "R224"
puts my_match[2] #=> "_"
puts my_match[3] #=> "00"
puts my_match[4] #=> "2003"
A MatchData object contains an array of each match group starting at index [1]. As you can see, index [0] returns the entire string. If you don't want the capture the "_" you can leave it's parentheses out.
Also, I'm not sure you are getting what you want with the part:
(.*?)
this basically says one or more of any single character followed by zero or one of any single character.

Resources