Select all characters in a string until a specific character Ruby-CodePudding

So what I want is to get n characters until it hits a specific character.

i have this String :

a='2.452811139617034,42.10874821716908|3.132087902867818,42.028314077306646|-0.07934861041448178,41.647538468746916|-0.07948265046522918,41.64754863599606'

How can I make it to only get to the character | , but without getting that character, and get it like this:

2.452811139617034,42.10874821716908

CodePudding user response：

You can simply do it like this:

a.split('|').first

CodePudding user response：

a[/[^|] /]
  #=> "2.452811139617034,42.10874821716908"

The regular expression simply matches as many characters other than '|' that it can.

CodePudding user response：

You can avoid creating an unnecessary Array (like Array#split) or using a Regex (like Array#gsub) by using.

a = "2.452811139617034,42.10874821716908|3.132087902867818,42.028314077306646|-0.07934861041448178,41.647538468746916|-0.07948265046522918,41.64754863599606"

a[0,a.index('|')]
#=>"2.452811139617034,42.1087482171"

This means select characters at positions 0 up to the index of the first pipe (|). Technically speaking it is start at position 0 and select the length of n where n is the index of the pipe character which works in this case because ruby uses 0 based indexing.

As @CarySwoveland astutely pointed out the string may not contain a pipe in which case my solution would need to change to

#to return entire string
a[0,a.index('|') || a.size]
# or
b = a.index(?|) ? a[0,b] : a
# or to return empty string
a[0,a.index('|').to_i]
# or to return nil
a[0,a.index(?|) || -1]

CodePudding user response：

One approach uses a gsub replacement:

a = "2.452811139617034,42.10874821716908|3.132087902867818,42.028314077306646|-0.07934861041448178,41.647538468746916|-0.07948265046522918,41.64754863599606"
output = a.gsub(/\|.*$/, "")
puts output  # 2.452811139617034,42.10874821716908

CodePudding user response：

I'm always curious as to the performance of various options and so I took the liberty of looping through some of the approaches suggested. I named them as follows:

split_all = a.split('|').first
partition = a.partition('|').first
split_two = a.split('|', 2).first
string_brack_args = a[0,a.index('|')]
string_brack_range = a[0...a.index('|')]
gsub_regex = a.gsub(/\|.*$/, "")
plain_regex = a[/[^|] /]

The results were as follows (slowest at top):

   user     system      total        real
gsub_regex           0.170000   0.000000   0.170000 (  0.162666)
split_all            0.110000   0.000000   0.110000 (  0.109498)
split_two            0.040000   0.000000   0.040000 (  0.041792)
partition            0.040000   0.000000   0.040000 (  0.037161)
string_brack_range   0.030000   0.000000   0.030000 (  0.034021)
plain_regex          0.040000   0.000000   0.040000 (  0.033468)
string_brack_args    0.020000   0.000000   0.020000 (  0.022455)

That's nearly an eight-fold increase in efficiency between the slowest and fastest. Even some of the little things make a huge difference. Granted, I looped through these 100_000 times, so the difference is pretty small for the one instance, but even something as simple as using using split(str, 2) vs split(str) is well over twice as fast. In fact, the faster half of the approaches listed average 3 times as fast as the slower half.