How to extract path from a string in ruby (between 1st and last fwd slash inclusive)

Question

I have been writing a ruby script that goes through a text file and locates all lines that begin with output path and stores it into a string (linefromtextfile) for that line. So typically it locates lines as below

"output_path":"/data/server/output/1/test_file.txt","text":
"output_path":"/data/server/output/2/test_file.txt","text":

And I want to extract from the lines the pathname (pathtokeep) only and write out to a file, i.e:

/data/server/output/1/
/data/server/output/2/

I have tried this RegEx but its not working:

pathtokeep=linefromtextfile.split(?:\$/.*?/)([^/]*?\.\S*)

Please someone advise here on my RegEx - is split the right way to go or is there an easier way to do this?

You don't need to thank the author of each answer. If you do, one day you'll face the choice of thanking someone for a poor answer or omitting the thank you for just that one person, leaving the obvious implication. If you look at other questions you'll see it's just not done at SO. — Cary Swoveland, Apr 11 '16 at 18:33
In future, when you give an example it's helpful to assign a variable to each input object. Here that might be `str = '"output_path":"...xt":'`. That way, readers can refer to those variables in answers and comments without having to define them. — Cary Swoveland, Apr 11 '16 at 18:40

guitarman · Accepted Answer · 2017-01-08T12:38:21.883

3

If your file has the always the same structure you could do it without a regex too.

line = '"output_path":"/data/server/output/1/test_file.txt","text":'

path = line.split(/:"|",/)[1]
# => "/data/server/output/1/test_file.txt"

basename = File.basename(path)
# => "test_file.txt"

File.dirname(path) + '/'
# => "/data/server/output/1/"

edited Jan 08 '17 at 12:38

answered Apr 09 '16 at 18:53

guitarman

3,290
1
17
27

I improved it with `File.dirname(path)` instead of `path.gsub(base name, '')` – guitarman Jan 08 '17 at 12:40

Cary Swoveland · Answer 2 · 2016-04-11T19:06:47.597

I suggest using Ruby methods to the extent that you can, employing a regex only to extract the path from the string.

str = '"output_path":"/data/server/output/1/test_file.txt","text":'

r = /
    :"      # match a colon and double quote
    (.+?)   # match one or more of any character, lazily, in capture group 1 
    "       # match a double quote
    /x      # free-spacing regex definition mode

File.dirname(str[r,1])
  #=> "/data/server/output/1"

If you really want the trailing forward slash,

File.dirname(str[r,1]) << "/"
  #=> "/data/server/output/1/"

Should you need it,

File.basename(str[r,1])
  #=> "test_file.txt"

I will leave it to the OP to read and write to files.

If you insist on using a single regex, you could write:

r = /
    (?<=:") # match a colon followed by a double-quote in a positive lookbehind
    .+      # match one more characters, greedily
    \/      # match a forward slash
    /x

str[r]
  #=> "/data/server/output/1/"

Note that .+, being greedy, gobbles up all characters until it reaches the last forward slash in the string.

Thanks to @guitarman, whose answer reminded me that I don't need to escape quotes in a regex. — Cary Swoveland, Apr 09 '16 at 19:33

score 0 · Answer 3 · answered Apr 09 '16 at 16:06

0

Try this RegEx:

(?<="output_path":")(.*?)(?=")

Live Demo on Regex101

How it works:

(?<="output_path":")     # Lookbehind for "output_path":"
(.*?)                    # Data inside "" (Lazy)
(?=")                    # Lookahead for closing "

answered Apr 09 '16 at 16:06

Kaspar Lee

5,446
4
31
54

Thanks for your feedback. Also for the link to Regex101 - handy! – adamjth Apr 11 '16 at 17:29
@adamjth If one of these answers answered your question, please mark it as accepted (press the tick underneath the voting buttons) – Kaspar Lee Apr 12 '16 at 11:06

How to extract path from a string in ruby (between 1st and last fwd slash inclusive)

3 Answers3