How to parse a text file and count occurance of a string?

Question

I have text file with a list of dates

How can I count how many times each date is recorded? So have an output something similar to:

2014-01-18 2
2014-01-19 0
2014-01-20 3
2014-01-21 2
2014-01-22 5

possible duplicate of [count duplicate elements in ruby array](http://stackoverflow.com/questions/569694/count-duplicate-elements-in-ruby-array) — Uri Agassi, Apr 08 '14 at 08:54
I agree with @Stefan that using a simple shell command-line is the best way. It's fast, and easy. — the Tin Man, Apr 08 '14 at 09:01

mdesantis · Answer 1 · 2014-04-08T08:59:23.487

4

path = '/path/to/file'
lines = File.readlines(path).map(&:chomp)

# At this point lines should look like below, this is just for testing
lines = ["2014-01-18", "2014-01-18", "2014-01-20", 
         "2014-01-20", "2014-01-20", "2014-01-21", 
         "2014-01-21", "2014-01-22", "2014-01-22", 
         "2014-01-22", "2014-01-22", "2014-01-22"]

# All Ruby versions (since you're using Ruby 1.9.3 you should use tihs)
Hash[ lines.group_by { |v| v }.map { |k, v| [k, v.size] } ]
# Ruby >= 2.1.0
lines.group_by { |v| v }.map { |k, v| [k, v.size] }.to_h

#=> {"2014-01-18"=>2, "2014-01-20"=>3, "2014-01-21"=>2, "2014-01-22"=>5}

edited Apr 08 '14 at 08:59

answered Apr 08 '14 at 08:08

mdesantis

8,257
4
31
63

`Hash.[]` is old fashioned. – sawa Apr 08 '14 at 08:09
I don't like it too, but I'm not aware of alternatives; What would you use? – mdesantis Apr 08 '14 at 08:11
Why not use `Array#to_h`? – sawa Apr 08 '14 at 08:11
How would I turn the text file lines into an array? – grabury Apr 08 '14 at 08:14
1

@sawa Oh! I didn't know it, it seems it was introduced in Ruby 2.1.0, thanks for the info! spuggy I'll write it, which Ruby version are you using? – mdesantis Apr 08 '14 at 08:16
1

@spuggy If you use a legacy version of Ruby, you should express that in the question by adding the relevant tag. – sawa Apr 08 '14 at 08:36

xlembouras · Answer 2 · 2014-04-08T12:35:54.857

3

I like using Hash.new

lines = ["2014-01-18", "2014-01-18", "2014-01-20", 
         "2014-01-20", "2014-01-20", "2014-01-21", 
         "2014-01-21", "2014-01-22", "2014-01-22", 
         "2014-01-22", "2014-01-22", "2014-01-22"]

result = Hash.new(0)
lines.each { |line| result[line] += 1 }

result
# => {"2014-01-18"=>2, "2014-01-20"=>3, "2014-01-21"=>2, "2014-01-22"=>5}

edited Apr 08 '14 at 12:35

answered Apr 08 '14 at 08:22

xlembouras

8,215
4
33
42

2

You don't have to pass a block, just `result = Hash.new(0)` – Stefan Apr 08 '14 at 08:49
1

And, a shorter version, since I'm a one-liner guy: `lines.each_with_object(Hash.new(0)) { |l, h| h[l] += 1 }` – mdesantis Apr 08 '14 at 08:51
@Stefan you are right, I usually use the more verbose block. I edited the answer. – xlembouras Apr 08 '14 at 12:36

score 2 · Answer 3 · answered Apr 08 '14 at 10:41

2

lines = File.readlines('file.txt').map(&:chomp)

op = Hash.new(0)

lines.each do |line|
  op[line.to_sym] += 1
end

puts op.sort_by { |k, v| v }

answered Apr 08 '14 at 10:41

Chezhian

773
9
15

How to parse a text file and count occurance of a string?

3 Answers3