# Parsing CSV file with ruby

**URL:** <https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967>\
**Category:** Ruby\
**Created:** [August 30, 2006, 5:57pm UTC](https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967 "2006-08-30T17:57:31Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![dfg59](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/e0b2c6/32.png) [@dfg59](https://www.ruby-forum.com/u/dfg59)\
**Post date:** [August 30, 2006, 5:57pm UTC](https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967/1 "2006-08-30T17:57:31Z")

</div>

I’m currently trying to do something that seems rather simple but I’m  
slightly new to ruby. I want to read in a cvs file, find rows that are  
distinct with respect to one of the elements in the row (for example,  
all rows in which the first element is “A”) and then do something with  
these rows (in this case, parse them, build some XML and write it to a  
file). I’m not familiar enough with iterators in ruby but I seem to  
remember there being functionality that will allow me to get distinct  
rows based on some element in the row. Let me know if this is possible  
and how I should approach it.

Thanks,  
Drew

---

<div class="post-metadata">

**Author:** ![dfg59](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/e0b2c6/32.png) [@dfg59](https://www.ruby-forum.com/u/dfg59)\
**Post date:** [August 30, 2006, 6:01pm UTC](https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967/2 "2006-08-30T18:01:01Z")

</div>

Let me be more specific: essentially I want to find the groups of rows  
that share an element. Let’s say each row in my CVS doc has 3 elements.  
I want to iterate across every group of rows that share the same value  
for the first element. Hope this makes sense.

---

<div class="post-metadata">

**Author:** ![dfg59](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/e0b2c6/32.png) [@dfg59](https://www.ruby-forum.com/u/dfg59)\
**Post date:** [August 30, 2006, 6:59pm UTC](https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967/3 "2006-08-30T18:59:42Z")

</div>

Drew O. wrote:

> Let me be more specific: essentially I want to find the groups of rows  
> that share an element. Let’s say each row in my CVS doc has 3 elements.  
> I want to iterate across every group of rows that share the same value  
> for the first element. Hope this makes sense.

#!/usr/bin/ruby -w

row\_hash = {}

File.open(“data.txt”).each { |record|  
fields = record.split(",")  
row\_hash[fields.first] = [] unless row\_hash[fields.first]  
row\_hash[fields.first] \<\< record  
}

row\_hash.keys.sort.each { |key|  
puts “Group: #{key}”  
row\_hash[key].each { |record|  
puts “\t#{record}”  
}  
}

data.txt:

a,this,is,one,record  
a,this,is,another,record  
b,this,is,one,record  
b,this,is,another,record  
c,this,is,one,record  
c,this,is,another,record

output:

Group: a  
a,this,is,one,record  
a,this,is,another,record  
Group: b  
b,this,is,one,record  
b,this,is,another,record  
Group: c  
c,this,is,one,record  
c,this,is,another,record

---

<div class="post-metadata">

**Author:** ![dfg59](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/e0b2c6/32.png) [@dfg59](https://www.ruby-forum.com/u/dfg59)\
**Post date:** [August 30, 2006, 8:47pm UTC](https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967/4 "2006-08-30T20:47:30Z")

</div>

Thank you both for the responses. Both seem to be EXTREMELY helpful.  
I’ll be sure to post issues I have in the form in the future.

Thanks,  
Drew

---

<div class="post-metadata">

**Author:** ![dfg59](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/e0b2c6/32.png) [@dfg59](https://www.ruby-forum.com/u/dfg59)\
**Post date:** [August 30, 2006, 7:07pm UTC](https://www.ruby-forum.com/t/parsing-csv-file-with-ruby/66967/5 "2006-08-30T19:07:44Z")

</div>

On Aug 30, 2006, at 11:01 AM, Drew O. wrote:

> Let me be more specific: essentially I want to find the groups of rows  
> that share an element. Let’s say each row in my CVS doc has 3  
> elements.  
> I want to iterate across every group of rows that share the same value  
> for the first element. Hope this makes sense.

I’m assuming you meant CSV (not CVS). 😉

See if this gets you going:

Firefly:~/Desktop$ cat data.csv  
one,1,A  
one,2,B  
one,3,C  
two,1,A  
two,2,B  
three,1,A  
Firefly:~/Desktop$ irb -r csv

> > rows = CSV.read(“data.csv”)  
> > =\> [[“one”, “1”, “A”], [“one”, “2”, “B”], [“one”, “3”, “C”], [“two”,  
> > “1”, “A”], [“two”, “2”, “B”], [“three”, “1”, “A”]]
> 
> > groups = rows.map { |row| row.first }.uniq  
> > =\> [“one”, “two”, “three”]
> 
> > groups.each do |group|  
> > ?\> puts group
> 
> > rows.select { |row| row.first == group }.each { |row| puts " #  
> > {row.inspect}" }
> 
> > end  
> > one  
> > [“one”, “1”, “A”]  
> > [“one”, “2”, “B”]  
> > [“one”, “3”, “C”]  
> > two  
> > [“two”, “1”, “A”]  
> > [“two”, “2”, “B”]  
> > three  
> > [“three”, “1”, “A”]  
> > =\> [“one”, “two”, “three”]

James Edward G. II