What's the best way to capture STDOUT into an Array? I looked at
popen3, open4, systemu but couldn't figure out how to do this.
In short, I want to execute a system call like "ls -l" and capture it
into an array. Of course, "ls -l" may not have proper delimiter but
assume that the delimiter is TAB or COMMA.
Thanks,
lines = `ls -l`.split( /\n/ )
On Dec 5, 4:57 pm, Venks <venkatesh.man...@gmail.com> wrote:
What's the best way to capture STDOUT into an Array? I looked at
popen3, open4, systemu but couldn't figure out how to do this.In short, I want to execute a system call like "ls -l" and capture it
into an array. Of course, "ls -l" may not have proper delimiter but
assume that the delimiter is TAB or COMMA.
You can use the IO objects returned by popen3 in any way that you'd use a
normal file. To get an array of lines, one string per line, use
#readlines. To parse it as a CSV, for example, the csv library in ruby
should work with CSV.parse
--Ken
On Wed, 05 Dec 2007 18:57:51 -0500, Venks wrote:
What's the best way to capture STDOUT into an Array? I looked at popen3,
open4, systemu but couldn't figure out how to do this.In short, I want to execute a system call like "ls -l" and capture it
into an array. Of course, "ls -l" may not have proper delimiter but
assume that the delimiter is TAB or COMMA.Thanks,
--
Ken (Chanoch) Bloom. PhD candidate. Linguistic Cognition Laboratory.
Department of Computer Science. Illinois Institute of Technology.
http://www.iit.edu/~kbloom1/
Thanks for your quick reply. I realize that you don't even need to use
split. What I am missing is how to create a 2 dimensional array with
each line contents in an array assuming that they are separated by a
COMMA.
Let me be more specific. Assume that we have a file under
/tmp/test.txt with the following contents.
aaa, 10, test1
bbb, 20, test2
ccc, 30, test3
ddd, 40, test4
I want to create a 2 dimensional array with each line being an array.
lines = `cat /tmp/test.txt` will create a single dimensional array by default.
Thanks,
On Dec 5, 2007 7:14 PM, Phrogz <phrogz@mac.com> wrote:
On Dec 5, 4:57 pm, Venks <venkatesh.man...@gmail.com> wrote:
> What's the best way to capture STDOUT into an Array? I looked at
> popen3, open4, systemu but couldn't figure out how to do this.
>
> In short, I want to execute a system call like "ls -l" and capture it
> into an array. Of course, "ls -l" may not have proper delimiter but
> assume that the delimiter is TAB or COMMA.lines = `ls -l`.split( /\n/ )
Actually, the output is not in a file. The output is generated by a
system call. I somehow think there is something simple in Ruby that I
can use to covert the STDOUT to a 2 dimensional array without actually
using any IO objects.
On Dec 5, 2007 7:39 PM, Ken Bloom <kbloom@gmail.com> wrote:
On Wed, 05 Dec 2007 18:57:51 -0500, Venks wrote:
> What's the best way to capture STDOUT into an Array? I looked at popen3,
> open4, systemu but couldn't figure out how to do this.
>
> In short, I want to execute a system call like "ls -l" and capture it
> into an array. Of course, "ls -l" may not have proper delimiter but
> assume that the delimiter is TAB or COMMA.
>
> Thanks,You can use the IO objects returned by popen3 in any way that you'd use a
normal file. To get an array of lines, one string per line, use
#readlines. To parse it as a CSV, for example, the csv library in ruby
should work with CSV.parse--Ken
--
Ken (Chanoch) Bloom. PhD candidate. Linguistic Cognition Laboratory.
Department of Computer Science. Illinois Institute of Technology.
http://www.iit.edu/~kbloom1/
I'm not so sure about that last comment. As far as I know backticks always return a String.
The easiest way to do what you want is just to split the String from backticks by line and then each line by whatever delimiter you chose. You can do this with map for instance:
[alexg@powerbook]/Users/alexg/Desktop(47): cat test.txt
aaa, 10, test1
bbb, 20, test2
ccc, 30, test3
ddd, 40, test4
[alexg@powerbook]/Users/alexg/Desktop(48): cat capture.rb
p `cat #{ARGV[0]}`.split(/\n/).map{|line| line.split(/, /)}
[alexg@powerbook]/Users/alexg/Desktop(49): ruby capture.rb test.txt
[["aaa", "10", "test1"], ["bbb", "20", "test2"], ["ccc", "30", "test3"], ["ddd", "40", "test4"]]
Alex Gutteridge
Bioinformatics Center
Kyoto University
On 6 Dec 2007, at 09:38, Venks wrote:
Thanks for your quick reply. I realize that you don't even need to use
split. What I am missing is how to create a 2 dimensional array with
each line contents in an array assuming that they are separated by a
COMMA.Let me be more specific. Assume that we have a file under
/tmp/test.txt with the following contents.aaa, 10, test1
bbb, 20, test2
ccc, 30, test3
ddd, 40, test4I want to create a 2 dimensional array with each line being an array.
lines = `cat /tmp/test.txt` will create a single dimensional array by default.
`cat /tmp/test.txt` will create a single string.
require 'csv'
nested_arrays=CSV.parse(`cat /tmp/test.txt`)
--Ken
On Wed, 05 Dec 2007 19:38:35 -0500, Venks wrote:
Thanks for your quick reply. I realize that you don't even need to use
split. What I am missing is how to create a 2 dimensional array with
each line contents in an array assuming that they are separated by a
COMMA.Let me be more specific. Assume that we have a file under /tmp/test.txt
with the following contents.aaa, 10, test1
bbb, 20, test2
ccc, 30, test3
ddd, 40, test4I want to create a 2 dimensional array with each line being an array.
lines = `cat /tmp/test.txt` will create a single dimensional array by
default.
--
Ken (Chanoch) Bloom. PhD candidate. Linguistic Cognition Laboratory.
Department of Computer Science. Illinois Institute of Technology.
http://www.iit.edu/~kbloom1/
Thanks for your reply. It works now. I was looking for the method
"map" over "split".
On Dec 5, 2007 7:53 PM, Alex Gutteridge <alexg@kuicr.kyoto-u.ac.jp> wrote:
On 6 Dec 2007, at 09:38, Venks wrote:
> Thanks for your quick reply. I realize that you don't even need to use
> split. What I am missing is how to create a 2 dimensional array with
> each line contents in an array assuming that they are separated by a
> COMMA.
>
> Let me be more specific. Assume that we have a file under
> /tmp/test.txt with the following contents.
>
> aaa, 10, test1
> bbb, 20, test2
> ccc, 30, test3
> ddd, 40, test4
>
> I want to create a 2 dimensional array with each line being an array.
>
> lines = `cat /tmp/test.txt` will create a single dimensional array
> by default.I'm not so sure about that last comment. As far as I know backticks
always return a String.The easiest way to do what you want is just to split the String from
backticks by line and then each line by whatever delimiter you chose.
You can do this with map for instance:[alexg@powerbook]/Users/alexg/Desktop(47): cat test.txt
aaa, 10, test1
bbb, 20, test2
ccc, 30, test3
ddd, 40, test4
[alexg@powerbook]/Users/alexg/Desktop(48): cat capture.rb
p `cat #{ARGV[0]}`.split(/\n/).map{|line| line.split(/, /)}
[alexg@powerbook]/Users/alexg/Desktop(49): ruby capture.rb test.txt
[["aaa", "10", "test1"], ["bbb", "20", "test2"], ["ccc", "30",
"test3"], ["ddd", "40", "test4"]]Alex Gutteridge
Bioinformatics Center
Kyoto University
Thanks.
Unfortunately I learned the hard way that if I use back ticks to
execute an external command STDERR is not captured. I need to check
the status of the command before proceeding. I am looking into
"systemu" library as I may need to use threads as part of this
development process.
On Dec 6, 2007 11:10 AM, Ken Bloom <kbloom@gmail.com> wrote:
On Wed, 05 Dec 2007 19:38:35 -0500, Venks wrote:
> Thanks for your quick reply. I realize that you don't even need to use
> split. What I am missing is how to create a 2 dimensional array with
> each line contents in an array assuming that they are separated by a
> COMMA.
>
> Let me be more specific. Assume that we have a file under /tmp/test.txt
> with the following contents.
>
> aaa, 10, test1
> bbb, 20, test2
> ccc, 30, test3
> ddd, 40, test4
>
> I want to create a 2 dimensional array with each line being an array.
>
> lines = `cat /tmp/test.txt` will create a single dimensional array by
> default.`cat /tmp/test.txt` will create a single string.
require 'csv'
nested_arrays=CSV.parse(`cat /tmp/test.txt`)--Ken
--
Ken (Chanoch) Bloom. PhD candidate. Linguistic Cognition Laboratory.
Department of Computer Science. Illinois Institute of Technology.
http://www.iit.edu/~kbloom1/
Also, I tried collect! method earlier, but didn't know how to apply
split within collect!. After seeing your example I understood the
proper syntax. I also understand that map and collect! are the same.
On Dec 5, 2007 9:16 PM, Venks <venkatesh.mantha@gmail.com> wrote:
Thanks for your reply. It works now. I was looking for the method
"map" over "split".On Dec 5, 2007 7:53 PM, Alex Gutteridge <alexg@kuicr.kyoto-u.ac.jp> wrote:
> On 6 Dec 2007, at 09:38, Venks wrote:
>
> > Thanks for your quick reply. I realize that you don't even need to use
> > split. What I am missing is how to create a 2 dimensional array with
> > each line contents in an array assuming that they are separated by a
> > COMMA.
> >
> > Let me be more specific. Assume that we have a file under
> > /tmp/test.txt with the following contents.
> >
> > aaa, 10, test1
> > bbb, 20, test2
> > ccc, 30, test3
> > ddd, 40, test4
> >
> > I want to create a 2 dimensional array with each line being an array.
> >
> > lines = `cat /tmp/test.txt` will create a single dimensional array
> > by default.
>
> I'm not so sure about that last comment. As far as I know backticks
> always return a String.
>
> The easiest way to do what you want is just to split the String from
> backticks by line and then each line by whatever delimiter you chose.
> You can do this with map for instance:
>
> [alexg@powerbook]/Users/alexg/Desktop(47): cat test.txt
> aaa, 10, test1
> bbb, 20, test2
> ccc, 30, test3
> ddd, 40, test4
> [alexg@powerbook]/Users/alexg/Desktop(48): cat capture.rb
> p `cat #{ARGV[0]}`.split(/\n/).map{|line| line.split(/, /)}
> [alexg@powerbook]/Users/alexg/Desktop(49): ruby capture.rb test.txt
> [["aaa", "10", "test1"], ["bbb", "20", "test2"], ["ccc", "30",
> "test3"], ["ddd", "40", "test4"]]
>
> Alex Gutteridge
>
> Bioinformatics Center
> Kyoto University
>
>
>
>
You can always use shell redirection to redirect stderr to stdout...
s = p `ls nonexistent_dir 2>&1`
p s
# => "ls: cannot access nonexistent_dir: No such file or directory\n"
Regards,
Jordan
On Dec 6, 10:45 am, Venks <venkatesh.man...@gmail.com> wrote:
Thanks.
Unfortunately I learned the hard way that if I use back ticks to
execute an external command STDERR is not captured. I need to check
the status of the command before proceeding. I am looking into
"systemu" library as I may need to use threads as part of this
development process.On Dec 6, 2007 11:10 AM, Ken Bloom <kbl...@gmail.com> wrote:
> On Wed, 05 Dec 2007 19:38:35 -0500, Venks wrote:
> > Thanks for your quick reply. I realize that you don't even need to use
> > split. What I am missing is how to create a 2 dimensional array with
> > each line contents in an array assuming that they are separated by a
> > COMMA.> > Let me be more specific. Assume that we have a file under /tmp/test.txt
> > with the following contents.> > aaa, 10, test1
> > bbb, 20, test2
> > ccc, 30, test3
> > ddd, 40, test4> > I want to create a 2 dimensional array with each line being an array.
> > lines = `cat /tmp/test.txt` will create a single dimensional array by
> > default.> `cat /tmp/test.txt` will create a single string.
> require 'csv'
> nested_arrays=CSV.parse(`cat /tmp/test.txt`)> --Ken
> --
> Ken (Chanoch) Bloom. PhD candidate. Linguistic Cognition Laboratory.
> Department of Computer Science. Illinois Institute of Technology.
>http://www.iit.edu/~kbloom1/
Venks wrote:
I also understand that map and collect! are the same.
map and collect are the same. As are map! and collect!.
map/collect and map!/collect! however are NOT the same.
Consider this:
arr = [1,2,3]
p arr.map {|x| x+1} # Prints [2, 3, 4]
p arr # Prints [1, 2, 3]
arr = [1,2,3]
p arr.map! {|x| x+1} # Prints [2, 3, 4]
p a # Prints [2, 3, 4] too.
HTH,
Sebastian
--
Jabber: sepp2k@jabber.org
ICQ: 205544826