Help a noob out.

Quick Perl example on a random XML feed, showing all the features you mention needing: #!/usr/bin/perl -luse strict;use LWP::Simple;use XML::Simple;use Text::CSV;my $csv = Text::CSV->new;my $xml = XMLin( get "http://feeds.delicious.com/v2/rss" );for my $item ( @{ $xml->{channel}->{item} } ) { $item->{title} =~ s/_/-/g; my @parts = split '/', $item->{link}; $csv->combine( $item->{title}, @parts ); print $csv->string; }

This would be trivial in Perl with the REST::Client, XML::Simple, and Text::CSV_XS modules. Perl was made for text transformations on data.

You're in good shape so far. You've got two basic philosophical choices here. 1 - You can walk through your xml data, pick out the elements you want to do stuff with, and write the CSV rows to a StringBuilder. This might be easier to wrap your head around. For small files, a DOM parser might be your best bet. 2 - Or, you can use XSLT to change the XML document directly to the CSV that you want. It's a bit more abstract, and the XSLT language is a beast unto itself. But, it does give you a terse and powerful way to do what you're looking for, with less lines of total code. I kind of wish I knew what the full ask is, so I can say one way or the other which is the best way to go. Anyway, since you say you already have a document in memory, this is a good place to start for option 1: http://docs.oracle.com/javase/6/docs/api/org/w3c/dom/Document.html Something like Document.getElementsByTagName(), along with a loop, might just get you the rest of the way. Good luck!

Putting in a plug for the http://groovy.codehaus.org/ programming language, which has taken this non-programmer further than I thought I could ever go. The syntax is amazingly intuitive: even if you don't know exactly what you're doing, a lot of times you can enter some parens or some other character that seems to make sense and lo and behold, it does what you want it to. XML processing with Groovy, in particular, is pretty pain-free. There is a great http://docs.codehaus.org/display/GROOVY/Mailing+Lists (user at groovy dot codehaus dot org) on which I frequently post "newb, please help" type questions and I always get at least one helpful answer. If you really want to learn programming, I've heard that it's best to start with Java or C/C#/C++ and go on to the "syntactic sugar" languages like Groovy, but if your job doesn't require you to be a full-fledged programmer, a language like Groovy can help you immensely. Python is also great, but be prepared to wrestle with getting packages if you are behind an NTLM (Windows) proxy at work and there aren't any Python programmers at your job to help you out. I haven't found that to be the case with Groovy.

Thanks for the updating and commenting zengargoyle! Aizkolari, is the Perl script clear to you now? Whatever solution you use, please don't try to make valid CSV "manually" by joining with commas as in substars' Ruby example – use some kind of CSV library. Otherwise a comma or newline in the data will break your script, and then you'll try surrounding all the values with double quotes, and then a double quote in the data will break your script...

Oh, you might want to play with the 'lwp-request' program that's installed with Strawberry / LWP::UserAgent. See: perldoc lwp-request...> lwp-request -m GET -H "ACCEPT: application/xml" ... https://...It's a simple CLI wrapper around LWP::UserAgent so you can play with flags and see results before/while hacking on your code. And your previous try with HTTP::Request would go something like:$res = $ua->request($r);$xml = XMLin( $res->content );

Yeah, +1 to nicwolff and mea culpa; was just trying not to confuse things with a bunch of libraries. Ruby has FasterCSV, which is analogous to Text::CSV in Perl.

See: LWP::UserAgent The non-Simple version of the doohickey.use LWP::UserAgent;my $ua = LWP::UserAgent->new;my $res = $ua->get( "http://localhost", "ACCEPT" => "application/xml", "KEY-HEADER-NAME" => "foo",);if ( $res->is_success ) { say $res->decoded_content;}else { die $res->status_line;}exit;perldoc will tell you what you need to know, LWP::UserAgent is the full web-client thing with user agent strings, cookie-jars, headers, SSL certs and the like. Here is the stuff the above sends to a server: # nc -l 80GET / HTTP/1.1TE: deflate,gzip;q=0.3Connection: TE, closeAccept: application/xmlHost: localhostUser-Agent: libwww-perl/6.04KEY-HEADER-NAME: foo Nitpick time...# 2-arg open is frowned upon (FH is a global here)open FH, '>>Output.csv' or die "$!";# -> 3 arg open with local varsopen my $outfile, '>>', 'Output.csv' or die "$!";print $outfile $csv->string, "\n"; # you can use , vs . here, print takes argument list.close $outfile or die "$!"; # if you're not checking for errors like full filesysem...# Perl will close the file automatically at exit.# optional: replace checking with 'autodie' pragma.use autodie;open my $fh, '>', 'Outfile.csv';# dies with "$!" on error automagically...# you may need to turn off ssl host verification for local certs$ua->ssl_opts( verify_hostname => 0 ); # see perldocLooks like you found the lower level objects used: HTTP::Request, HTTP::Response...

I'll add that it may be helpful to know about the MetaCPAN site. Pretty much every Perl module somehow comes from the CPAN and you'll find decently formatted documentation: https://metacpan.org/module/XML::Simple And you can look at some of the files that aren't installed with the module but are part of the distibution as a whole. Since Perl is generally awesome when it comes to testing, you can get a bunch of example usage by browsing some of the tests: https://metacpan.org/source/GRANTM/XML-Simple-2.18/t/1_XMLin.t And the while you can use 'perldoc' in a terminal, often the HTML version on MetaCPAN is a much easier read with all the linking and syntax-highlighting. The core Perl documentation is also available: http://perldoc.perl.org/perl.html The first 3 tutorials may be of help in digging out your data from $xml. perlreftut Perl references short introduction perldsc Perl data structures intro perllol Perl data structures: arrays of arrays If you get lost along the way memail me. From the original description, what you're trying to do isn't that hard, but the evil of XML and deep data structures may not be the easiest first Perl hackery.