Import CSV data into a wiki

From Organic Design wiki
Revision as of 03:47, 29 May 2008 by Sven (talk | contribs) (Add API url to comment header)
  1. !/usr/bin/perl
  2. Our Perl scripts.{{#security:*|sysop}}Automated scripts to perform batch automation.
  3. - Licenced under LGPL (http://www.gnu.org/copyleft/lesser.html)
  4. - Author: http://www.organicdesign.co.nz/nad
  5. - Source: http://www.organicdesign.co.nz/scraper.pl
  6. - Started: 2008-03-21
  7. - API: http://en.wikipedia.org/w/api.php

require('wiki.pl');

  1. Job, log and error files

$ARGV[0] or die "No job file specified!"; $ARGV[0] =~ /^(.+?)(\..+?)?$/;

  1. Set a debug conditional

$::debug = 0;

$::log = "$1.log"; $::err = "$1.err"; $::sep = ','; $::title = 0; $::template = 'Record';

  1. Parse the job file

if (open JOB,'<',$ARGV[0]) { for (<JOB>) { if (/^\*?\s*csv\s*:\s*(.+?)\s*$/i) { $::csv = $1 } if (/^\*?\s*wiki\s*:\s*(.+?)\s*$/i) { $::wiki = $1 } if (/^\*?\s*user\s*:\s*(.+?)\s*$/i) { $::user = $1 } if (/^\*?\s*pass\s*:\s*(.+?)\s*$/i) { $::pass = $1 } if (/^\*?\s*separator\s*:\s*"(.+?)"\s*$/i) { $::sep = $1 } if (/^\*?\s*title\s*:\s*(.+?)\s*$/i) { $::title = $1 } if (/^\*?\s*template\s*:\s*(.+?)\s*$/i) { $::template = $1 } } close JOB; } else { die "Couldn't parse job file!" }


  1. Open CSV file and read in headings line

if (open CSV,'<',$::csv) { $_ = <CSV>; /^\s*(.+?)\s*$/; @headings = split /$::sep/i, $1; } else { die "Could not open CSV file!" }

  1. Log in to the wiki

wikiLogin($::wiki,$::user,$::pass) or exit;

  1. fetch the template if it exists

$response = $client->get("$wiki?title=$template&action=raw"); if( $response->is_success ) {

 $wikitext = $response->content;
 # Remove noinclude areas
 $wikitext =~ s/.+?<\/noinclude>//gs;
 # Find all unique {{{parameters}}}
 # http://en.wikipedia.org/wiki/Help:Templates#Parameters
 while ($wikitext =~ m/\{{{(.+?)}}}/g ) {
   $key = $1;
   # Remove default anchors if there are any
   $key =~ s/\|.*//;
   $params{$key} = undef;
 }

 # Need to check @headings elements that are identical to keys(%param) 
 for( $i = 0; $i < @headings; $i++ ) {
     
   if(exists($params{$headings[$i]})) {
     print "$headings[$i] matches!\n";
     push(@cols ,$i);
   }
 }
 
 if($::debug) {
   print "\@A: @headings\n";
   print "%H: @{[%params]}\n";
   print "\@cols @cols\n";
   die "[\$::debug set exiting]\n" ;
 }

}


  1. Get batch size and current number (also later account for n-bots)
  1. todo: log batch start
  1. Process the records

$n = 1; while (<CSV>) { /^\s*(.+?)\s*$/; @record = split /$::sep/, $1; $tmpl = "{{$template\n"; # JUST NEED TO FIX UP HERE $tmpl .= "|$headings[$_] = $record[$_]\n" for 0..$#headings; $tmpl .= "}}"; print "Processing record ".$n++."\n";

# Update the record $text = wikiRawPage($::wiki,$record[$::title],0); $text .= "\n$tmpl" unless $text =~ s/\{\{$template.+?\}\}/$tmpl/is; $done = wikiPageEdit($::wiki,$record[$::title],$text,"$template updated by csv2wiki.pl");

# log a row error if any }

close CSV;