$| = 1; my $baseURL = 'http://union.ndltd.org/OAI-PMH/'; my $filename = 'aaaaafna'; my $resumptionToken = '2019-01-22T00:51:30Z!2037-01-01T00:00:00Z!!oai_dc!6045389!7351939!oai:union.ndltd.org:IBICT/aoi:localhost:jspui/2251'; use LWP::UserAgent; $ua = LWP::UserAgent->new; # before running this script, execute: # export http_proxy=http://localhost:/ where is your cntlm port $ua->env_proxy(); do { my $reqURL = $baseURL.'?verb=ListRecords&'.(($ eq '')?'metadataPrefix=oai_dc':'resumptionToken='.$resumptionToken); # my $reqURL = $baseURL.'?verb=Identify'; my $req = HTTP::Request->new( GET => $reqURL ); print "Harvesting $reqURL\n"; my $state = 0; my $res; while ($state == 0) { $res = $ua->request($req); if ($res->code == 503) { my $sleep = $res->header ('Retry-After'); if (not defined ($sleep) || ($sleep < 0) || ($sleep > 86400)) { $state = 1;} else { print "Sleeping for $sleep seconds\n"; sleep ($sleep); } } else { $state = 1; } } my $content = $res->content; my $records = (split (//, $content))-1; print "Saving response with $records records to $filename.xml\n"; open (FILE, ">$filename.xml"); print FILE $content; close (FILE); $filename++; $resumptionToken = ''; if ($content =~ /]*>([^<]+)<\/resumptionToken>/) { $resumptionToken = $1; } } while ($resumptionToken ne '');