--- trunk/all2xml.pl 2004/12/31 04:22:49 619
+++ trunk/all2xml.pl 2006/04/13 19:44:51 730
@@ -1,7 +1,7 @@
#!/usr/bin/perl -w
use strict;
-use IsisDB;
+use Biblio::Isis;
use Getopt::Std;
use Data::Dumper;
use XML::Simple;
@@ -11,6 +11,7 @@
#use GDBM_File;
use Fcntl; # for O_RDWR
use TDB_File;
+use Carp;
$|=1;
@@ -22,7 +23,8 @@
my $config;
#use index_DBI; # default DBI module for index
-use index_DBI_cache; # faster DBI module using memory cache
+#use index_DBI_cache; # faster DBI module using memory cache
+use index_DBI_filter; # filter support for indexes
my $index;
my %opts;
@@ -61,7 +63,7 @@
'isis' => 'isis',
'excel' => 'column',
'marc' => 'marc',
- 'feed' => 'feed'
+ 'feed' => 'feed',
);
my $cache; # for cacheing
@@ -111,6 +113,10 @@
$cache->{tags_by_order} = \@sorted_tags;
}
+ if (! @sorted_tags) {
+ print STDERR "WARNING: no tags for this type found in import_xml file!\n";
+ }
+
# lookup key
my $lookup_key;
@@ -120,6 +126,7 @@
delete $cache->{swish_exact_data};
delete $cache->{index_data};
delete $cache->{index_delimiter};
+ delete $cache->{distinct};
my @page_fields; # names of fields
@@ -219,7 +226,7 @@
my ($swish,$display);
- my $tag = $type2tag{$type} || die "can't find which tag to use for type $type";
+ my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type} || die "can't find which tag to use for type $type";
# is this field page-by-page?
my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page};
@@ -228,6 +235,11 @@
# default line_delimiter if using
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '
';
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter};
+ my $distinct = $config->{indexer}->{$field}->{distinct};
+ if ($distinct && !$iterate_by_page) {
+ warn "WARNING: distinct is currently not supported without iterate_by_page!\n";
+ $distinct = 0;
+ }
my $format_name = $config->{indexer}->{$field}->{format_name};
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter};
@@ -257,8 +269,6 @@
# init vars so that we go into while...
($swish,$display) = (1,1);
- # placeholder for all repeatable entries for index
-
sub mkformat($$) {
my $x = shift || die "mkformat needs tag reference";
my $data = shift || return;
@@ -408,9 +418,10 @@
$ldel = " " if ($append);
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
if (! $cache->{$what}->{$field}->[$page]) {
- $cache->{$what}->{$field}->[$page] = $data;
- } else {
- $cache->{$what}->{$field}->[$page] .= $ldel.$data;
+ push @{$cache->{$what}->{$field}->[$page]}, {
+ data => $data,
+ delimiter => $ldel,
+ };
}
}
@@ -485,15 +496,15 @@
# on first page!!!
my $page = 0;
if ($display_data) {
- $cache->{display_data}->{$field}->[$page] = $display_data;
+ push @{$cache->{display_data}->{$field}->[$page]}, { data => $display_data };
$display_data = "";
}
if ($swish_data) {
- $cache->{swish_data}->{$field}->[$page] = $swish_data;
+ push @{$cache->{swish_data}->{$field}->[$page]}, { data => $swish_data };
$swish_data = "";
}
if ($swish_exact_data) {
- $cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data;
+ push @{$cache->{swish_exact_data}->{$field}->[$page]}, { data => $swish_exact_data };
$swish_exact_data = "";
}
}
@@ -504,14 +515,28 @@
my $nr_pages = $page_max{$field} || next;
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
#print STDERR Dumper($cache->{display_data});
+ my $seen; # used for distinct
for (my $page=0; $page <= $nr_pages; $page++) {
my $display_data;
- if ($cache->{format}->{$field}) {
- my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]);
- $display_data=$tmp if ($tmp);
- } else {
- $display_data = $cache->{display_data}->{$field}->[$page];
+ my $delimiter = '';
+ foreach my $element (@{ $cache->{display_data}->{$field}->[$page] }) {
+ my $data = $element->{data};
+ die "BUG! no data in element?" unless ($data);
+
+ if ($distinct) {
+ next if ($cache->{distinct}->{$field}->{ $data });
+ $cache->{distinct}->{$field}->{ $data } = 1;
+ }
+
+ if ($cache->{format}->{$field}) {
+ my $tmp = mkformat($cache->{format}->{$field},$data);
+ $display_data .= $delimiter . $tmp if ($tmp);
+ } else {
+ $display_data .= $delimiter . $data;
+ }
+ $delimiter = $element->{delimiter} if ($element->{delimiter});
}
+
if ($display_data) { # default
if ($field eq "headline") {
$xml .= xmlify("headline", $display_data);
@@ -523,7 +548,7 @@
}
}
- my $swish_data = $cache->{swish_data}->{$field}->[$page];
+ my $swish_data = join(" ",map { $_->{data} } @{ $cache->{swish_data}->{$field}->[$page] });
if ($swish_data) {
# remove extra spaces
$swish_data =~ s/ +/ /g;
@@ -532,7 +557,7 @@
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
}
- my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
+ my $swish_exact_data = join(" ", map { $_->{data} } @{ $cache->{swish_exact_data}->{$field}->[$page] });
if ($swish_exact_data) {
$swish_exact_data =~ s/ +/ /g;
$swish_exact_data =~ s/ +$//g;
@@ -634,6 +659,10 @@
foreach my $database ($cfg->Sections) {
+ # save database name in global variable path for later
+ # (need for index filter creation)
+ $path = $database;
+
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
my $add_xml = $cfg -> val($database, 'xml'); # optional
@@ -659,13 +688,22 @@
print STDERR "opening lookup file '$lookup_file'\n";
}
-print STDERR "reading ./import_xml/$type.xml\n";
+ my $import_xml_type = $cfg->val($database, 'import_xml_file') || $type;
+ my $import_xml_file = "./import_xml/$import_xml_type.xml";
+
+ if (! -r $import_xml_file) {
+ print STDERR "ERROR: file $import_xml_file not readable skipping!\n";
+ next;
+ }
+
+ print STDERR "reading $import_xml_file\n";
# extract just type basic
my $type_base = $type;
$type_base =~ s/_.+$//g;
- $config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 );
+ my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type_base} || die "can't find which tag to use for type $type";
+ $config=XMLin($import_xml_file, ForceArray => [ $tag, 'config', 'format' ], ForceContent => 1 );
# helper for progress bar
sub fmt_time {
@@ -684,10 +722,13 @@
my $start_t = time();
sub progress {
return if (! $show_progress);
- my $current = shift || 1;
+ my $current = shift;
my $total = shift || 1;
my $p = int($current * 100 / $total);
- if ($p != $last_p) {
+ if ($p < $last_p || $current == 1) {
+ $start_t = time();
+ $last_p = 0;
+ } elsif ($p != $last_p) {
my $rate = ($current / (time() - $start_t || 1));
my $eta = ($total-$current) / ($rate || 1);
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta));
@@ -696,21 +737,34 @@
}
my $fake_dir = 1;
+ my $fake_pos = 0;
+ my $last_fake_t = time();
sub fakeprogress {
return if (! $show_progress);
my $current = shift @_;
- my @ind = ('-','\\','|','/','-','\\','|','/', '-');
+ my @ind = ('-','\\','|','/','-','\\','|','/');
- $last_p += $fake_dir;
- $fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0);
- if ($last_p % 10 == 0) {
- printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]);
+ if ($current < $fake_pos) {
+ $start_t = time();
+ $last_fake_t = 0;
+ $fake_dir = 1;
+ $fake_pos = 0;
+ }
+
+ if (time()-$last_fake_t >= 1) {
+ $last_fake_t = time();
+ $fake_pos += $fake_dir;
+ $fake_dir = -$fake_dir if ($fake_pos > 38);
+ }
+
+ if ($current % 10 == 0) {
+ my $rate = ($current / (time() - $start_t || 1));
+ printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate);
}
}
# now read database
-print STDERR "using: $type...\n";
# erase cache for tags by order in this database
delete $cache->{tags_by_order};
@@ -720,13 +774,16 @@
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
- my $db = new IsisDB( isisdb => $isis_db );
+ my $db = new Biblio::Isis( isisdb => $isis_db );
- my $max_rowid = $db->{'maxmfn'} || die "can't find maxmfn";
+ if (! $db) {
+ print STDERR "FATAL: can't read ISIS database: $isis_db, skipping...\n";
+ next;
+ }
- print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
+ my $max_rowid = $db->count if ($db);
- my $path = $database;
+ print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
my $row = $db->to_hash( $row_id );
@@ -759,7 +816,8 @@
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
- my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml";
+ my $start_row = x($config->{start_row}) || die "no start_row in $type.xml";
+ $start_row--;
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
@@ -809,44 +867,57 @@
print "Document-Type: XML\n\n$xml\n";
}
}
+
+ print STDERR "\n";
+
} elsif ($type_base eq "marc") {
- require MARC;
+ require MARC::File::USMARC;
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
# optional argument is format
- my $format = x($config->{marc_format}) || 'usmarc';
-
+ warn "marc_format is no longer used!" if ($config->{marc_format});
print STDERR "Reading MARC file '$marc_file'\n";
- my $marc = new MARC;
- my $nr = $marc->openmarc({
- file=>$marc_file, format=>$format
- }) || die "Can't open MARC file '$marc_file' with format '$format'";
+ my $marc = MARC::File::USMARC->in( $marc_file );
+
+ if (! $marc) {
+ print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n";
+ next;
+ }
- # read MARC file in memory
- $marc->nextmarc(-1);
+ # count records in MARC file
+ sub marc_count {
+ my $filename = shift || die;
+ my $file = MARC::File::USMARC->in($filename) || return;
+ my $count = 0;
+ while ($file->skip()) {
+ $count++;
+ }
+ return $count;
+ }
- my $max_rec = $marc->marc_count();
+ my $count = marc_count($marc_file) || warn "no records in '$marc_file'?";
- for(my $i=1; $i<=$max_rec; $i++) {
+ my $i = 1;
- progress($i,$max_rec);
+ while( my $rec = $marc->next() ) {
- # store value for marc_sf.pm
- $main::cache->{marc_record} = $i;
+ progress($i,$count);
my $swishpath = $database."#".$i;
- if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) {
+ if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) {
$xml = $cp2utf->convert($xml);
use bytes; # as opposed to chars
print "Path-Name: $swishpath\n";
print "Content-Length: ".(length($xml)+1)."\n";
print "Document-Type: XML\n\n$xml\n";
}
+
+ $i++;
}
print STDERR "\n";
@@ -891,6 +962,72 @@
}
# close lookup
untie %lhash if (%lhash);
+
+ } elsif ($type_base eq "dbf") {
+
+ my $dbf_file = $cfg -> val($database, 'dbf_file') || die "$database doesn't have 'dbf_file' defined!";
+ my $dbf_codepage = $cfg -> val($database, 'dbf_codepage') || die "$database doesn't have 'dbf_codepage' defined!";
+ my $dbf_mapping = $cfg -> val($database, 'dbf_mapping') || die "$database doesn't have 'dbf_mapping' defined!";
+
+ $import2cp = Text::Iconv->new($dbf_codepage,$codepage);
+ require XBase;
+ my $db = new XBase $dbf_file;
+
+ if (! $db) {
+ print STDERR "ERROR: can't read DBF database: $dbf_file, skipping...\n";
+ next;
+ }
+
+ my $max_rowid = $db->last_record;
+
+ print STDERR "Reading database: $dbf_file [$max_rowid rows]\n";
+
+ my %dbf2iso;
+ foreach my $m (split(/[\n\r]+/,$dbf_mapping)) {
+ my ($col,$fld) = split(/\s+/,$m,2);
+ $dbf2iso{$col} = $fld;
+ }
+
+#print STDERR "## dbf2iso: ",Dumper(\%dbf2iso),"\n## /dbf2iso\n";
+
+ # bad, bad...
+ require "to_hash.pm";
+
+ foreach my $row_id (0 .. $max_rowid) {
+ my $dbf_row = $db->get_record_as_hash($row_id);
+ if ($dbf_row) {
+
+#print STDERR "## dbf_row: ",Dumper($dbf_row),"\n## /dbf_row\n";
+ # apply mapping from config file
+ # all unspecified records will get _ in
+ # front of them - _DELETE will be __DELETE
+ my $rec;
+ map {
+ my $new_fld = $dbf2iso{$_} || '_'.$_;
+ my $data = $dbf_row->{$_};
+ push @{ $rec->{$new_fld} }, $data if ($data && $data !~ /^(?:\s+|\$a\.|)$/);
+ } keys %{$dbf_row};
+#print STDERR "## rec: ",Dumper($rec),"\n## /rec\n";
+ my $row = to_hash($row_id+1, $rec);
+
+ $row->{mfn} = $row_id+1;
+ $row->{record} = $rec;
+
+#print STDERR "## row: ",Dumper($row),"\n## /row\n";
+ progress($row->{mfn}, $max_rowid);
+
+ my $swishpath = $path."#".int($row->{mfn});
+
+ if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
+ $xml = $cp2utf->convert($xml);
+ use bytes; # as opposed to chars
+ print "Path-Name: $swishpath\n";
+ print "Content-Length: ".(length($xml)+1)."\n";
+ print "Document-Type: XML\n\n$xml\n";
+ }
+ }
+ }
+ print STDERR "\n";
}
}
@@ -911,8 +1048,8 @@
=head1 DESCRIPTION
-This command will read ISIS data file using OpenIsis perl module, MARC
-records using MARC module and optionally Micro$oft Excel files to
+This command will read ISIS data file using Biblio::Isis perl module, MARC
+records using MARC::File module and optionally Micro$oft Excel files to
create one XML file for usage with I indexer. Dispite it's name,
this script B from isis files (isis allready
has something like that). Output of this script is tailor-made for SWISH-E.