13 |
|
|
14 |
use Data::Dumper; |
use Data::Dumper; |
15 |
|
|
16 |
|
my ($have_biblio_isis, $have_openisis) = (0,0); |
17 |
|
|
18 |
|
eval "use Biblio::Isis 0.13;"; |
19 |
|
unless ($@) { |
20 |
|
$have_biblio_isis = 1 |
21 |
|
} else { |
22 |
|
eval "use OpenIsis;"; |
23 |
|
$have_openisis = 1 unless ($@); |
24 |
|
} |
25 |
|
|
26 |
#my $LOOKUP_REGEX = '\[[^\[\]]+\]'; |
#my $LOOKUP_REGEX = '\[[^\[\]]+\]'; |
27 |
#my $LOOKUP_REGEX_SAVE = '\[([^\[\]]+)\]'; |
#my $LOOKUP_REGEX_SAVE = '\[([^\[\]]+)\]'; |
28 |
my $LOOKUP_REGEX = 'lookup{[^\{\}]+}'; |
my $LOOKUP_REGEX = 'lookup{[^\{\}]+}'; |
46 |
config_file => 'name.conf', |
config_file => 'name.conf', |
47 |
code_page => 'ISO-8859-2', |
code_page => 'ISO-8859-2', |
48 |
low_mem => 1, |
low_mem => 1, |
49 |
|
filter => { |
50 |
|
'lower' => sub { lc($_[0]) }, |
51 |
|
}, |
52 |
); |
); |
53 |
|
|
54 |
Default C<code_page> is C<ISO-8859-2>. |
Default C<code_page> is C<ISO-8859-2>. |
55 |
|
|
56 |
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
Default is not to use C<low_mem> options (see L<MEMORY USAGE> below). |
57 |
|
|
58 |
|
There is optinal parametar C<filter> which specify different filters which |
59 |
|
can be applied using C<filter{name}> notation. |
60 |
|
Same filters can be used in Template Toolkit files. |
61 |
|
|
62 |
This method will also read configuration files |
This method will also read configuration files |
63 |
C<global.conf> (used by indexer and Web font-end) |
C<global.conf> (used by indexer and Web font-end) |
64 |
and configuration file specified by C<config_file> |
and configuration file specified by C<config_file> |
123 |
# create Template toolkit instance |
# create Template toolkit instance |
124 |
$self->{'tt'} = Template->new( |
$self->{'tt'} = Template->new( |
125 |
INCLUDE_PATH => ($self->{'global_config_file'}->{'output_template'} || './output_template'), |
INCLUDE_PATH => ($self->{'global_config_file'}->{'output_template'} || './output_template'), |
126 |
# FILTERS => { |
FILTERS => $self->{'filter'}, |
|
# 'foo' => \&foo_filter, |
|
|
# }, |
|
127 |
EVAL_PERL => 1, |
EVAL_PERL => 1, |
128 |
); |
); |
129 |
|
|
153 |
$self->{'db'} = $db; |
$self->{'db'} = $db; |
154 |
} |
} |
155 |
|
|
156 |
|
$log->debug("filters defined: ",Dumper($self->{'filter'})); |
157 |
|
|
158 |
return $self; |
return $self; |
159 |
} |
} |
160 |
|
|
161 |
=head2 open_isis |
=head2 open_isis |
162 |
|
|
163 |
Open CDS/ISIS database using OpenIsis module and read all records to memory. |
Open CDS/ISIS, WinISIS or IsisMarc database using IsisDB or OpenIsis module |
164 |
|
and read all records to memory. |
165 |
|
|
166 |
$webpac->open_isis( |
$webpac->open_isis( |
167 |
filename => '/data/ISIS/ISIS', |
filename => '/data/ISIS/ISIS', |
209 |
$self->{'isis_filename'} = $arg->{'filename'}; |
$self->{'isis_filename'} = $arg->{'filename'}; |
210 |
$self->{'isis_code_page'} = $code_page; |
$self->{'isis_code_page'} = $code_page; |
211 |
|
|
|
use OpenIsis; |
|
|
|
|
212 |
#$self->{'isis_code_page'} = $code_page; |
#$self->{'isis_code_page'} = $code_page; |
213 |
|
|
214 |
# create Text::Iconv object |
# create Text::Iconv object |
217 |
$log->info("reading ISIS database '",$arg->{'filename'},"'"); |
$log->info("reading ISIS database '",$arg->{'filename'},"'"); |
218 |
$log->debug("isis code page: $code_page"); |
$log->debug("isis code page: $code_page"); |
219 |
|
|
220 |
my $isis_db = OpenIsis::open($arg->{'filename'}); |
my ($isis_db,$maxmfn); |
221 |
|
|
222 |
|
if ($have_openisis) { |
223 |
|
$log->debug("using OpenIsis perl bindings"); |
224 |
|
$isis_db = OpenIsis::open($arg->{'filename'}); |
225 |
|
$maxmfn = OpenIsis::maxRowid( $isis_db ) || 1; |
226 |
|
} elsif ($have_biblio_isis) { |
227 |
|
$log->debug("using Biblio::Isis"); |
228 |
|
use Biblio::Isis; |
229 |
|
$isis_db = new Biblio::Isis( |
230 |
|
isisdb => $arg->{'filename'}, |
231 |
|
include_deleted => 1, |
232 |
|
hash_filter => sub { |
233 |
|
my $l = shift || return; |
234 |
|
$l = $cp->convert($l); |
235 |
|
return $l; |
236 |
|
}, |
237 |
|
); |
238 |
|
$maxmfn = $isis_db->{'maxmfn'}; |
239 |
|
} else { |
240 |
|
$log->logdie("Can't find supported ISIS library for perl. I suggent that you install Bilbio::Isis from CPAN."); |
241 |
|
} |
242 |
|
|
243 |
|
|
|
my $maxmfn = OpenIsis::maxRowid( $isis_db ) || 1; |
|
244 |
my $startmfn = 1; |
my $startmfn = 1; |
245 |
|
|
246 |
if (my $s = $self->{'start_mfn'}) { |
if (my $s = $self->{'start_mfn'}) { |
252 |
|
|
253 |
$maxmfn = $startmfn + $self->{limit_mfn} if ($self->{limit_mfn}); |
$maxmfn = $startmfn + $self->{limit_mfn} if ($self->{limit_mfn}); |
254 |
|
|
255 |
$log->info("processing ",($maxmfn-$startmfn)." records..."); |
$log->info("processing ",($maxmfn-$startmfn)." records using ",( $have_openisis ? 'OpenIsis' : 'Biblio::Isis')); |
256 |
|
|
257 |
|
|
258 |
# read database |
# read database |
259 |
for (my $mfn = $startmfn; $mfn <= $maxmfn; $mfn++) { |
for (my $mfn = $startmfn; $mfn <= $maxmfn; $mfn++) { |
263 |
|
|
264 |
my $rec; |
my $rec; |
265 |
|
|
266 |
# read record |
if ($have_openisis) { |
267 |
my $row = OpenIsis::read( $isis_db, $mfn ); |
|
268 |
foreach my $k (keys %{$row}) { |
# read record using OpenIsis |
269 |
if ($k ne "mfn") { |
my $row = OpenIsis::read( $isis_db, $mfn ); |
270 |
foreach my $l (@{$row->{$k}}) { |
foreach my $k (keys %{$row}) { |
271 |
$l = $cp->convert($l); |
if ($k ne "mfn") { |
272 |
# has subfields? |
foreach my $l (@{$row->{$k}}) { |
273 |
my $val; |
$l = $cp->convert($l); |
274 |
if ($l =~ m/\^/) { |
# has subfields? |
275 |
foreach my $t (split(/\^/,$l)) { |
my $val; |
276 |
next if (! $t); |
if ($l =~ m/\^/) { |
277 |
$val->{substr($t,0,1)} = substr($t,1); |
foreach my $t (split(/\^/,$l)) { |
278 |
|
next if (! $t); |
279 |
|
$val->{substr($t,0,1)} = substr($t,1); |
280 |
|
} |
281 |
|
} else { |
282 |
|
$val = $l; |
283 |
} |
} |
|
} else { |
|
|
$val = $l; |
|
|
} |
|
284 |
|
|
285 |
push @{$rec->{$k}}, $val; |
push @{$rec->{$k}}, $val; |
286 |
|
} |
287 |
|
} else { |
288 |
|
push @{$rec->{'000'}}, $mfn; |
289 |
} |
} |
|
} else { |
|
|
push @{$rec->{'000'}}, $mfn; |
|
290 |
} |
} |
291 |
|
|
292 |
|
} elsif ($have_biblio_isis) { |
293 |
|
$rec = $isis_db->to_hash($mfn); |
294 |
|
} else { |
295 |
|
$log->logdie("hum? implementation missing?"); |
296 |
} |
} |
297 |
|
|
298 |
$log->confess("record $mfn empty?") unless ($rec); |
$log->confess("record $mfn empty?") unless ($rec); |
397 |
|
|
398 |
$self->{'last_pcnt'} ||= 1; |
$self->{'last_pcnt'} ||= 1; |
399 |
|
|
400 |
my $p = int($curr * 100 / $max); |
my $p = int($curr * 100 / $max) || 1; |
401 |
|
|
402 |
# reset on re-run |
# reset on re-run |
403 |
if ($p < $self->{'last_pcnt'}) { |
if ($p < $self->{'last_pcnt'}) { |
404 |
$self->{'last_pcnt'} = $p; |
$self->{'last_pcnt'} = $p; |
405 |
$self->{'last_t'} = time(); |
$self->{'start_t'} = time(); |
|
$self->{'last_curr'} = 1; |
|
406 |
} |
} |
407 |
|
|
408 |
if ($p != $self->{'last_pcnt'}) { |
if ($p != $self->{'last_pcnt'}) { |
409 |
|
|
|
my $last_curr = $self->{'last_curr'} || $curr; |
|
410 |
my $t = time(); |
my $t = time(); |
411 |
my $rate = ($curr - $last_curr) / (($t - $self->{'last_t'} || 1)); |
my $rate = ($curr / ($t - $self->{'start_t'} || 1)); |
412 |
my $eta = ($max-$curr) / ($rate || 1); |
my $eta = ($max-$curr) / ($rate || 1); |
413 |
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$curr,"=" x ($p/3)."$p%>", $max, $rate, $self->fmt_time($eta)); |
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$curr,"=" x ($p/3)."$p%>", $max, $rate, $self->fmt_time($eta)); |
414 |
$self->{'last_pcnt'} = $p; |
$self->{'last_pcnt'} = $p; |
|
$self->{'last_t'} = time(); |
|
415 |
$self->{'last_curr'} = $curr; |
$self->{'last_curr'} = $curr; |
416 |
} |
} |
417 |
print STDERR "\n" if ($p == 100); |
print STDERR "\n" if ($p == 100); |
602 |
# iteration (for repeatable fields) |
# iteration (for repeatable fields) |
603 |
my $i = shift || 0; |
my $i = shift || 0; |
604 |
|
|
605 |
|
$log->logdie("infitite loop in format $format") if ($i > ($self->{'max_mfn'} || 9999)); |
606 |
|
|
607 |
# FIXME remove for speedup? |
# FIXME remove for speedup? |
608 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
609 |
|
|
617 |
# remove eval{...} from beginning |
# remove eval{...} from beginning |
618 |
$eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s); |
$eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s); |
619 |
|
|
620 |
|
my $filter_name; |
621 |
|
# remove filter{...} from beginning |
622 |
|
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
623 |
|
|
624 |
# do actual replacement of placeholders |
# do actual replacement of placeholders |
625 |
|
# repeatable fields |
626 |
$format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges; |
$format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges; |
627 |
|
# non-repeatable fields |
628 |
|
$format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found)/ges; |
629 |
|
|
630 |
if ($found) { |
if ($found) { |
631 |
$log->debug("format: $format"); |
$log->debug("format: $format"); |
633 |
my $eval = $self->fill_in($rec,$eval_code,$i); |
my $eval = $self->fill_in($rec,$eval_code,$i); |
634 |
return if (! $self->_eval($eval)); |
return if (! $self->_eval($eval)); |
635 |
} |
} |
636 |
|
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
637 |
|
$log->debug("filter '$filter_name' for $format"); |
638 |
|
$format = $self->{'filter'}->{$filter_name}->($format); |
639 |
|
return unless(defined($format)); |
640 |
|
$log->debug("filter result: $format"); |
641 |
|
} |
642 |
# do we have lookups? |
# do we have lookups? |
643 |
if ($format =~ /$LOOKUP_REGEX/o) { |
if ($format =~ /$LOOKUP_REGEX/o) { |
644 |
$log->debug("format '$format' has lookup"); |
$log->debug("format '$format' has lookup"); |
731 |
# remove eval{...} from beginning |
# remove eval{...} from beginning |
732 |
$eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s); |
$eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s); |
733 |
|
|
734 |
|
my $filter_name; |
735 |
|
# remove filter{...} from beginning |
736 |
|
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
737 |
|
|
738 |
my $prefix; |
my $prefix; |
739 |
my $all_found=0; |
my $all_found=0; |
740 |
|
|
741 |
while ($format =~ s/^(.*?)v(\d+)(?:\^(\w))?//s) { |
while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) { |
742 |
|
|
743 |
my $del = $1 || ''; |
my $del = $1 || ''; |
744 |
$prefix ||= $del if ($all_found == 0); |
$prefix ||= $del if ($all_found == 0); |
745 |
|
|
746 |
|
# repeatable index |
747 |
|
my $r = $i; |
748 |
|
$r = 0 if (lc("$2") eq 's'); |
749 |
|
|
750 |
my $found = 0; |
my $found = 0; |
751 |
my $tmp = $self->get_data(\$rec,$2,$3,$i,\$found); |
my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found); |
752 |
|
|
753 |
if ($found) { |
if ($found) { |
754 |
push @out, $del; |
push @out, $del; |
772 |
} |
} |
773 |
|
|
774 |
if ($eval_code) { |
if ($eval_code) { |
775 |
my $eval = $self->fill_in($rec,$eval_code,$i); |
my $eval = $self->fill_in($rec,$eval_code,$i) || return; |
776 |
$log->debug("about to eval{",$eval,"} format: $out"); |
$log->debug("about to eval{$eval} format: $out"); |
777 |
return if (! $self->_eval($eval)); |
return if (! $self->_eval($eval)); |
778 |
} |
} |
779 |
|
|
780 |
|
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
781 |
|
$log->debug("about to filter{$filter_name} format: $out"); |
782 |
|
$out = $self->{'filter'}->{$filter_name}->($out); |
783 |
|
return unless(defined($out)); |
784 |
|
$log->debug("filter result: $out"); |
785 |
|
} |
786 |
|
|
787 |
return $out; |
return $out; |
788 |
} |
} |
932 |
|
|
933 |
if ($tag->{'sort'}) { |
if ($tag->{'sort'}) { |
934 |
@v = $self->sort_arr(@v); |
@v = $self->sort_arr(@v); |
|
$log->warn("sort within tag is usually not what you want!"); |
|
935 |
} |
} |
936 |
|
|
937 |
# use format? |
# use format? |
967 |
# delimiter (or ,) |
# delimiter (or ,) |
968 |
my $d = $tag->{'delimiter'}; |
my $d = $tag->{'delimiter'}; |
969 |
# default delimiter |
# default delimiter |
970 |
$d ||= ", "; |
$d ||= " "; |
971 |
|
|
972 |
my $last = pop @{$row->{$type}}; |
my $last = pop @{$row->{$type}}; |
973 |
$d = "" if (! $last); |
$d = "" if (! $last); |
1139 |
|
|
1140 |
$log->debug("eval: ",$code," [",$ret,"]"); |
$log->debug("eval: ",$code," [",$ret,"]"); |
1141 |
|
|
1142 |
return $ret || 0; |
return $ret || undef; |
1143 |
} |
} |
1144 |
|
|
1145 |
=head2 _sort_by_order |
=head2 _sort_by_order |