1 |
#!/usr/bin/perl -w |
#!/usr/bin/perl -w |
2 |
|
|
3 |
use strict; |
use strict; |
4 |
use OpenIsis; |
use Biblio::Isis; |
5 |
use Getopt::Std; |
use Getopt::Std; |
6 |
use Data::Dumper; |
use Data::Dumper; |
7 |
use XML::Simple; |
use XML::Simple; |
|
use Text::Unaccent 1.02; # 1.01 won't compile on my platform, |
|
8 |
use Text::Iconv; |
use Text::Iconv; |
9 |
use Config::IniFiles; |
use Config::IniFiles; |
10 |
use Encode; |
use Encode; |
16 |
|
|
17 |
my $config_file = $0; |
my $config_file = $0; |
18 |
$config_file =~ s/\.pl$/.conf/; |
$config_file =~ s/\.pl$/.conf/; |
19 |
|
$config_file = $ARGV[0] if ($ARGV[0] && -f $ARGV[0]); |
20 |
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
21 |
|
|
22 |
my $config; |
my $config; |
23 |
|
|
24 |
#use index_DBI; # default DBI module for index |
#use index_DBI; # default DBI module for index |
25 |
use index_DBI_cache; # faster DBI module using memory cache |
#use index_DBI_cache; # faster DBI module using memory cache |
26 |
|
use index_DBI_tag; # tag support for indexes |
27 |
my $index; |
my $index; |
28 |
|
|
29 |
my %opts; |
my %opts; |
73 |
# if you are tight on memory, turn this off |
# if you are tight on memory, turn this off |
74 |
my $use_lhash_cache = 1; |
my $use_lhash_cache = 1; |
75 |
|
|
76 |
|
my $last_field_name; # cache to prevent repeated fields |
77 |
|
|
78 |
sub data2xml { |
sub data2xml { |
79 |
|
|
80 |
use xmlify; |
use xmlify; |
105 |
} |
} |
106 |
|
|
107 |
my @sorted_tags; |
my @sorted_tags; |
108 |
if ($cache->{tags_by_order}->{$type}) { |
if ($cache->{tags_by_order}) { |
109 |
@sorted_tags = @{$cache->{tags_by_order}->{$type}}; |
@sorted_tags = @{$cache->{tags_by_order}}; |
110 |
} else { |
} else { |
111 |
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
112 |
$cache->{tags_by_order}->{$type} = \@sorted_tags; |
$cache->{tags_by_order} = \@sorted_tags; |
113 |
} |
} |
114 |
|
|
115 |
# lookup key |
# lookup key |
116 |
my $lookup_key; |
my $lookup_key; |
117 |
|
|
118 |
# cache for field in pages |
# cache for field in pages |
119 |
$cache->{display_data} = (); |
delete $cache->{display_data}; |
120 |
$cache->{swish_data} = (); |
delete $cache->{swish_data}; |
121 |
$cache->{swish_exact_data} = (); |
delete $cache->{swish_exact_data}; |
122 |
|
delete $cache->{index_data}; |
123 |
|
delete $cache->{index_delimiter}; |
124 |
my @page_fields; # names of fields |
my @page_fields; # names of fields |
125 |
|
|
126 |
|
|
140 |
} else { |
} else { |
141 |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
142 |
} |
} |
143 |
|
|
144 |
if ($field_name) { |
if ($field_name) { |
145 |
return x($field_name); |
$field_name = x($field_name); |
146 |
|
if (! $last_field_name) { |
147 |
|
$last_field_name = $field_name; |
148 |
|
return $last_field_name; |
149 |
|
} elsif ($field_name ne $last_field_name) { |
150 |
|
$last_field_name = $field_name; |
151 |
|
return $last_field_name; |
152 |
|
} |
153 |
} |
} |
154 |
} |
} |
155 |
|
|
156 |
|
|
157 |
|
# init variables for different types |
158 |
|
sub init_visible_type($) { |
159 |
|
my $type = shift; |
160 |
|
|
161 |
|
# swish, swish_exact, display, index, index_lookup |
162 |
|
# swish and display defaults |
163 |
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
164 |
|
if (lc($type) eq "display") { |
165 |
|
$s = 0; |
166 |
|
} elsif (lc($type) eq "swish") { |
167 |
|
$d = 0; |
168 |
|
} elsif (lc($type) eq "index") { |
169 |
|
($s,$se,$d,$i) = (0,1,0,1); |
170 |
|
} elsif (lc($type) eq "swish_exact") { |
171 |
|
($s,$se,$d,$i) = (0,1,0,0); |
172 |
|
} elsif (lc($type) =~ /^lookup/) { |
173 |
|
($s,$se,$d,$i,$il) = (0,1,0,0,1); |
174 |
|
} elsif ($type) { |
175 |
|
print STDERR "WARNING: unknown type: $type\n"; |
176 |
|
} |
177 |
|
return ($s,$se,$d,$i,$il); |
178 |
|
} |
179 |
|
|
180 |
|
|
181 |
|
# convert |
182 |
|
# |
183 |
|
# <tag> |
184 |
|
# <delimiter>, </delimiter> |
185 |
|
# <value>200a</value> |
186 |
|
# </tag> |
187 |
|
# |
188 |
|
# to |
189 |
|
# |
190 |
|
# <tag delimiter=", ">200a</tag> |
191 |
|
# |
192 |
|
# but without loosing spaces in delimiter (becasue |
193 |
|
# new XML::Simple strips spaces in attribute values |
194 |
|
# as defined in XML specification) |
195 |
|
# |
196 |
|
sub unroll_x($) { |
197 |
|
my $x = shift; |
198 |
|
|
199 |
|
if (defined $x->{value}) { |
200 |
|
my ($v,$d) = ($x->{value}->{content}, $x->{delimiter}->{content}); |
201 |
|
delete $x->{value}; |
202 |
|
delete $x->{delimiter}; |
203 |
|
$x->{content} = $v; |
204 |
|
$x->{delimiter} = $d; |
205 |
|
} |
206 |
|
return $x; |
207 |
|
} |
208 |
|
|
209 |
# begin real work: go field by field |
# begin real work: go field by field |
210 |
foreach my $field (@sorted_tags) { |
foreach my $field (@sorted_tags) { |
211 |
|
|
215 |
my $swish_data = ""; |
my $swish_data = ""; |
216 |
my $swish_exact_data = ""; |
my $swish_exact_data = ""; |
217 |
my $display_data = ""; |
my $display_data = ""; |
218 |
|
my @index_data; |
219 |
my $line_delimiter; |
my $line_delimiter; |
220 |
|
|
221 |
my ($swish,$display); |
my ($swish,$display); |
228 |
my %page_max = (); |
my %page_max = (); |
229 |
# default line_delimiter if using |
# default line_delimiter if using |
230 |
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
231 |
|
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter}; |
232 |
|
|
233 |
|
my $format_name = $config->{indexer}->{$field}->{format_name}; |
234 |
|
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter}; |
235 |
|
if ($format_name && $format_delimiter) { |
236 |
|
$cache->{format}->{$field}->{format_name} = $format_name; |
237 |
|
$cache->{format}->{$field}->{format_delimiter} = $format_delimiter; |
238 |
|
} |
239 |
|
|
240 |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
241 |
|
|
242 |
|
$x = unroll_x($x); |
243 |
|
|
244 |
my $format = x($x->{content}); |
my $format = x($x->{content}); |
245 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
246 |
|
|
247 |
my $repeat_off = 0; # init repeatable offset |
my $repeat_off = 0; # init repeatable offset |
248 |
|
|
249 |
# swish, swish_exact, display, index, index_lookup |
my ($s,$se,$d,$i,$il) = init_visible_type($x->{type}); |
|
# swish and display defaults |
|
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
|
|
$s = 0 if (lc($x->{type}) eq "display"); |
|
|
$d = 0 if (lc($x->{type}) eq "swish"); |
|
|
$se = 1 if (lc($x->{type}) eq "swish_exact"); |
|
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
|
|
$il = 1 if (lc($x->{type}) =~ /^lookup/); |
|
|
|
|
250 |
|
|
251 |
# what will separate last line from this one? |
# what will separate last line from this one? |
252 |
if ($display_data && $x->{append} && $x->{append} eq "1") { |
if ($display_data && $x->{append}) { |
253 |
$line_delimiter = ' '; |
$line_delimiter = $delimiter; |
254 |
} elsif ($display_data) { |
} elsif ($display_data) { |
255 |
$line_delimiter = '<br/>'; |
$line_delimiter = '<br/>'; |
256 |
} |
} |
258 |
# init vars so that we go into while... |
# init vars so that we go into while... |
259 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
260 |
|
|
261 |
# placeholder for all repeatable entries for index |
sub mkformat($$) { |
|
my @index_data; |
|
|
|
|
|
sub mkformat { |
|
262 |
my $x = shift || die "mkformat needs tag reference"; |
my $x = shift || die "mkformat needs tag reference"; |
263 |
my $data = shift || return; |
my $data = shift || return; |
264 |
my $format_name = x($x->{format_name}) || return $data; |
my $format_name = x($x->{format_name}) || return $data; |
276 |
if (($#data+1) == $nr) { |
if (($#data+1) == $nr) { |
277 |
return sprintf($fmt,@data); |
return sprintf($fmt,@data); |
278 |
} else { |
} else { |
279 |
print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
#print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
280 |
return $data; |
return $data; |
281 |
} |
} |
282 |
} else { |
} else { |
305 |
$display = $new_display; |
$display = $new_display; |
306 |
$cache->{lhash}->{$display} = $new_display; |
$cache->{lhash}->{$display} = $new_display; |
307 |
} else { |
} else { |
308 |
print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
# print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
309 |
$display = ""; |
$display = ""; |
310 |
$cache->{lhash}->{$display} = $null; |
$cache->{lhash}->{$display} = $null; |
311 |
} |
} |
326 |
} |
} |
327 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
328 |
if ($swish) { |
if ($swish) { |
329 |
|
my $tmp = $swish; |
330 |
if ($filter && ($s || $se)) { |
if ($filter && ($s || $se)) { |
331 |
no strict 'refs'; |
no strict 'refs'; |
332 |
my $tmp = join(" ",&$filter($swish)) if ($s || $se); |
$tmp = join(" ",&$filter($tmp)) if ($s || $se); |
|
$swish_data .= $tmp if ($s); |
|
|
$swish_exact_data .= $tmp if ($se); |
|
|
|
|
|
} else { |
|
|
$swish_data .= $swish if ($s); |
|
|
$swish_exact_data .= $swish if ($se); |
|
333 |
} |
} |
334 |
|
|
335 |
|
$swish_data .= $tmp if ($s && $tmp); |
336 |
|
$swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se); |
337 |
} |
} |
338 |
|
|
339 |
# type="display" ; field for display |
# type="display" ; field for display |
340 |
if ($d && $display) { |
if ($d && $display) { |
341 |
|
my $ldel = $delimiter; |
342 |
if ($line_delimiter && $display_data) { |
if ($line_delimiter && $display_data) { |
343 |
$display_data .= $line_delimiter; |
$ldel = $line_delimiter; |
344 |
} |
} |
345 |
if ($filter) { |
if ($filter) { |
346 |
no strict 'refs'; |
no strict 'refs'; |
347 |
if ($display_data) { |
my @arr; |
348 |
$display_data .= $delimiter.join($delimiter,mkformat($x,&$filter($display))); |
foreach my $tmp (&$filter($display)) { |
349 |
} else { |
my $tmp2 = mkformat($x,$tmp); |
350 |
$display_data = join($delimiter,mkformat($x,&$filter($display))); |
push @arr,$tmp2 if ($tmp2); |
351 |
} |
} |
352 |
|
$display_data .= $ldel if ($display_data && @arr); |
353 |
|
$display_data .= join($delimiter,@arr); |
354 |
} else { |
} else { |
355 |
if ($display_data) { |
$display_data .= $ldel if ($display_data); |
356 |
$display_data .= $delimiter.mkformat($x,$display); |
my $tmp = mkformat($x,$display); |
357 |
} else { |
$display_data .= $tmp if ($tmp); |
|
$display_data = mkformat($x,$display); |
|
|
} |
|
358 |
} |
} |
359 |
} |
} |
360 |
|
|
361 |
# type="index" ; insert into index |
# type="index" ; insert into index |
362 |
|
my $idisplay; |
363 |
if ($i && $display) { |
if ($i && $display) { |
364 |
|
$idisplay = $display; |
365 |
if ($filter) { |
if ($filter) { |
366 |
no strict 'refs'; |
no strict 'refs'; |
367 |
$display = &$filter($display); |
$idisplay = &$filter($idisplay); |
|
} |
|
|
if ($x->{append} && @index_data) { |
|
|
$index_data[$#index_data].=$display; |
|
|
} else { |
|
|
push @index_data, $display; |
|
368 |
} |
} |
369 |
|
push @index_data, $idisplay if ($idisplay && !$iterate_by_page); |
370 |
} |
} |
371 |
|
|
372 |
# store fields in lookup |
# store fields in lookup |
375 |
if ($lookup_key) { |
if ($lookup_key) { |
376 |
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
377 |
} else { |
} else { |
378 |
$lookup_key = $display; |
if ($filter) { |
379 |
|
no strict 'refs'; |
380 |
|
$lookup_key = &$filter($display); |
381 |
|
} else { |
382 |
|
$lookup_key = $display; |
383 |
|
} |
384 |
} |
} |
385 |
} elsif (lc($x->{type}) eq "lookup_val") { |
} elsif (lc($x->{type}) eq "lookup_val") { |
386 |
if ($lookup_key) { |
if ($lookup_key) { |
387 |
$lhash{$lookup_key} = $display; |
if ($filter) { |
388 |
|
no strict 'refs'; |
389 |
|
$lhash{$lookup_key} = &$filter($display); |
390 |
|
} else { |
391 |
|
$lhash{$lookup_key} = $display; |
392 |
|
} |
393 |
} else { |
} else { |
394 |
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
395 |
} |
} |
402 |
sub iterate_fld($$$$$$) { |
sub iterate_fld($$$$$$) { |
403 |
my ($cache,$what,$field,$page,$data,$append) = @_; |
my ($cache,$what,$field,$page,$data,$append) = @_; |
404 |
return if (!$data); |
return if (!$data); |
405 |
my $line_delimiter = $page_line_delimiter; |
|
406 |
$line_delimiter = " " if ($append); |
my $ldel = $page_line_delimiter; |
407 |
|
$ldel = " " if ($append); |
408 |
|
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
409 |
if (! $cache->{$what}->{$field}->[$page]) { |
if (! $cache->{$what}->{$field}->[$page]) { |
410 |
$cache->{$what}->{$field}->[$page] = $data; |
$cache->{$what}->{$field}->[$page] = $data; |
411 |
} else { |
} else { |
412 |
$cache->{$what}->{$field}->[$page] .= $line_delimiter.$data; |
$cache->{$what}->{$field}->[$page] .= $ldel.$data; |
413 |
} |
} |
414 |
} |
} |
415 |
|
|
416 |
if ($display_data) { |
if ($display_data) { |
|
print STDERR "line delimiter: ",Dumper($line_delimiter) if ($line_delimiter); |
|
417 |
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
418 |
} |
} |
419 |
$display_data = ""; |
$display_data = ""; |
425 |
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
426 |
$swish_exact_data = ""; |
$swish_exact_data = ""; |
427 |
} |
} |
428 |
|
|
429 |
|
if ($idisplay) { |
430 |
|
my $ldel=$page_line_delimiter; |
431 |
|
my @index_data; |
432 |
|
if ($cache->{index_data}->{$field}->[$page]) { |
433 |
|
|
434 |
|
@index_data = @{$cache->{index_data}->{$field}->[$page]}; |
435 |
|
} |
436 |
|
if ($x->{append}) { |
437 |
|
if (@index_data) { |
438 |
|
$index_data[$#index_data] .= $idisplay; |
439 |
|
} else { |
440 |
|
push @index_data, $idisplay; |
441 |
|
} |
442 |
|
} else { |
443 |
|
push @index_data, $idisplay; |
444 |
|
} |
445 |
|
$idisplay = ""; |
446 |
|
@{$cache->{index_data}->{$field}->[$page]} = @index_data; |
447 |
|
} |
448 |
} |
} |
449 |
} |
} |
450 |
|
|
451 |
# fill data in index |
if (! $iterate_by_page) { |
452 |
foreach my $d (@index_data) { |
my $idel = $x->{index_delimiter}; |
453 |
$index->insert($field, $d, $path); |
# fill data in index |
454 |
|
foreach my $tmp (@index_data) { |
455 |
|
my $i = $d = $tmp; |
456 |
|
if ($idel && $tmp =~ m/$idel/) { |
457 |
|
($i,$d) = split(/$idel/,$tmp); |
458 |
|
} |
459 |
|
$index->insert($field, $i, $d, $path); |
460 |
|
} |
461 |
|
@index_data = (); |
462 |
} |
} |
463 |
} |
} |
464 |
|
|
465 |
# now try to parse variables from configuration file |
# now try to parse variables from configuration file |
466 |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
467 |
|
|
468 |
|
$x = unroll_x($x); |
469 |
|
|
470 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
471 |
my $val = $cfg->val($database, x($x->{content})); |
my $val = $cfg->val($database, x($x->{content})); |
472 |
|
|
473 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
# FIXME index_lookup is not supported! |
474 |
$s = 0 if (lc($x->{type}) eq "display"); |
my ($s,$se,$d,$i,$il) = init_visible_type($x->{type}); |
|
$d = 0 if (lc($x->{type}) eq "swish"); |
|
|
# no support for swish exact in config. |
|
|
# IMHO, it's useless |
|
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
|
475 |
|
|
476 |
if ($val) { |
if ($val) { |
477 |
$display_data .= $delimiter.$val if ($d); |
$display_data .= $delimiter.$val if ($d); |
478 |
$swish_data .= $val if ($s); |
$swish_data .= " ".$val if ($s); |
479 |
$index->insert($field, $val, $path) if ($i); |
$index->insert($field, $val, $path) if ($i); |
480 |
} |
} |
481 |
|
|
504 |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
505 |
#print STDERR Dumper($cache->{display_data}); |
#print STDERR Dumper($cache->{display_data}); |
506 |
for (my $page=0; $page <= $nr_pages; $page++) { |
for (my $page=0; $page <= $nr_pages; $page++) { |
507 |
|
my $display_data; |
508 |
my $display_data = $cache->{display_data}->{$field}->[$page]; |
if ($cache->{format}->{$field}) { |
509 |
|
my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]); |
510 |
|
$display_data=$tmp if ($tmp); |
511 |
|
} else { |
512 |
|
$display_data = $cache->{display_data}->{$field}->[$page]; |
513 |
|
} |
514 |
if ($display_data) { # default |
if ($display_data) { # default |
515 |
if ($field eq "headline") { |
if ($field eq "headline") { |
516 |
$xml .= xmlify("headline", $display_data); |
$xml .= xmlify("headline", $display_data); |
528 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
529 |
$swish_data =~ s/ +$//g; |
$swish_data =~ s/ +$//g; |
530 |
|
|
531 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
532 |
} |
} |
533 |
|
|
534 |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
538 |
|
|
539 |
# add delimiters before and after word. |
# add delimiters before and after word. |
540 |
# That is required to produce exact match |
# That is required to produce exact match |
541 |
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
542 |
|
} |
543 |
|
|
544 |
|
my $idel = $cache->{index_delimiter}->{$field}; |
545 |
|
foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) { |
546 |
|
my $i = $tmp; |
547 |
|
my $d = $tmp; |
548 |
|
if ($idel && $tmp =~ m/$idel/) { |
549 |
|
($i,$d) = split(/$idel/,$tmp); |
550 |
|
} |
551 |
|
$index->insert($field, $i, $d, $path); |
552 |
|
#print STDERR "index [$idel] $field: $i --> $d [$path]\n"; |
553 |
} |
} |
554 |
} |
} |
555 |
|
|
571 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
572 |
$swish_data =~ s/ +$//g; |
$swish_data =~ s/ +$//g; |
573 |
|
|
574 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
575 |
} |
} |
576 |
|
|
577 |
if ($swish_exact_data) { |
if ($swish_exact_data) { |
580 |
|
|
581 |
# add delimiters before and after word. |
# add delimiters before and after word. |
582 |
# That is required to produce exact match |
# That is required to produce exact match |
583 |
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
584 |
} |
} |
585 |
} |
} |
586 |
} |
} |
617 |
|
|
618 |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
619 |
|
|
620 |
my $unac_filter = $cfg_global->val('global', 'unac_filter'); |
my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter'); |
621 |
if ($unac_filter) { |
if ($my_unac_filter) { |
622 |
require $unac_filter; |
print STDERR "using $my_unac_filter to filter characters for search\n"; |
623 |
|
require $my_unac_filter; |
624 |
|
} else { |
625 |
|
print STDERR "### fallback to default my_unac_string!\n"; |
626 |
|
eval q{ |
627 |
|
sub main::my_unac_string($$) { |
628 |
|
my ($charset, $string) = (@_); |
629 |
|
return $string; |
630 |
|
} |
631 |
|
}; |
632 |
} |
} |
633 |
|
|
634 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
640 |
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
641 |
if ($lookup_file) { |
if ($lookup_file) { |
642 |
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
643 |
|
if (! -e $lookup_file) { |
644 |
|
open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!"; |
645 |
|
close(LOOKUP); |
646 |
|
} |
647 |
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
648 |
print STDERR "creating lookup file '$lookup_file'\n"; |
print STDERR "creating lookup file '$lookup_file'\n"; |
649 |
|
# delete memory cache for lookup file |
650 |
|
delete $cache->{lhash}; |
651 |
} |
} |
652 |
|
|
653 |
# open existing lookup file |
# open existing lookup file |
664 |
my $type_base = $type; |
my $type_base = $type; |
665 |
$type_base =~ s/_.+$//g; |
$type_base =~ s/_.+$//g; |
666 |
|
|
667 |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1); |
$config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 ); |
668 |
|
|
669 |
|
# helper for progress bar |
670 |
|
sub fmt_time { |
671 |
|
my $t = shift || 0; |
672 |
|
my $out = ""; |
673 |
|
|
674 |
|
my ($ss,$mm,$hh) = gmtime($t); |
675 |
|
$out .= "${hh}h" if ($hh); |
676 |
|
$out .= sprintf("%02d:%02d", $mm,$ss); |
677 |
|
$out .= " " if ($hh == 0); |
678 |
|
return $out; |
679 |
|
} |
680 |
|
|
681 |
# output current progress indicator |
# output current progress indicator |
682 |
my $last_p = 0; |
my $last_p = 0; |
683 |
|
my $start_t = time(); |
684 |
sub progress { |
sub progress { |
685 |
return if (! $show_progress); |
return if (! $show_progress); |
686 |
my $current = shift; |
my $current = shift; |
687 |
my $total = shift || 1; |
my $total = shift || 1; |
688 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
689 |
if ($p != $last_p) { |
if ($p < $last_p || $current == 1) { |
690 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
$start_t = time(); |
691 |
|
$last_p = 0; |
692 |
|
} elsif ($p != $last_p) { |
693 |
|
my $rate = ($current / (time() - $start_t || 1)); |
694 |
|
my $eta = ($total-$current) / ($rate || 1); |
695 |
|
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta)); |
696 |
$last_p = $p; |
$last_p = $p; |
697 |
} |
} |
698 |
} |
} |
699 |
|
|
700 |
my $fake_dir = 1; |
my $fake_dir = 1; |
701 |
|
my $fake_pos = 0; |
702 |
|
my $last_fake_t = time(); |
703 |
sub fakeprogress { |
sub fakeprogress { |
704 |
return if (! $show_progress); |
return if (! $show_progress); |
705 |
my $current = shift @_; |
my $current = shift @_; |
706 |
|
|
707 |
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
my @ind = ('-','\\','|','/','-','\\','|','/'); |
708 |
|
|
709 |
|
if ($current < $fake_pos) { |
710 |
|
$start_t = time(); |
711 |
|
$last_fake_t = 0; |
712 |
|
$fake_dir = 1; |
713 |
|
$fake_pos = 0; |
714 |
|
} |
715 |
|
|
716 |
$last_p += $fake_dir; |
if (time()-$last_fake_t >= 1) { |
717 |
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
$last_fake_t = time(); |
718 |
if ($last_p % 10 == 0) { |
$fake_pos += $fake_dir; |
719 |
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
$fake_dir = -$fake_dir if ($fake_pos > 38); |
720 |
|
} |
721 |
|
|
722 |
|
if ($current % 10 == 0) { |
723 |
|
my $rate = ($current / (time() - $start_t || 1)); |
724 |
|
printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate); |
725 |
} |
} |
726 |
} |
} |
727 |
|
|
728 |
# now read database |
# now read database |
729 |
print STDERR "using: $type...\n"; |
print STDERR "using: $type...\n"; |
730 |
|
|
731 |
|
# erase cache for tags by order in this database |
732 |
|
delete $cache->{tags_by_order}; |
733 |
|
|
734 |
if ($type_base eq "isis") { |
if ($type_base eq "isis") { |
735 |
|
|
736 |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
737 |
|
|
738 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
739 |
my $db = OpenIsis::open( $isis_db ); |
my $db = new Biblio::Isis( isisdb => $isis_db ); |
740 |
|
|
741 |
# check if .txt database for OpenIsis is zero length, |
my $max_rowid = $db->count || die "can't find maxmfn"; |
|
# if so, erase it and re-open database |
|
|
sub check_txt_db { |
|
|
my $isis_db = shift || die "need isis database name"; |
|
|
my $reopen = 0; |
|
|
|
|
|
if (-e $isis_db.".TXT") { |
|
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
|
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
|
|
$reopen++; |
|
|
} |
|
|
if (-e $isis_db.".PTR") { |
|
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
|
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
|
|
$reopen++; |
|
|
} |
|
|
return OpenIsis::open( $isis_db ) if ($reopen); |
|
|
} |
|
|
|
|
|
# EOF error |
|
|
if ($db == -1) { |
|
|
$db = check_txt_db($isis_db); |
|
|
if ($db == -1) { |
|
|
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
|
|
next; |
|
|
} |
|
|
} |
|
|
|
|
|
# OpenIsis::ERR_BADF |
|
|
if ($db == -4) { |
|
|
print STDERR "FATAL: OpenIsis can't find file $isis_db\n"; |
|
|
next; |
|
|
# OpenIsis::ERR_IO |
|
|
} elsif ($db == -5) { |
|
|
print STDERR "FATAL: OpenIsis can't access file $isis_db\n"; |
|
|
next; |
|
|
} elsif ($db < 0) { |
|
|
print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n"; |
|
|
next; |
|
|
} |
|
|
|
|
|
my $max_rowid = OpenIsis::maxRowid( $db ); |
|
|
|
|
|
# if 0 records, try to rease isis .txt database |
|
|
if ($max_rowid == 0) { |
|
|
# force removal of database |
|
|
$db = check_txt_db($isis_db); |
|
|
$max_rowid = OpenIsis::maxRowid( $db ); |
|
|
} |
|
742 |
|
|
743 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
744 |
|
|
745 |
my $path = $database; |
$path = $database; |
746 |
|
|
747 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
748 |
my $row = OpenIsis::read( $db, $row_id ); |
my $row = $db->to_hash( $row_id ); |
749 |
if ($row && $row->{mfn}) { |
if ($row) { |
750 |
|
|
751 |
|
$row->{mfn} = $row_id; |
752 |
|
$row->{record} = $db->{record}; |
753 |
|
|
754 |
progress($row->{mfn}, $max_rowid); |
progress($row->{mfn}, $max_rowid); |
755 |
|
|
756 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
764 |
} |
} |
765 |
} |
} |
766 |
} |
} |
|
# for this to work with current version of OpenIsis (0.9.0) |
|
|
# you might need my patch from |
|
|
# https://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
|
|
OpenIsis::close($db); |
|
767 |
print STDERR "\n"; |
print STDERR "\n"; |
768 |
|
|
769 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
770 |
use Spreadsheet::ParseExcel; |
require Spreadsheet::ParseExcel; |
771 |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
require Spreadsheet::ParseExcel::Utility; |
772 |
|
import Spreadsheet::ParseExcel::Utility qw(int2col); |
773 |
|
|
774 |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
775 |
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!"; |
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!"; |
797 |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
798 |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
799 |
if ($cell) { |
if ($cell) { |
800 |
$row->{int2col($iC)} = $cell->Value; |
# this conversion is a cludge. |
801 |
|
# Files from Excell could have |
802 |
|
# characters which don't fit into |
803 |
|
# destination encoding. |
804 |
|
$row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value; |
805 |
} |
} |
806 |
} |
} |
807 |
|
|
827 |
} |
} |
828 |
} elsif ($type_base eq "marc") { |
} elsif ($type_base eq "marc") { |
829 |
|
|
830 |
use MARC; |
require MARC::File::USMARC; |
831 |
|
|
832 |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
833 |
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
834 |
|
|
835 |
# optional argument is format |
# optional argument is format |
836 |
my $format = x($config->{format}) || 'usmarc'; |
warn "marc_format is no longer used!" if ($config->{marc_format}); |
|
|
|
837 |
print STDERR "Reading MARC file '$marc_file'\n"; |
print STDERR "Reading MARC file '$marc_file'\n"; |
838 |
|
|
839 |
my $marc = new MARC; |
my $marc = MARC::File::USMARC->in( $marc_file ) |
840 |
my $nr = $marc->openmarc({ |
|| die "Can't open MARC file '$marc_file': ".$MARC::File::ERROR; |
|
file=>$marc_file, format=>$format |
|
|
}) || die "Can't open MARC file '$marc_file'"; |
|
841 |
|
|
842 |
my $i=0; # record nr. |
# count records in MARC file |
843 |
|
sub marc_count { |
844 |
|
my $filename = shift || die; |
845 |
|
my $file = MARC::File::USMARC->in($filename) || die $MARC::File::ERROR; |
846 |
|
my $count = 0; |
847 |
|
while ($file->skip()) { |
848 |
|
$count++; |
849 |
|
} |
850 |
|
return $count; |
851 |
|
} |
852 |
|
|
853 |
|
my $count = marc_count($marc_file) || warn "no records in '$marc_file'?"; |
854 |
|
|
855 |
my $rec; |
my $i = 0; |
856 |
|
|
857 |
while ($marc->nextmarc(1)) { |
while( my $rec = $marc->next() ) { |
858 |
|
|
859 |
# XXX |
progress($i++,$count); |
|
fakeprogress($i++); |
|
860 |
|
|
861 |
my $swishpath = $database."#".$i; |
my $swishpath = $database."#".$i; |
862 |
|
|
863 |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) { |
864 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
865 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
866 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
868 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
869 |
} |
} |
870 |
} |
} |
871 |
|
|
872 |
|
print STDERR "\n"; |
873 |
|
|
874 |
} elsif ($type_base eq "feed") { |
} elsif ($type_base eq "feed") { |
875 |
|
|
876 |
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
925 |
|
|
926 |
all2xml.pl - read various file formats and dump XML for SWISH-E |
all2xml.pl - read various file formats and dump XML for SWISH-E |
927 |
|
|
928 |
|
=head1 SYNOPSYS |
929 |
|
|
930 |
|
$ all2xml.pl [test.conf] |
931 |
|
|
932 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
933 |
|
|
934 |
This command will read ISIS data file using OpenIsis perl module, MARC |
This command will read ISIS data file using IsisDB perl module, MARC |
935 |
records using MARC module and optionally Micro$oft Excel files to |
records using MARC module and optionally Micro$oft Excel files to |
936 |
create one XML file for usage with I<SWISH-E> indexer. Dispite it's name, |
create one XML file for usage with I<SWISH-E> indexer. Dispite it's name, |
937 |
this script B<isn't general xml generator> from isis files (isis allready |
this script B<isn't general xml generator> from isis files (isis allready |
938 |
has something like that). Output of this script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |
939 |
|
|
940 |
|
If no configuration file is specified, it will use default one called |
941 |
|
C<all2xml.conf>. |
942 |
|
|
943 |
=head1 BUGS |
=head1 BUGS |
944 |
|
|
945 |
Documentation is really lacking. However, in true Open Source spirit, source |
Documentation is really lacking. However, in true Open Source spirit, source |