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; |
11 |
|
#use GDBM_File; |
12 |
|
use Fcntl; # for O_RDWR |
13 |
|
use TDB_File; |
14 |
|
|
15 |
$|=1; |
$|=1; |
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_filter; # filter support for indexes |
27 |
my $index; |
my $index; |
28 |
|
|
29 |
my %opts; |
my %opts; |
65 |
'feed' => 'feed' |
'feed' => 'feed' |
66 |
); |
); |
67 |
|
|
68 |
|
my $cache; # for cacheing |
69 |
|
|
70 |
|
# lookup hash (tied to file) |
71 |
|
my %lhash; |
72 |
|
# this option will cache all lookup entries in memory. |
73 |
|
# if you are tight on memory, turn this off |
74 |
|
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; |
104 |
return $va <=> $vb; |
return $va <=> $vb; |
105 |
} |
} |
106 |
|
|
107 |
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
my @sorted_tags; |
108 |
|
if ($cache->{tags_by_order}) { |
109 |
|
@sorted_tags = @{$cache->{tags_by_order}}; |
110 |
|
} else { |
111 |
|
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
112 |
|
$cache->{tags_by_order} = \@sorted_tags; |
113 |
|
} |
114 |
|
|
115 |
|
# lookup key |
116 |
|
my $lookup_key; |
117 |
|
|
118 |
|
# cache for field in pages |
119 |
|
delete $cache->{display_data}; |
120 |
|
delete $cache->{swish_data}; |
121 |
|
delete $cache->{swish_exact_data}; |
122 |
|
delete $cache->{index_data}; |
123 |
|
delete $cache->{index_delimiter}; |
124 |
|
my @page_fields; # names of fields |
125 |
|
|
126 |
|
|
127 |
|
# subs used to produce output |
128 |
|
|
129 |
|
sub get_field_name($$$) { |
130 |
|
my ($config,$field,$field_usage) = @_; |
131 |
|
|
132 |
|
# find field name (signular, plural) |
133 |
|
my $field_name = ""; |
134 |
|
if ($config->{indexer}->{$field}->{name_singular} && $field_usage == 1) { |
135 |
|
$field_name = $config->{indexer}->{$field}->{name_singular}; |
136 |
|
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
137 |
|
$field_name = $config->{indexer}->{$field}->{name_plural}; |
138 |
|
} elsif ($config->{indexer}->{$field}->{name}) { |
139 |
|
$field_name = $config->{indexer}->{$field}->{name}; |
140 |
|
} else { |
141 |
|
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
142 |
|
} |
143 |
|
|
144 |
|
if ($field_name) { |
145 |
|
$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 |
210 |
|
foreach my $field (@sorted_tags) { |
211 |
|
|
212 |
$field=x($field); |
$field=x($field); |
213 |
$field_usage{$field}++; |
$field_usage{$field}++; |
214 |
|
|
215 |
my $swish_data = ""; |
my $swish_data = ""; |
216 |
|
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); |
222 |
|
|
223 |
my $tag = $type2tag{$type} || die "can't find which tag to use for type $type"; |
my $tag = $type2tag{$type} || die "can't find which tag to use for type $type"; |
224 |
|
|
225 |
|
# is this field page-by-page? |
226 |
|
my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page}; |
227 |
|
push @page_fields,$field if ($iterate_by_page); |
228 |
|
my %page_max = (); |
229 |
|
# default line_delimiter if using |
230 |
|
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; # repeatable offset |
my $repeat_off = 0; # init repeatable offset |
248 |
|
|
249 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
my ($s,$se,$d,$i,$il) = init_visible_type($x->{type}); |
|
$s = 0 if (lc($x->{type}) eq "display"); |
|
|
$d = 0 if (lc($x->{type}) eq "swish"); |
|
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
|
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($$) { |
262 |
my @index_data; |
my $x = shift || die "mkformat needs tag reference"; |
263 |
my $index_filter; |
my $data = shift || return; |
264 |
|
my $format_name = x($x->{format_name}) || return $data; |
265 |
|
my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!"; |
266 |
|
my $format_delimiter = x($x->{format_delimiter}); |
267 |
|
my @data; |
268 |
|
if ($format_delimiter) { |
269 |
|
@data = split(/$format_delimiter/,$data); |
270 |
|
} else { |
271 |
|
push @data,$data; |
272 |
|
} |
273 |
|
|
274 |
|
if ($fmt) { |
275 |
|
my $nr = scalar $fmt =~ s/%s/%s/g; |
276 |
|
if (($#data+1) == $nr) { |
277 |
|
return sprintf($fmt,@data); |
278 |
|
} else { |
279 |
|
#print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
280 |
|
return $data; |
281 |
|
} |
282 |
|
} else { |
283 |
|
print STDERR "usage of link '$format_name' without defined format (<link> tag)\n"; |
284 |
|
} |
285 |
|
} |
286 |
|
|
287 |
# while because of repeatable fields |
# while because of repeatable fields |
288 |
while ($swish || $display) { |
while ($swish || $display) { |
289 |
|
my $page = $repeat_off; |
290 |
|
$page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0)); |
291 |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
292 |
if ($repeat_off > 1000) { |
if ($repeat_off > 1000) { |
293 |
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
294 |
last; |
last; |
295 |
} |
} |
296 |
|
|
297 |
|
# is this field is lookup? |
298 |
|
if ($display && $x->{lookup}) { |
299 |
|
my $null = "<!-- null -->"; |
300 |
|
if ($use_lhash_cache) { |
301 |
|
if (!defined($cache->{lhash}->{$display})) { |
302 |
|
my $new_display = $lhash{$display}; |
303 |
|
if (defined($new_display)) { |
304 |
|
#print STDERR "lookup cache store '$display' = '$new_display'\n"; |
305 |
|
$display = $new_display; |
306 |
|
$cache->{lhash}->{$display} = $new_display; |
307 |
|
} else { |
308 |
|
# print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
309 |
|
$display = ""; |
310 |
|
$cache->{lhash}->{$display} = $null; |
311 |
|
} |
312 |
|
} else { |
313 |
|
$display = $cache->{lhash}->{$display}; |
314 |
|
} |
315 |
|
} else { |
316 |
|
$display = $lhash{$display} || $null; |
317 |
|
} |
318 |
|
} |
319 |
|
|
320 |
# filter="name" ; filter this field through |
# filter="name" ; filter this field through |
321 |
# filter/[name].pm |
# filter/[name].pm |
322 |
my $filter = $x->{filter}; |
my $filter = $x->{filter}; |
323 |
if ($filter) { |
if ($filter && !$cache->{filter_loaded}->{$filter}) { |
324 |
require "filter/".$filter.".pm"; |
require "filter/".$filter.".pm"; |
325 |
|
$cache->{filter_loaded}->{$filter}++; |
326 |
} |
} |
327 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
328 |
if ($s && $swish) { |
if ($swish) { |
329 |
if ($filter) { |
my $tmp = $swish; |
330 |
|
if ($filter && ($s || $se)) { |
331 |
no strict 'refs'; |
no strict 'refs'; |
332 |
$swish_data .= join(" ",&$filter($swish)); |
$tmp = join(" ",&$filter($tmp)) if ($s || $se); |
|
} else { |
|
|
$swish_data .= $swish; |
|
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; |
|
undef $line_delimiter; |
|
344 |
} |
} |
345 |
if ($filter) { |
if ($filter) { |
346 |
no strict 'refs'; |
no strict 'refs'; |
347 |
$display_data .= join($delimiter,&$filter($display)); |
my @arr; |
348 |
} else { |
foreach my $tmp (&$filter($display)) { |
349 |
if ($display_data) { |
my $tmp2 = mkformat($x,$tmp); |
350 |
$display_data .= $delimiter.$display; |
push @arr,$tmp2 if ($tmp2); |
|
} else { |
|
|
$display_data .= $display; |
|
351 |
} |
} |
352 |
|
$display_data .= $ldel if ($display_data && @arr); |
353 |
|
$display_data .= join($delimiter,@arr); |
354 |
|
} else { |
355 |
|
$display_data .= $ldel if ($display_data); |
356 |
|
my $tmp = mkformat($x,$display); |
357 |
|
$display_data .= $tmp if ($tmp); |
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 |
push @index_data, $display; |
$idisplay = $display; |
365 |
$index_filter = $filter if ($filter); |
if ($filter) { |
366 |
|
no strict 'refs'; |
367 |
|
$idisplay = &$filter($idisplay); |
368 |
|
} |
369 |
|
push @index_data, $idisplay if ($idisplay && !$iterate_by_page); |
370 |
} |
} |
|
} |
|
371 |
|
|
372 |
# fill data in index |
# store fields in lookup |
373 |
if (@index_data) { |
if ($il && $display) { |
374 |
if ($index_filter) { |
if (lc($x->{type}) eq "lookup_key") { |
375 |
no strict 'refs'; |
if ($lookup_key) { |
376 |
foreach my $d (&$index_filter(@index_data)) { |
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
377 |
$index->insert($field, $d, $path); |
} else { |
378 |
|
if ($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") { |
386 |
|
if ($lookup_key) { |
387 |
|
if ($filter) { |
388 |
|
no strict 'refs'; |
389 |
|
$lhash{$lookup_key} = &$filter($display); |
390 |
|
} else { |
391 |
|
$lhash{$lookup_key} = $display; |
392 |
|
} |
393 |
|
} else { |
394 |
|
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
395 |
|
} |
396 |
} |
} |
397 |
} else { |
|
398 |
foreach my $d (@index_data) { |
} |
399 |
$index->insert($field, $d, $path); |
|
400 |
|
# store data for page-by-page repeatable fields |
401 |
|
if ($iterate_by_page) { |
402 |
|
sub iterate_fld($$$$$$) { |
403 |
|
my ($cache,$what,$field,$page,$data,$append) = @_; |
404 |
|
return if (!$data); |
405 |
|
|
406 |
|
my $ldel = $page_line_delimiter; |
407 |
|
$ldel = " " if ($append); |
408 |
|
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
409 |
|
if (! $cache->{$what}->{$field}->[$page]) { |
410 |
|
$cache->{$what}->{$field}->[$page] = $data; |
411 |
|
} else { |
412 |
|
$cache->{$what}->{$field}->[$page] .= $ldel.$data; |
413 |
|
} |
414 |
|
} |
415 |
|
|
416 |
|
if ($display_data) { |
417 |
|
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
418 |
|
} |
419 |
|
$display_data = ""; |
420 |
|
if ($swish_data) { |
421 |
|
iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append}); |
422 |
|
$swish_data = ""; |
423 |
} |
} |
424 |
|
if ($swish_exact_data) { |
425 |
|
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
426 |
|
$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 |
|
if (! $iterate_by_page) { |
452 |
|
my $idel = $x->{index_delimiter}; |
453 |
|
# 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"); |
|
|
($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 |
|
|
482 |
|
if ($iterate_by_page) { |
483 |
|
# FIXME data from config tag will appear just |
484 |
|
# on first page!!! |
485 |
|
my $page = 0; |
486 |
|
if ($display_data) { |
487 |
|
$cache->{display_data}->{$field}->[$page] = $display_data; |
488 |
|
$display_data = ""; |
489 |
|
} |
490 |
|
if ($swish_data) { |
491 |
|
$cache->{swish_data}->{$field}->[$page] = $swish_data; |
492 |
|
$swish_data = ""; |
493 |
|
} |
494 |
|
if ($swish_exact_data) { |
495 |
|
$cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data; |
496 |
|
$swish_exact_data = ""; |
497 |
|
} |
498 |
|
} |
499 |
} |
} |
500 |
|
|
501 |
|
# save data page-by-page |
502 |
if ($display_data) { |
foreach my $field (@page_fields) { |
503 |
|
my $nr_pages = $page_max{$field} || next; |
504 |
if ($field eq "headline") { |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
505 |
$xml .= xmlify("headline", $display_data); |
#print STDERR Dumper($cache->{display_data}); |
506 |
} else { |
for (my $page=0; $page <= $nr_pages; $page++) { |
507 |
|
my $display_data; |
508 |
# find field name (signular, plural) |
if ($cache->{format}->{$field}) { |
509 |
my $field_name = ""; |
my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]); |
510 |
if ($config->{indexer}->{$field}->{name_singular} && $field_usage{$field} == 1) { |
$display_data=$tmp if ($tmp); |
|
$field_name = $config->{indexer}->{$field}->{name_singular}."#-#"; |
|
|
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
|
|
$field_name = $config->{indexer}->{$field}->{name_plural}."#-#"; |
|
|
} elsif ($config->{indexer}->{$field}->{name}) { |
|
|
$field_name = $config->{indexer}->{$field}->{name}."#-#"; |
|
511 |
} else { |
} else { |
512 |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
$display_data = $cache->{display_data}->{$field}->[$page]; |
513 |
} |
} |
514 |
if ($field_name) { |
if ($display_data) { # default |
515 |
$html .= x($field_name); |
if ($field eq "headline") { |
516 |
|
$xml .= xmlify("headline", $display_data); |
517 |
|
} else { |
518 |
|
|
519 |
|
# fallback to empty field name if needed |
520 |
|
$html .= get_field_name($config,$field,$field_usage{$field}) || ''; |
521 |
|
$html .= "#-#".$display_data."###\n"; |
522 |
|
} |
523 |
|
} |
524 |
|
|
525 |
|
my $swish_data = $cache->{swish_data}->{$field}->[$page]; |
526 |
|
if ($swish_data) { |
527 |
|
# remove extra spaces |
528 |
|
$swish_data =~ s/ +/ /g; |
529 |
|
$swish_data =~ s/ +$//g; |
530 |
|
|
531 |
|
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
532 |
|
} |
533 |
|
|
534 |
|
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
535 |
|
if ($swish_exact_data) { |
536 |
|
$swish_exact_data =~ s/ +/ /g; |
537 |
|
$swish_exact_data =~ s/ +$//g; |
538 |
|
|
539 |
|
# add delimiters before and after word. |
540 |
|
# That is required to produce exact match |
541 |
|
$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 |
} |
} |
|
$html .= $display_data."###\n"; |
|
554 |
} |
} |
555 |
|
|
556 |
} |
} |
557 |
if ($swish_data) { |
|
558 |
# remove extra spaces |
if (! $iterate_by_page) { |
559 |
$swish_data =~ s/ +/ /g; |
if ($display_data) { |
560 |
$swish_data =~ s/ +$//g; |
if ($field eq "headline") { |
561 |
|
$xml .= xmlify("headline", $display_data); |
562 |
|
} else { |
563 |
|
|
564 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
# fallback to empty field name if needed |
565 |
} |
$html .= get_field_name($config,$field,$field_usage{$field}) || ''; |
566 |
|
$html .= "#-#".$display_data."###\n"; |
567 |
|
} |
568 |
|
} |
569 |
|
if ($swish_data) { |
570 |
|
# remove extra spaces |
571 |
|
$swish_data =~ s/ +/ /g; |
572 |
|
$swish_data =~ s/ +$//g; |
573 |
|
|
574 |
|
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
575 |
|
} |
576 |
|
|
577 |
|
if ($swish_exact_data) { |
578 |
|
$swish_exact_data =~ s/ +/ /g; |
579 |
|
$swish_exact_data =~ s/ +$//g; |
580 |
|
|
581 |
|
# add delimiters before and after word. |
582 |
|
# That is required to produce exact match |
583 |
|
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
584 |
|
} |
585 |
|
} |
586 |
} |
} |
587 |
|
|
588 |
# dump formatted output in <html> |
# dump formatted output in <html> |
617 |
|
|
618 |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
619 |
|
|
620 |
|
my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter'); |
621 |
|
if ($my_unac_filter) { |
622 |
|
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) { |
635 |
|
|
636 |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
637 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
638 |
|
|
639 |
|
# create new lookup file |
640 |
|
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
641 |
|
if ($lookup_file) { |
642 |
|
#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', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
648 |
|
print STDERR "creating lookup file '$lookup_file'\n"; |
649 |
|
# delete memory cache for lookup file |
650 |
|
delete $cache->{lhash}; |
651 |
|
} |
652 |
|
|
653 |
|
# open existing lookup file |
654 |
|
$lookup_file = $cfg -> val($database, 'lookup_open'); # optional |
655 |
|
if ($lookup_file) { |
656 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644; |
657 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644; |
658 |
|
print STDERR "opening lookup file '$lookup_file'\n"; |
659 |
|
} |
660 |
|
|
661 |
print STDERR "reading ./import_xml/$type.xml\n"; |
print STDERR "reading ./import_xml/$type.xml\n"; |
662 |
|
|
663 |
# extract just type basic |
# extract just type basic |
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' ], 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 |
|
$fake_pos = 0; |
714 |
|
} |
715 |
|
|
716 |
|
if (time()-$last_fake_t >= 1) { |
717 |
|
$last_fake_t = time(); |
718 |
|
$fake_pos += $fake_dir; |
719 |
|
$fake_dir = -$fake_dir if ($fake_pos > 38); |
720 |
|
} |
721 |
|
|
722 |
$last_p += $fake_dir; |
if ($current % 10 == 0) { |
723 |
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
my $rate = ($current / (time() - $start_t || 1)); |
724 |
if ($last_p % 10 == 0) { |
printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate); |
|
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
|
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 |
# OpenIsis::ERR_BADF |
my $max_rowid = $db->count || die "can't find maxmfn"; |
742 |
if ($db == -4) { |
$max_rowid = 200; # XXX |
|
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 ); |
|
743 |
|
|
744 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
745 |
|
|
746 |
my $path = $database; |
$path = $database; |
747 |
|
|
748 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
749 |
my $row = OpenIsis::read( $db, $row_id ); |
my $row = $db->to_hash( $row_id ); |
750 |
if ($row && $row->{mfn}) { |
if ($row) { |
751 |
|
|
752 |
|
$row->{mfn} = $row_id; |
753 |
|
$row->{record} = $db->{record}; |
754 |
|
|
755 |
progress($row->{mfn}, $max_rowid); |
progress($row->{mfn}, $max_rowid); |
756 |
|
|
757 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
768 |
print STDERR "\n"; |
print STDERR "\n"; |
769 |
|
|
770 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
771 |
use Spreadsheet::ParseExcel; |
require Spreadsheet::ParseExcel; |
772 |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
require Spreadsheet::ParseExcel::Utility; |
773 |
|
import Spreadsheet::ParseExcel::Utility qw(int2col); |
774 |
|
|
775 |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
776 |
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!"; |
798 |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
799 |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
800 |
if ($cell) { |
if ($cell) { |
801 |
$row->{int2col($iC)} = $cell->Value; |
# this conversion is a cludge. |
802 |
|
# Files from Excell could have |
803 |
|
# characters which don't fit into |
804 |
|
# destination encoding. |
805 |
|
$row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value; |
806 |
} |
} |
807 |
} |
} |
808 |
|
|
828 |
} |
} |
829 |
} elsif ($type_base eq "marc") { |
} elsif ($type_base eq "marc") { |
830 |
|
|
831 |
use MARC; |
require MARC::File::USMARC; |
832 |
|
|
833 |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
834 |
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!"; |
835 |
|
|
836 |
# optional argument is format |
# optional argument is format |
837 |
my $format = x($config->{format}) || 'usmarc'; |
warn "marc_format is no longer used!" if ($config->{marc_format}); |
|
|
|
838 |
print STDERR "Reading MARC file '$marc_file'\n"; |
print STDERR "Reading MARC file '$marc_file'\n"; |
839 |
|
|
840 |
my $marc = new MARC; |
my $marc = MARC::File::USMARC->in( $marc_file ) |
841 |
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'"; |
|
842 |
|
|
843 |
my $i=0; # record nr. |
# count records in MARC file |
844 |
|
sub marc_count { |
845 |
|
my $filename = shift || die; |
846 |
|
my $file = MARC::File::USMARC->in($filename) || die $MARC::File::ERROR; |
847 |
|
my $count = 0; |
848 |
|
while ($file->skip()) { |
849 |
|
$count++; |
850 |
|
} |
851 |
|
return $count; |
852 |
|
} |
853 |
|
|
854 |
my $rec; |
my $count = marc_count($marc_file) || warn "no records in '$marc_file'?"; |
855 |
|
|
856 |
while ($marc->nextmarc(1)) { |
my $i = 0; |
857 |
|
|
858 |
# XXX |
while( my $rec = $marc->next() ) { |
859 |
fakeprogress($i++); |
|
860 |
|
progress($i++,$count); |
861 |
|
|
862 |
my $swishpath = $database."#".$i; |
my $swishpath = $database."#".$i; |
863 |
|
|
864 |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) { |
865 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
866 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
867 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
869 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
870 |
} |
} |
871 |
} |
} |
872 |
|
|
873 |
|
print STDERR "\n"; |
874 |
|
|
875 |
} elsif ($type_base eq "feed") { |
} elsif ($type_base eq "feed") { |
876 |
|
|
877 |
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
910 |
fakeprogress($i); |
fakeprogress($i); |
911 |
|
|
912 |
} |
} |
913 |
|
# close lookup |
914 |
|
untie %lhash if (%lhash); |
915 |
} |
} |
916 |
} |
} |
917 |
|
|
926 |
|
|
927 |
all2xml.pl - read various file formats and dump XML for SWISH-E |
all2xml.pl - read various file formats and dump XML for SWISH-E |
928 |
|
|
929 |
|
=head1 SYNOPSYS |
930 |
|
|
931 |
|
$ all2xml.pl [test.conf] |
932 |
|
|
933 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
934 |
|
|
935 |
This command will read ISIS data file using OpenIsis perl module, MARC |
This command will read ISIS data file using IsisDB perl module, MARC |
936 |
records using MARC module and optionally Micro$oft Excel files to |
records using MARC module and optionally Micro$oft Excel files to |
937 |
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, |
938 |
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 |
939 |
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. |
940 |
|
|
941 |
|
If no configuration file is specified, it will use default one called |
942 |
|
C<all2xml.conf>. |
943 |
|
|
944 |
=head1 BUGS |
=head1 BUGS |
945 |
|
|
946 |
Documentation is really lacking. However, in true Open Source spirit, source |
Documentation is really lacking. However, in true Open Source spirit, source |