9 |
use Text::Iconv; |
use Text::Iconv; |
10 |
use Config::IniFiles; |
use Config::IniFiles; |
11 |
use Encode; |
use Encode; |
12 |
|
#use GDBM_File; |
13 |
|
use Fcntl; # for O_RDWR |
14 |
|
use TDB_File; |
15 |
|
|
16 |
$|=1; |
$|=1; |
17 |
|
|
59 |
# format in XML file |
# format in XML file |
60 |
my %type2tag = ( |
my %type2tag = ( |
61 |
'isis' => 'isis', |
'isis' => 'isis', |
62 |
'excel' => 'column' |
'excel' => 'column', |
63 |
|
'marc' => 'marc', |
64 |
|
'feed' => 'feed' |
65 |
); |
); |
66 |
|
|
67 |
|
my $cache; # for cacheing |
68 |
|
|
69 |
|
# lookup hash (tied to file) |
70 |
|
my %lhash; |
71 |
|
# this option will cache all lookup entries in memory. |
72 |
|
# if you are tight on memory, turn this off |
73 |
|
my $use_lhash_cache = 1; |
74 |
|
|
75 |
|
my $last_field_name; # cache to prevent repeated fields |
76 |
|
|
77 |
sub data2xml { |
sub data2xml { |
78 |
|
|
79 |
use xmlify; |
use xmlify; |
95 |
|
|
96 |
# sort subrouting using order="" attribute |
# sort subrouting using order="" attribute |
97 |
sub by_order { |
sub by_order { |
98 |
return 0 if (! $config->{indexer}->{$a}->{order}); |
my $va = $config->{indexer}->{$a}->{order} || |
99 |
return 0 if (! $config->{indexer}->{$b}->{order}); |
$config->{indexer}->{$a}; |
100 |
|
my $vb = $config->{indexer}->{$b}->{order} || |
101 |
|
$config->{indexer}->{$b}; |
102 |
|
|
103 |
return $config->{indexer}->{$a}->{order} <=> |
return $va <=> $vb; |
|
$config->{indexer}->{$b}->{order} ; |
|
104 |
} |
} |
105 |
|
|
106 |
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
my @sorted_tags; |
107 |
|
if ($cache->{tags_by_order}) { |
108 |
|
@sorted_tags = @{$cache->{tags_by_order}}; |
109 |
|
} else { |
110 |
|
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
111 |
|
$cache->{tags_by_order} = \@sorted_tags; |
112 |
|
} |
113 |
|
|
114 |
$field=x($field); |
# lookup key |
115 |
|
my $lookup_key; |
116 |
|
|
117 |
|
# cache for field in pages |
118 |
|
delete $cache->{display_data}; |
119 |
|
delete $cache->{swish_data}; |
120 |
|
delete $cache->{swish_exact_data}; |
121 |
|
delete $cache->{index_data}; |
122 |
|
delete $cache->{index_delimiter}; |
123 |
|
my @page_fields; # names of fields |
124 |
|
|
125 |
|
|
126 |
|
# subs used to produce output |
127 |
|
|
128 |
|
sub get_field_name($$$) { |
129 |
|
my ($config,$field,$field_usage) = @_; |
130 |
|
|
131 |
|
# find field name (signular, plural) |
132 |
|
my $field_name = ""; |
133 |
|
if ($config->{indexer}->{$field}->{name_singular} && $field_usage == 1) { |
134 |
|
$field_name = $config->{indexer}->{$field}->{name_singular}; |
135 |
|
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
136 |
|
$field_name = $config->{indexer}->{$field}->{name_plural}; |
137 |
|
} elsif ($config->{indexer}->{$field}->{name}) { |
138 |
|
$field_name = $config->{indexer}->{$field}->{name}; |
139 |
|
} else { |
140 |
|
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
141 |
|
} |
142 |
|
if ($field_name) { |
143 |
|
if (! $last_field_name) { |
144 |
|
$last_field_name = x($field_name); |
145 |
|
return $last_field_name; |
146 |
|
} elsif ($field_name ne $last_field_name) { |
147 |
|
$last_field_name = x($field_name); |
148 |
|
return $last_field_name; |
149 |
|
} |
150 |
|
} |
151 |
|
} |
152 |
|
|
153 |
|
|
154 |
|
# begin real work: go field by field |
155 |
|
foreach my $field (@sorted_tags) { |
156 |
|
|
157 |
|
$field=x($field); |
158 |
$field_usage{$field}++; |
$field_usage{$field}++; |
159 |
|
|
160 |
my $swish_data = ""; |
my $swish_data = ""; |
161 |
|
my $swish_exact_data = ""; |
162 |
my $display_data = ""; |
my $display_data = ""; |
163 |
|
my @index_data; |
164 |
my $line_delimiter; |
my $line_delimiter; |
165 |
|
|
166 |
my ($swish,$display); |
my ($swish,$display); |
167 |
|
|
168 |
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"; |
169 |
|
|
170 |
|
# is this field page-by-page? |
171 |
|
my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page}; |
172 |
|
push @page_fields,$field if ($iterate_by_page); |
173 |
|
my %page_max = (); |
174 |
|
# default line_delimiter if using |
175 |
|
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
176 |
|
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter}; |
177 |
|
|
178 |
|
my $format_name = $config->{indexer}->{$field}->{format_name}; |
179 |
|
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter}; |
180 |
|
if ($format_name && $format_delimiter) { |
181 |
|
$cache->{format}->{$field}->{format_name} = $format_name; |
182 |
|
$cache->{format}->{$field}->{format_delimiter} = $format_delimiter; |
183 |
|
} |
184 |
|
|
185 |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
186 |
|
|
187 |
my $format = x($x->{content}); |
my $format = x($x->{content}); |
188 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
189 |
|
|
190 |
my $repeat_off = 0; # repeatable offset |
my $repeat_off = 0; # init repeatable offset |
191 |
|
|
192 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
# swish, swish_exact, display, index, index_lookup |
193 |
|
# swish and display defaults |
194 |
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
195 |
$s = 0 if (lc($x->{type}) eq "display"); |
$s = 0 if (lc($x->{type}) eq "display"); |
196 |
$d = 0 if (lc($x->{type}) eq "swish"); |
$d = 0 if (lc($x->{type}) eq "swish"); |
197 |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
($s,$se,$d,$i) = (0,1,0,1) if (lc($x->{type}) eq "index"); |
198 |
|
($s,$se,$d,$i) = (0,1,0,0) if (lc($x->{type}) eq "swish_exact"); |
199 |
|
($s,$se,$d,$i,$il) = (0,1,0,0,1) if (lc($x->{type}) =~ /^lookup/); |
200 |
|
|
201 |
# what will separate last line from this one? |
# what will separate last line from this one? |
202 |
if ($display_data && $x->{append} && $x->{append} eq "1") { |
if ($display_data && $x->{append}) { |
203 |
$line_delimiter = ' '; |
$line_delimiter = ' '; |
204 |
} elsif ($display_data) { |
} elsif ($display_data) { |
205 |
$line_delimiter = '<br/>'; |
$line_delimiter = '<br/>'; |
208 |
# init vars so that we go into while... |
# init vars so that we go into while... |
209 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
210 |
|
|
211 |
if ($swish || $display) { |
# placeholder for all repeatable entries for index |
212 |
|
|
213 |
|
sub chk_eval($) { |
214 |
|
my $data = shift; |
215 |
|
return if (! defined($data)); |
216 |
|
if ($data && $data =~ s/\s*eval{([^}]+)}\s*//) { |
217 |
|
if (eval "$1") { |
218 |
|
return $data; |
219 |
|
} else { |
220 |
|
return undef; |
221 |
|
} |
222 |
|
} else { |
223 |
|
return $data; |
224 |
|
} |
225 |
|
} |
226 |
|
|
227 |
|
sub mkformat($$) { |
228 |
|
my $x = shift || die "mkformat needs tag reference"; |
229 |
|
my $data = shift || return; |
230 |
|
my $format_name = x($x->{format_name}) || return chk_eval($data); |
231 |
|
my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!"; |
232 |
|
my $format_delimiter = x($x->{format_delimiter}); |
233 |
|
my @data; |
234 |
|
if ($format_delimiter) { |
235 |
|
@data = split(/$format_delimiter/,$data); |
236 |
|
} else { |
237 |
|
push @data,$data; |
238 |
|
} |
239 |
|
|
240 |
|
if ($fmt) { |
241 |
|
my $nr = scalar $fmt =~ s/%s/%s/g; |
242 |
|
if (($#data+1) == $nr) { |
243 |
|
return chk_eval(sprintf($fmt,@data)); |
244 |
|
} else { |
245 |
|
#print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
246 |
|
return chk_eval($data); |
247 |
|
} |
248 |
|
} else { |
249 |
|
print STDERR "usage of link '$format_name' without defined format (<link> tag)\n"; |
250 |
|
} |
251 |
|
} |
252 |
|
|
253 |
|
# while because of repeatable fields |
254 |
|
while ($swish || $display) { |
255 |
|
my $page = $repeat_off; |
256 |
|
$page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0)); |
257 |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
258 |
|
if ($repeat_off > 1000) { |
259 |
|
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
260 |
|
last; |
261 |
|
} |
262 |
|
|
263 |
|
# is this field is lookup? |
264 |
|
if ($display && $x->{lookup}) { |
265 |
|
my $null = "<!-- null -->"; |
266 |
|
if ($use_lhash_cache) { |
267 |
|
if (!defined($cache->{lhash}->{$display})) { |
268 |
|
my $new_display = $lhash{$display}; |
269 |
|
if (defined($new_display)) { |
270 |
|
#print STDERR "lookup cache store '$display' = '$new_display'\n"; |
271 |
|
$display = $new_display; |
272 |
|
$cache->{lhash}->{$display} = $new_display; |
273 |
|
} else { |
274 |
|
# print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
275 |
|
$display = ""; |
276 |
|
$cache->{lhash}->{$display} = $null; |
277 |
|
} |
278 |
|
} else { |
279 |
|
$display = $cache->{lhash}->{$display}; |
280 |
|
} |
281 |
|
} else { |
282 |
|
$display = $lhash{$display} || $null; |
283 |
|
} |
284 |
|
} |
285 |
|
|
286 |
# filter="name" ; filter this field through |
# filter="name" ; filter this field through |
287 |
# filter/[name].pm |
# filter/[name].pm |
288 |
my $filter = $x->{filter}; |
my $filter = $x->{filter}; |
289 |
if ($filter) { |
if ($filter && !$cache->{filter_loaded}->{$filter}) { |
290 |
require "filter/".$filter.".pm"; |
require "filter/".$filter.".pm"; |
291 |
|
$cache->{filter_loaded}->{$filter}++; |
292 |
} |
} |
293 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
294 |
if ($s && $swish) { |
if ($swish) { |
295 |
if ($filter) { |
if ($filter && ($s || $se)) { |
296 |
no strict 'refs'; |
no strict 'refs'; |
297 |
$swish_data .= join(" ",&$filter($swish)); |
my $tmp = join(" ",&$filter($swish)) if ($s || $se); |
298 |
|
$swish_data .= $tmp if ($s); |
299 |
|
$swish_exact_data .= "xxbxx $tmp xxexx " if ($se && $tmp ne ""); |
300 |
|
|
301 |
} else { |
} else { |
302 |
$swish_data .= $swish; |
$swish_data .= $swish if ($s); |
303 |
|
$swish_exact_data .= "xxbxx $swish xxexx " if ($se && $swish ne ""); |
304 |
} |
} |
305 |
|
|
306 |
|
$tmp = chk_eval($tmp); |
307 |
|
$swish_data .= $tmp if ($s && $tmp); |
308 |
|
$swish_exact_data .= "xxbxx $tmp xxexx " if ($se && $tmp ne ""); |
309 |
} |
} |
310 |
|
|
311 |
# type="display" ; field for display |
# type="display" ; field for display |
312 |
if ($d && $display) { |
if ($d && $display) { |
313 |
|
my $ldel = $delimiter; |
314 |
if ($line_delimiter && $display_data) { |
if ($line_delimiter && $display_data) { |
315 |
$display_data .= $line_delimiter; |
$ldel = $line_delimiter; |
|
undef $line_delimiter; |
|
316 |
} |
} |
317 |
if ($filter) { |
if ($filter) { |
318 |
no strict 'refs'; |
no strict 'refs'; |
319 |
$display_data .= join($delimiter,&$filter($display)); |
my @arr; |
320 |
} else { |
foreach my $tmp (&$filter($display)) { |
321 |
if ($display_data) { |
my $tmp2 = mkformat($x,$tmp); |
322 |
$display_data .= $delimiter.$display; |
push @arr,$tmp2 if ($tmp2); |
|
} else { |
|
|
$display_data .= $display; |
|
323 |
} |
} |
324 |
|
$display_data .= $ldel if ($display_data && @arr); |
325 |
|
$display_data .= join($delimiter,@arr); |
326 |
|
} else { |
327 |
|
$display_data .= $ldel if ($display_data); |
328 |
|
my $tmp = mkformat($x,$display); |
329 |
|
$display_data .= $tmp if ($tmp); |
330 |
} |
} |
331 |
} |
} |
332 |
|
|
333 |
# type="index" ; insert into index |
# type="index" ; insert into index |
334 |
|
my $idisplay; |
335 |
if ($i && $display) { |
if ($i && $display) { |
336 |
my $index_data = $display; |
$idisplay = $display; |
337 |
if ($filter) { |
if ($filter) { |
338 |
no strict 'refs'; |
no strict 'refs'; |
339 |
foreach my $d (&$filter($index_data)) { |
$idisplay = &$filter($idisplay); |
340 |
$index->insert($field, $d, $path); |
} |
341 |
|
$idisplay = chk_eval($idisplay); |
342 |
|
push @index_data, $idisplay if ($idisplay && !$iterate_by_page); |
343 |
|
} |
344 |
|
|
345 |
|
# store fields in lookup |
346 |
|
if ($il && $display) { |
347 |
|
if (lc($x->{type}) eq "lookup_key") { |
348 |
|
if ($lookup_key) { |
349 |
|
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
350 |
|
} else { |
351 |
|
if ($filter) { |
352 |
|
no strict 'refs'; |
353 |
|
$lookup_key = &$filter($display); |
354 |
|
} else { |
355 |
|
$lookup_key = $display; |
356 |
|
} |
357 |
} |
} |
358 |
} else { |
} elsif (lc($x->{type}) eq "lookup_val") { |
359 |
$index->insert($field, $index_data, $path); |
if ($lookup_key) { |
360 |
|
if ($filter) { |
361 |
|
no strict 'refs'; |
362 |
|
$lhash{$lookup_key} = &$filter($display); |
363 |
|
} else { |
364 |
|
$lhash{$lookup_key} = $display; |
365 |
|
} |
366 |
|
} else { |
367 |
|
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
368 |
|
} |
369 |
|
} |
370 |
|
|
371 |
|
} |
372 |
|
|
373 |
|
# store data for page-by-page repeatable fields |
374 |
|
if ($iterate_by_page) { |
375 |
|
sub iterate_fld($$$$$$) { |
376 |
|
my ($cache,$what,$field,$page,$data,$append) = @_; |
377 |
|
return if (!$data); |
378 |
|
|
379 |
|
my $ldel = $page_line_delimiter; |
380 |
|
$ldel = " " if ($append); |
381 |
|
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
382 |
|
if (! $cache->{$what}->{$field}->[$page]) { |
383 |
|
$cache->{$what}->{$field}->[$page] = $data; |
384 |
|
} else { |
385 |
|
$cache->{$what}->{$field}->[$page] .= $ldel.$data; |
386 |
|
} |
387 |
|
} |
388 |
|
|
389 |
|
if ($display_data) { |
390 |
|
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
391 |
|
} |
392 |
|
$display_data = ""; |
393 |
|
if ($swish_data) { |
394 |
|
iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append}); |
395 |
|
$swish_data = ""; |
396 |
|
} |
397 |
|
if ($swish_exact_data) { |
398 |
|
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
399 |
|
$swish_exact_data = ""; |
400 |
|
} |
401 |
|
|
402 |
|
if ($idisplay) { |
403 |
|
my $ldel=$page_line_delimiter; |
404 |
|
my @index_data; |
405 |
|
if ($cache->{index_data}->{$field}->[$page]) { |
406 |
|
|
407 |
|
@index_data = @{$cache->{index_data}->{$field}->[$page]}; |
408 |
|
} |
409 |
|
if ($x->{append}) { |
410 |
|
if (@index_data) { |
411 |
|
$index_data[$#index_data] .= $idisplay; |
412 |
|
} else { |
413 |
|
push @index_data, $idisplay; |
414 |
|
} |
415 |
|
} else { |
416 |
|
push @index_data, $idisplay; |
417 |
|
} |
418 |
|
$idisplay = ""; |
419 |
|
@{$cache->{index_data}->{$field}->[$page]} = @index_data; |
420 |
|
} |
421 |
|
} |
422 |
|
} |
423 |
|
|
424 |
|
if (! $iterate_by_page) { |
425 |
|
my $idel = $x->{index_delimiter}; |
426 |
|
# fill data in index |
427 |
|
foreach my $tmp (@index_data) { |
428 |
|
my $i = $d = $tmp; |
429 |
|
if ($idel && $tmp =~ m/$idel/) { |
430 |
|
($i,$d) = split(/$idel/,$tmp); |
431 |
} |
} |
432 |
|
$index->insert($field, $i, $d, $path); |
433 |
} |
} |
434 |
|
@index_data = (); |
435 |
} |
} |
436 |
} |
} |
437 |
|
|
438 |
# now try to parse variables from configuration file |
# now try to parse variables from configuration file |
439 |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
440 |
|
|
441 |
|
my $delimiter = x($x->{delimiter}) || ' '; |
442 |
my $val = $cfg->val($database, x($x->{content})); |
my $val = $cfg->val($database, x($x->{content})); |
443 |
|
|
444 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
my ($s,$d,$i) = (1,1,0); # swish, display default |
445 |
$s = 0 if (lc($x->{type}) eq "display"); |
$s = 0 if (lc($x->{type}) eq "display"); |
446 |
$d = 0 if (lc($x->{type}) eq "swish"); |
$d = 0 if (lc($x->{type}) eq "swish"); |
447 |
|
# no support for swish exact in config. |
448 |
|
# IMHO, it's useless |
449 |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
450 |
|
|
451 |
if ($val) { |
if ($val) { |
452 |
$display_data .= $val if ($d); |
$display_data .= $delimiter.$val if ($d); |
453 |
$swish_data .= $val if ($s); |
$swish_data .= $val if ($s); |
454 |
$index->insert($field, $val, $path) if ($i); |
$index->insert($field, $val, $path) if ($i); |
455 |
} |
} |
456 |
|
|
457 |
|
if ($iterate_by_page) { |
458 |
|
# FIXME data from config tag will appear just |
459 |
|
# on first page!!! |
460 |
|
my $page = 0; |
461 |
|
if ($display_data) { |
462 |
|
$cache->{display_data}->{$field}->[$page] = $display_data; |
463 |
|
$display_data = ""; |
464 |
|
} |
465 |
|
if ($swish_data) { |
466 |
|
$cache->{swish_data}->{$field}->[$page] = $swish_data; |
467 |
|
$swish_data = ""; |
468 |
|
} |
469 |
|
if ($swish_exact_data) { |
470 |
|
$cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data; |
471 |
|
$swish_exact_data = ""; |
472 |
|
} |
473 |
|
} |
474 |
} |
} |
475 |
|
|
476 |
|
# save data page-by-page |
477 |
|
foreach my $field (@page_fields) { |
478 |
|
my $nr_pages = $page_max{$field} || next; |
479 |
|
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
480 |
|
#print STDERR Dumper($cache->{display_data}); |
481 |
|
for (my $page=0; $page <= $nr_pages; $page++) { |
482 |
|
my $display_data; |
483 |
|
if ($cache->{format}->{$field}) { |
484 |
|
my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]); |
485 |
|
$display_data=$tmp if ($tmp); |
486 |
|
} else { |
487 |
|
$display_data = $cache->{display_data}->{$field}->[$page]; |
488 |
|
} |
489 |
|
if ($display_data) { # default |
490 |
|
if ($field eq "headline") { |
491 |
|
$xml .= xmlify("headline", $display_data); |
492 |
|
} else { |
493 |
|
|
494 |
if ($display_data) { |
# fallback to empty field name if needed |
495 |
|
$html .= get_field_name($config,$field,$field_usage{$field}) || ''; |
496 |
|
$html .= "#-#".$display_data."###\n"; |
497 |
|
} |
498 |
|
} |
499 |
|
|
500 |
|
my $swish_data = $cache->{swish_data}->{$field}->[$page]; |
501 |
|
if ($swish_data) { |
502 |
|
# remove extra spaces |
503 |
|
$swish_data =~ s/ +/ /g; |
504 |
|
$swish_data =~ s/ +$//g; |
505 |
|
|
506 |
if ($field eq "headline") { |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
|
$xml .= xmlify("headline", $display_data); |
|
|
} else { |
|
|
|
|
|
# find field name (signular, plural) |
|
|
my $field_name = ""; |
|
|
if ($config->{indexer}->{$field}->{name_singular} && $field_usage{$field} == 1) { |
|
|
$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}."#-#"; |
|
|
} else { |
|
|
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
|
507 |
} |
} |
508 |
if ($field_name) { |
|
509 |
$html .= x($field_name); |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
510 |
|
if ($swish_exact_data) { |
511 |
|
$swish_exact_data =~ s/ +/ /g; |
512 |
|
$swish_exact_data =~ s/ +$//g; |
513 |
|
|
514 |
|
# add delimiters before and after word. |
515 |
|
# That is required to produce exact match |
516 |
|
$xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data)); |
517 |
|
} |
518 |
|
|
519 |
|
my $idel = $cache->{index_delimiter}->{$field}; |
520 |
|
foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) { |
521 |
|
my $i = $tmp; |
522 |
|
my $d = $tmp; |
523 |
|
if ($idel && $tmp =~ m/$idel/) { |
524 |
|
($i,$d) = split(/$idel/,$tmp); |
525 |
|
} |
526 |
|
$index->insert($field, $i, $d, $path); |
527 |
|
#print STDERR "index [$idel] $field: $i --> $d [$path]\n"; |
528 |
} |
} |
|
$html .= $display_data."###\n"; |
|
529 |
} |
} |
530 |
|
|
531 |
} |
} |
532 |
if ($swish_data) { |
|
533 |
# remove extra spaces |
if (! $iterate_by_page) { |
534 |
$swish_data =~ s/ +/ /g; |
if ($display_data) { |
535 |
$swish_data =~ s/ +$//g; |
if ($field eq "headline") { |
536 |
|
$xml .= xmlify("headline", $display_data); |
537 |
|
} else { |
538 |
|
|
539 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
# fallback to empty field name if needed |
540 |
} |
$html .= get_field_name($config,$field,$field_usage{$field}) || ''; |
541 |
|
$html .= "#-#".$display_data."###\n"; |
542 |
|
} |
543 |
|
} |
544 |
|
if ($swish_data) { |
545 |
|
# remove extra spaces |
546 |
|
$swish_data =~ s/ +/ /g; |
547 |
|
$swish_data =~ s/ +$//g; |
548 |
|
|
549 |
|
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
550 |
|
} |
551 |
|
|
552 |
|
if ($swish_exact_data) { |
553 |
|
$swish_exact_data =~ s/ +/ /g; |
554 |
|
$swish_exact_data =~ s/ +$//g; |
555 |
|
|
556 |
|
# add delimiters before and after word. |
557 |
|
# That is required to produce exact match |
558 |
|
$xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data)); |
559 |
|
} |
560 |
|
} |
561 |
} |
} |
562 |
|
|
563 |
# dump formatted output in <html> |
# dump formatted output in <html> |
564 |
if ($html) { |
if ($html) { |
565 |
$xml .= xmlify("html",$html); |
#$xml .= xmlify("html",$html); |
566 |
|
$xml .= "<html><![CDATA[ $html ]]></html>"; |
567 |
} |
} |
568 |
|
|
569 |
if ($xml) { |
if ($xml) { |
590 |
$cfg_global->val('global', 'dbi_passwd') || '', |
$cfg_global->val('global', 'dbi_passwd') || '', |
591 |
); |
); |
592 |
|
|
593 |
|
my $show_progress = $cfg_global->val('global', 'show_progress'); |
594 |
|
|
595 |
|
my $unac_filter = $cfg_global->val('global', 'unac_filter'); |
596 |
|
if ($unac_filter) { |
597 |
|
require $unac_filter; |
598 |
|
} |
599 |
|
|
600 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
601 |
|
|
602 |
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"; |
603 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
604 |
|
|
605 |
|
# create new lookup file |
606 |
|
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
607 |
|
if ($lookup_file) { |
608 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
609 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
610 |
|
print STDERR "creating lookup file '$lookup_file'\n"; |
611 |
|
# delete memory cache for lookup file |
612 |
|
delete $cache->{lhash}; |
613 |
|
} |
614 |
|
|
615 |
|
# open existing lookup file |
616 |
|
$lookup_file = $cfg -> val($database, 'lookup_open'); # optional |
617 |
|
if ($lookup_file) { |
618 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644; |
619 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644; |
620 |
|
print STDERR "opening lookup file '$lookup_file'\n"; |
621 |
|
} |
622 |
|
|
623 |
print STDERR "reading ./import_xml/$type.xml\n"; |
print STDERR "reading ./import_xml/$type.xml\n"; |
624 |
|
|
625 |
# extract just type basic |
# extract just type basic |
626 |
my $type_base = $type; |
my $type_base = $type; |
627 |
$type_base =~ s/_.+$//g; |
$type_base =~ s/_.+$//g; |
628 |
|
|
629 |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config' ], forcecontent => 1); |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1); |
630 |
|
|
631 |
# output current progress indicator |
# output current progress indicator |
632 |
my $last_p = 0; |
my $last_p = 0; |
633 |
sub progress { |
sub progress { |
634 |
#return if (! $opts{q}); # FIXME |
return if (! $show_progress); |
635 |
my $current = shift; |
my $current = shift; |
636 |
my $total = shift || 1; |
my $total = shift || 1; |
637 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
641 |
} |
} |
642 |
} |
} |
643 |
|
|
644 |
|
my $fake_dir = 1; |
645 |
|
sub fakeprogress { |
646 |
|
return if (! $show_progress); |
647 |
|
my $current = shift @_; |
648 |
|
|
649 |
|
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
650 |
|
|
651 |
|
$last_p += $fake_dir; |
652 |
|
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
653 |
|
if ($last_p % 10 == 0) { |
654 |
|
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
655 |
|
} |
656 |
|
} |
657 |
|
|
658 |
# now read database |
# now read database |
659 |
print STDERR "using: $type...\n"; |
print STDERR "using: $type...\n"; |
660 |
|
|
661 |
|
# erase cache for tags by order in this database |
662 |
|
delete $cache->{tags_by_order}; |
663 |
|
|
664 |
if ($type_base eq "isis") { |
if ($type_base eq "isis") { |
665 |
|
|
666 |
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!"; |
668 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
669 |
my $db = OpenIsis::open( $isis_db ); |
my $db = OpenIsis::open( $isis_db ); |
670 |
|
|
671 |
|
# check if .txt database for OpenIsis is zero length, |
672 |
|
# if so, erase it and re-open database |
673 |
|
sub check_txt_db { |
674 |
|
my $isis_db = shift || die "need isis database name"; |
675 |
|
my $reopen = 0; |
676 |
|
|
677 |
|
if (-e $isis_db.".TXT") { |
678 |
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
679 |
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
680 |
|
$reopen++; |
681 |
|
} |
682 |
|
if (-e $isis_db.".PTR") { |
683 |
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
684 |
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
685 |
|
$reopen++; |
686 |
|
} |
687 |
|
return OpenIsis::open( $isis_db ) if ($reopen); |
688 |
|
} |
689 |
|
|
690 |
|
# EOF error |
691 |
|
if ($db == -1) { |
692 |
|
$db = check_txt_db($isis_db); |
693 |
|
if ($db == -1) { |
694 |
|
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
695 |
|
next; |
696 |
|
} |
697 |
|
} |
698 |
|
|
699 |
|
# OpenIsis::ERR_BADF |
700 |
|
if ($db == -4) { |
701 |
|
print STDERR "FATAL: OpenIsis can't find file $isis_db\n"; |
702 |
|
next; |
703 |
|
# OpenIsis::ERR_IO |
704 |
|
} elsif ($db == -5) { |
705 |
|
print STDERR "FATAL: OpenIsis can't access file $isis_db\n"; |
706 |
|
next; |
707 |
|
} elsif ($db < 0) { |
708 |
|
print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n"; |
709 |
|
next; |
710 |
|
} |
711 |
|
|
712 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
my $max_rowid = OpenIsis::maxRowid( $db ); |
713 |
|
|
714 |
|
# if 0 records, try to rease isis .txt database |
715 |
|
if ($max_rowid == 0) { |
716 |
|
# force removal of database |
717 |
|
$db = check_txt_db($isis_db); |
718 |
|
$max_rowid = OpenIsis::maxRowid( $db ); |
719 |
|
} |
720 |
|
|
721 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
722 |
|
|
723 |
my $path = $database; |
my $path = $database; |
739 |
} |
} |
740 |
} |
} |
741 |
} |
} |
742 |
|
# for this to work with current version of OpenIsis (0.9.0) |
743 |
|
# you might need my patch from |
744 |
|
# https://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
745 |
|
OpenIsis::close($db); |
746 |
print STDERR "\n"; |
print STDERR "\n"; |
747 |
|
|
748 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
749 |
use Spreadsheet::ParseExcel; |
require Spreadsheet::ParseExcel; |
750 |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
require Spreadsheet::ParseExcel::Utility; |
751 |
|
import Spreadsheet::ParseExcel::Utility qw(int2col); |
752 |
|
|
753 |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
754 |
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 "$database doesn't have 'excel_file' defined!"; |
800 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
801 |
} |
} |
802 |
} |
} |
803 |
|
} elsif ($type_base eq "marc") { |
804 |
|
|
805 |
|
require MARC; |
806 |
|
|
807 |
|
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
808 |
|
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
809 |
|
|
810 |
|
# optional argument is format |
811 |
|
my $format = x($config->{marc_format}) || 'usmarc'; |
812 |
|
|
813 |
|
print STDERR "Reading MARC file '$marc_file'\n"; |
814 |
|
|
815 |
|
my $marc = new MARC; |
816 |
|
my $nr = $marc->openmarc({ |
817 |
|
file=>$marc_file, format=>$format |
818 |
|
}) || die "Can't open MARC file '$marc_file' with format '$format'"; |
819 |
|
|
820 |
|
# read MARC file in memory |
821 |
|
$marc->nextmarc(-1); |
822 |
|
|
823 |
|
my $max_rec = $marc->marc_count(); |
824 |
|
|
825 |
|
for(my $i=1; $i<=$max_rec; $i++) { |
826 |
|
|
827 |
|
progress($i,$max_rec); |
828 |
|
|
829 |
|
# store value for marc_sf.pm |
830 |
|
$main::cache->{marc_record} = $i; |
831 |
|
|
832 |
|
my $swishpath = $database."#".$i; |
833 |
|
|
834 |
|
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
835 |
|
$xml = $cp2utf->convert($xml); |
836 |
|
use bytes; # as opposed to chars |
837 |
|
print "Path-Name: $swishpath\n"; |
838 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
839 |
|
print "Document-Type: XML\n\n$xml\n"; |
840 |
|
} |
841 |
|
} |
842 |
|
|
843 |
|
print STDERR "\n"; |
844 |
|
|
845 |
|
} elsif ($type_base eq "feed") { |
846 |
|
|
847 |
|
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
848 |
|
my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!"; |
849 |
|
|
850 |
|
print STDERR "Reading feed from program '$prog'\n"; |
851 |
|
|
852 |
|
open(FEED,"feeds/$prog |") || die "can't start $prog: $!"; |
853 |
|
|
854 |
|
my $i=1; # record nr. |
855 |
|
|
856 |
|
my $data; |
857 |
|
my $line=1; |
858 |
|
|
859 |
|
while (<FEED>) { |
860 |
|
chomp; |
861 |
|
|
862 |
|
if (/^$/) { |
863 |
|
my $swishpath = $database."#".$i++; |
864 |
|
|
865 |
|
if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) { |
866 |
|
$xml = $cp2utf->convert($xml); |
867 |
|
use bytes; # as opposed to chars |
868 |
|
print "Path-Name: $swishpath\n"; |
869 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
870 |
|
print "Document-Type: XML\n\n$xml\n"; |
871 |
|
} |
872 |
|
$line = 1; |
873 |
|
$data = {}; |
874 |
|
next; |
875 |
|
} |
876 |
|
|
877 |
|
$line = $1 if (s/^(\d+):\s*//); |
878 |
|
$data->{$line++} = $_; |
879 |
|
|
880 |
|
fakeprogress($i); |
881 |
|
|
882 |
|
} |
883 |
|
# close lookup |
884 |
|
untie %lhash if (%lhash); |
885 |
} |
} |
886 |
} |
} |
887 |
|
|
894 |
|
|
895 |
=head1 NAME |
=head1 NAME |
896 |
|
|
897 |
isis2xml.pl - read isis file and dump XML |
all2xml.pl - read various file formats and dump XML for SWISH-E |
898 |
|
|
899 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
900 |
|
|
901 |
This command will read ISIS data file using OpenIsis perl module and |
This command will read ISIS data file using OpenIsis perl module, MARC |
902 |
create XML file for usage with I<SWISH-E> |
records using MARC module and optionally Micro$oft Excel files to |
903 |
indexer. Dispite it's name, this script B<isn't general xml generator> |
create one XML file for usage with I<SWISH-E> indexer. Dispite it's name, |
904 |
from isis files (isis allready has something like that). Output of this |
this script B<isn't general xml generator> from isis files (isis allready |
905 |
script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |
906 |
|
|
907 |
|
=head1 BUGS |
908 |
|
|
909 |
|
Documentation is really lacking. However, in true Open Source spirit, source |
910 |
|
is best documentation. I even made considerable effort to comment parts |
911 |
|
which are not intuitively clear, so... |
912 |
|
|
913 |
=head1 AUTHOR |
=head1 AUTHOR |
914 |
|
|