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 |
|
use Carp; |
15 |
|
|
16 |
$|=1; |
$|=1; |
17 |
|
|
18 |
my $config_file = $0; |
my $config_file = $0; |
19 |
$config_file =~ s/\.pl$/.conf/; |
$config_file =~ s/\.pl$/.conf/; |
20 |
|
$config_file = $ARGV[0] if ($ARGV[0] && -f $ARGV[0]); |
21 |
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
22 |
|
|
23 |
my $config; |
my $config; |
24 |
|
|
25 |
#use index_DBI; # default DBI module for index |
#use index_DBI; # default DBI module for index |
26 |
use index_DBI_cache; # faster DBI module using memory cache |
#use index_DBI_cache; # faster DBI module using memory cache |
27 |
|
use index_DBI_filter; # filter support for indexes |
28 |
my $index; |
my $index; |
29 |
|
|
30 |
my %opts; |
my %opts; |
63 |
'isis' => 'isis', |
'isis' => 'isis', |
64 |
'excel' => 'column', |
'excel' => 'column', |
65 |
'marc' => 'marc', |
'marc' => 'marc', |
66 |
|
'feed' => 'feed', |
67 |
); |
); |
68 |
|
|
69 |
|
my $cache; # for cacheing |
70 |
|
|
71 |
|
# lookup hash (tied to file) |
72 |
|
my %lhash; |
73 |
|
# this option will cache all lookup entries in memory. |
74 |
|
# if you are tight on memory, turn this off |
75 |
|
my $use_lhash_cache = 1; |
76 |
|
|
77 |
|
my $last_field_name; # cache to prevent repeated fields |
78 |
|
|
79 |
|
my $broken_cdata = XMLin('<foo><![CDATA[<bar>]]></foo>') eq '<bar>>'; |
80 |
|
warn "XML::Simple on this system seems broken with <![CDATA[..]]>.\n" if ($broken_cdata); |
81 |
|
|
82 |
sub data2xml { |
sub data2xml { |
83 |
|
|
84 |
use xmlify; |
use xmlify; |
100 |
|
|
101 |
# sort subrouting using order="" attribute |
# sort subrouting using order="" attribute |
102 |
sub by_order { |
sub by_order { |
103 |
return 0 if (! $config->{indexer}->{$a}->{order}); |
my $va = $config->{indexer}->{$a}->{order} || |
104 |
return 0 if (! $config->{indexer}->{$b}->{order}); |
$config->{indexer}->{$a}; |
105 |
|
my $vb = $config->{indexer}->{$b}->{order} || |
106 |
|
$config->{indexer}->{$b}; |
107 |
|
|
108 |
return $config->{indexer}->{$a}->{order} <=> |
return $va <=> $vb; |
|
$config->{indexer}->{$b}->{order} ; |
|
109 |
} |
} |
110 |
|
|
111 |
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
my @sorted_tags; |
112 |
|
if ($cache->{tags_by_order}) { |
113 |
|
@sorted_tags = @{$cache->{tags_by_order}}; |
114 |
|
} else { |
115 |
|
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
116 |
|
$cache->{tags_by_order} = \@sorted_tags; |
117 |
|
} |
118 |
|
|
119 |
|
if (! @sorted_tags) { |
120 |
|
print STDERR "WARNING: no tags for this type found in import_xml file!\n"; |
121 |
|
} |
122 |
|
|
123 |
|
# lookup key |
124 |
|
my $lookup_key; |
125 |
|
|
126 |
|
# cache for field in pages |
127 |
|
delete $cache->{display_data}; |
128 |
|
delete $cache->{swish_data}; |
129 |
|
delete $cache->{swish_exact_data}; |
130 |
|
delete $cache->{index_data}; |
131 |
|
delete $cache->{index_delimiter}; |
132 |
|
delete $cache->{distinct}; |
133 |
|
my @page_fields; # names of fields |
134 |
|
|
135 |
|
|
136 |
|
# subs used to produce output |
137 |
|
|
138 |
|
sub get_field_name($$$) { |
139 |
|
my ($config,$field,$field_usage) = @_; |
140 |
|
|
141 |
|
# find field name (signular, plural) |
142 |
|
my $field_name = ""; |
143 |
|
if ($config->{indexer}->{$field}->{name_singular} && $field_usage == 1) { |
144 |
|
$field_name = $config->{indexer}->{$field}->{name_singular}; |
145 |
|
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
146 |
|
$field_name = $config->{indexer}->{$field}->{name_plural}; |
147 |
|
} elsif ($config->{indexer}->{$field}->{name}) { |
148 |
|
$field_name = $config->{indexer}->{$field}->{name}; |
149 |
|
} else { |
150 |
|
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
151 |
|
} |
152 |
|
|
153 |
|
if ($field_name) { |
154 |
|
$field_name = x($field_name); |
155 |
|
if (! $last_field_name) { |
156 |
|
$last_field_name = $field_name; |
157 |
|
return $last_field_name; |
158 |
|
} elsif ($field_name ne $last_field_name) { |
159 |
|
$last_field_name = $field_name; |
160 |
|
return $last_field_name; |
161 |
|
} |
162 |
|
} |
163 |
|
} |
164 |
|
|
165 |
|
|
166 |
|
# init variables for different types |
167 |
|
sub init_visible_type($) { |
168 |
|
my $type = shift; |
169 |
|
|
170 |
|
# swish, swish_exact, display, index, index_lookup |
171 |
|
# swish and display defaults |
172 |
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
173 |
|
if (lc($type) eq "display") { |
174 |
|
$s = 0; |
175 |
|
} elsif (lc($type) eq "swish") { |
176 |
|
$d = 0; |
177 |
|
} elsif (lc($type) eq "index") { |
178 |
|
($s,$se,$d,$i) = (0,1,0,1); |
179 |
|
} elsif (lc($type) eq "swish_exact") { |
180 |
|
($s,$se,$d,$i) = (0,1,0,0); |
181 |
|
} elsif (lc($type) =~ /^lookup/) { |
182 |
|
($s,$se,$d,$i,$il) = (0,1,0,0,1); |
183 |
|
} elsif ($type) { |
184 |
|
print STDERR "WARNING: unknown type: $type\n"; |
185 |
|
} |
186 |
|
return ($s,$se,$d,$i,$il); |
187 |
|
} |
188 |
|
|
|
$field=x($field); |
|
189 |
|
|
190 |
|
# convert |
191 |
|
# |
192 |
|
# <tag> |
193 |
|
# <delimiter>, </delimiter> |
194 |
|
# <value>200a</value> |
195 |
|
# </tag> |
196 |
|
# |
197 |
|
# to |
198 |
|
# |
199 |
|
# <tag delimiter=", ">200a</tag> |
200 |
|
# |
201 |
|
# but without loosing spaces in delimiter (becasue |
202 |
|
# new XML::Simple strips spaces in attribute values |
203 |
|
# as defined in XML specification) |
204 |
|
# |
205 |
|
sub unroll_x($) { |
206 |
|
my $x = shift; |
207 |
|
|
208 |
|
if (defined $x->{value}) { |
209 |
|
my ($v,$d) = ($x->{value}->{content}, $x->{delimiter}->{content}); |
210 |
|
delete $x->{value}; |
211 |
|
delete $x->{delimiter}; |
212 |
|
$x->{content} = $v; |
213 |
|
$d =~ s#>$## if ($d && $broken_cdata); |
214 |
|
$x->{delimiter} = $d; |
215 |
|
} |
216 |
|
return $x; |
217 |
|
} |
218 |
|
|
219 |
|
# begin real work: go field by field |
220 |
|
foreach my $field (@sorted_tags) { |
221 |
|
|
222 |
|
$field=x($field); |
223 |
$field_usage{$field}++; |
$field_usage{$field}++; |
224 |
|
|
225 |
my $swish_data = ""; |
my $swish_data = ""; |
226 |
|
my $swish_exact_data = ""; |
227 |
my $display_data = ""; |
my $display_data = ""; |
228 |
|
my @index_data; |
229 |
my $line_delimiter; |
my $line_delimiter; |
230 |
|
|
231 |
my ($swish,$display); |
my ($swish,$display); |
232 |
|
|
233 |
my $tag = $type2tag{$type} || die "can't find which tag to use for type $type"; |
my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type} || die "can't find which tag to use for type $type"; |
234 |
|
|
235 |
|
# is this field page-by-page? |
236 |
|
my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page}; |
237 |
|
push @page_fields,$field if ($iterate_by_page); |
238 |
|
my %page_max = (); |
239 |
|
# default line_delimiter if using |
240 |
|
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
241 |
|
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter}; |
242 |
|
my $distinct = $config->{indexer}->{$field}->{distinct}; |
243 |
|
if ($distinct && !$iterate_by_page) { |
244 |
|
warn "WARNING: distinct is currently not supported without iterate_by_page!\n"; |
245 |
|
$distinct = 0; |
246 |
|
} |
247 |
|
|
248 |
|
my $format_name = $config->{indexer}->{$field}->{format_name}; |
249 |
|
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter}; |
250 |
|
if ($format_name && $format_delimiter) { |
251 |
|
$cache->{format}->{$field}->{format_name} = $format_name; |
252 |
|
$cache->{format}->{$field}->{format_delimiter} = $format_delimiter; |
253 |
|
} |
254 |
|
|
255 |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
256 |
|
|
257 |
|
$x = unroll_x($x); |
258 |
|
|
259 |
my $format = x($x->{content}); |
my $format = x($x->{content}); |
260 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
261 |
|
|
262 |
my $repeat_off = 0; # repeatable offset |
my $repeat_off = 0; # init repeatable offset |
263 |
|
|
264 |
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"); |
|
265 |
|
|
266 |
# what will separate last line from this one? |
# what will separate last line from this one? |
267 |
if ($display_data && $x->{append} && $x->{append} eq "1") { |
if ($display_data && $x->{append}) { |
268 |
$line_delimiter = ' '; |
$line_delimiter = $delimiter; |
269 |
} elsif ($display_data) { |
} elsif ($display_data) { |
270 |
$line_delimiter = '<br/>'; |
$line_delimiter = '<br/>'; |
271 |
} |
} |
273 |
# init vars so that we go into while... |
# init vars so that we go into while... |
274 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
275 |
|
|
276 |
if ($swish || $display) { |
sub mkformat($$) { |
277 |
|
my $x = shift || die "mkformat needs tag reference"; |
278 |
|
my $data = shift || return; |
279 |
|
my $format_name = x($x->{format_name}) || return $data; |
280 |
|
my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!"; |
281 |
|
my $format_delimiter = x($x->{format_delimiter}); |
282 |
|
my @data; |
283 |
|
if ($format_delimiter) { |
284 |
|
@data = split(/$format_delimiter/,$data); |
285 |
|
} else { |
286 |
|
push @data,$data; |
287 |
|
} |
288 |
|
|
289 |
|
if ($fmt) { |
290 |
|
my $nr = scalar $fmt =~ s/%s/%s/g; |
291 |
|
if (($#data+1) == $nr) { |
292 |
|
return sprintf($fmt,@data); |
293 |
|
} else { |
294 |
|
#print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
295 |
|
return $data; |
296 |
|
} |
297 |
|
} else { |
298 |
|
print STDERR "usage of link '$format_name' without defined format (<link> tag)\n"; |
299 |
|
} |
300 |
|
} |
301 |
|
|
302 |
|
# while because of repeatable fields |
303 |
|
while ($swish || $display) { |
304 |
|
my $page = $repeat_off; |
305 |
|
$page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0)); |
306 |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
307 |
|
if ($repeat_off > 1000) { |
308 |
|
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
309 |
|
last; |
310 |
|
} |
311 |
|
|
312 |
|
# is this field is lookup? |
313 |
|
if ($display && $x->{lookup}) { |
314 |
|
my $null = "<!-- null -->"; |
315 |
|
if ($use_lhash_cache) { |
316 |
|
if (!defined($cache->{lhash}->{$display})) { |
317 |
|
my $new_display = $lhash{$display}; |
318 |
|
if (defined($new_display)) { |
319 |
|
#print STDERR "lookup cache store '$display' = '$new_display'\n"; |
320 |
|
$display = $new_display; |
321 |
|
$cache->{lhash}->{$display} = $new_display; |
322 |
|
} else { |
323 |
|
# print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
324 |
|
$display = ""; |
325 |
|
$cache->{lhash}->{$display} = $null; |
326 |
|
} |
327 |
|
} else { |
328 |
|
$display = $cache->{lhash}->{$display}; |
329 |
|
} |
330 |
|
} else { |
331 |
|
$display = $lhash{$display} || $null; |
332 |
|
} |
333 |
|
} |
334 |
|
|
335 |
# filter="name" ; filter this field through |
# filter="name" ; filter this field through |
336 |
# filter/[name].pm |
# filter/[name].pm |
337 |
my $filter = $x->{filter}; |
my $filter = $x->{filter}; |
338 |
if ($filter) { |
if ($filter && !$cache->{filter_loaded}->{$filter}) { |
339 |
require "filter/".$filter.".pm"; |
require "filter/".$filter.".pm"; |
340 |
|
$cache->{filter_loaded}->{$filter}++; |
341 |
} |
} |
342 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
343 |
if ($s && $swish) { |
if ($swish) { |
344 |
if ($filter) { |
my $tmp = $swish; |
345 |
|
if ($filter && ($s || $se)) { |
346 |
no strict 'refs'; |
no strict 'refs'; |
347 |
$swish_data .= join(" ",&$filter($swish)); |
$tmp = join(" ",&$filter($tmp)) if ($s || $se); |
|
} else { |
|
|
$swish_data .= $swish; |
|
348 |
} |
} |
349 |
|
|
350 |
|
$swish_data .= $tmp if ($s && $tmp); |
351 |
|
$swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se); |
352 |
} |
} |
353 |
|
|
354 |
# type="display" ; field for display |
# type="display" ; field for display |
355 |
if ($d && $display) { |
if ($d && $display) { |
356 |
|
my $ldel = $delimiter; |
357 |
if ($line_delimiter && $display_data) { |
if ($line_delimiter && $display_data) { |
358 |
$display_data .= $line_delimiter; |
$ldel = $line_delimiter; |
|
undef $line_delimiter; |
|
359 |
} |
} |
360 |
if ($filter) { |
if ($filter) { |
361 |
no strict 'refs'; |
no strict 'refs'; |
362 |
$display_data .= join($delimiter,&$filter($display)); |
my @arr; |
363 |
} else { |
foreach my $tmp (&$filter($display)) { |
364 |
if ($display_data) { |
my $tmp2 = mkformat($x,$tmp); |
365 |
$display_data .= $delimiter.$display; |
push @arr,$tmp2 if ($tmp2); |
|
} else { |
|
|
$display_data .= $display; |
|
366 |
} |
} |
367 |
|
$display_data .= $ldel if ($display_data && @arr); |
368 |
|
$display_data .= join($delimiter,@arr); |
369 |
|
} else { |
370 |
|
$display_data .= $ldel if ($display_data); |
371 |
|
my $tmp = mkformat($x,$display); |
372 |
|
$display_data .= $tmp if ($tmp); |
373 |
} |
} |
374 |
} |
} |
375 |
|
|
376 |
# type="index" ; insert into index |
# type="index" ; insert into index |
377 |
|
my $idisplay; |
378 |
if ($i && $display) { |
if ($i && $display) { |
379 |
my $index_data = $display; |
$idisplay = $display; |
380 |
if ($filter) { |
if ($filter) { |
381 |
no strict 'refs'; |
no strict 'refs'; |
382 |
foreach my $d (&$filter($index_data)) { |
$idisplay = &$filter($idisplay); |
383 |
$index->insert($field, $d, $path); |
} |
384 |
|
push @index_data, $idisplay if ($idisplay && !$iterate_by_page); |
385 |
|
} |
386 |
|
|
387 |
|
# store fields in lookup |
388 |
|
if ($il && $display) { |
389 |
|
if (lc($x->{type}) eq "lookup_key") { |
390 |
|
if ($lookup_key) { |
391 |
|
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
392 |
|
} else { |
393 |
|
if ($filter) { |
394 |
|
no strict 'refs'; |
395 |
|
$lookup_key = &$filter($display); |
396 |
|
} else { |
397 |
|
$lookup_key = $display; |
398 |
|
} |
399 |
} |
} |
400 |
} else { |
} elsif (lc($x->{type}) eq "lookup_val") { |
401 |
$index->insert($field, $index_data, $path); |
if ($lookup_key) { |
402 |
|
if ($filter) { |
403 |
|
no strict 'refs'; |
404 |
|
$lhash{$lookup_key} = &$filter($display); |
405 |
|
} else { |
406 |
|
$lhash{$lookup_key} = $display; |
407 |
|
} |
408 |
|
} else { |
409 |
|
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
410 |
|
} |
411 |
|
} |
412 |
|
|
413 |
|
} |
414 |
|
|
415 |
|
# store data for page-by-page repeatable fields |
416 |
|
if ($iterate_by_page) { |
417 |
|
sub iterate_fld($$$$$$) { |
418 |
|
my ($cache,$what,$field,$page,$data,$append) = @_; |
419 |
|
return if (!$data); |
420 |
|
|
421 |
|
my $ldel = $page_line_delimiter; |
422 |
|
$ldel = " " if ($append); |
423 |
|
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
424 |
|
if (! $cache->{$what}->{$field}->[$page]) { |
425 |
|
push @{$cache->{$what}->{$field}->[$page]}, { |
426 |
|
data => $data, |
427 |
|
delimiter => $ldel, |
428 |
|
}; |
429 |
|
} |
430 |
|
} |
431 |
|
|
432 |
|
if ($display_data) { |
433 |
|
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
434 |
|
} |
435 |
|
$display_data = ""; |
436 |
|
if ($swish_data) { |
437 |
|
iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append}); |
438 |
|
$swish_data = ""; |
439 |
|
} |
440 |
|
if ($swish_exact_data) { |
441 |
|
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
442 |
|
$swish_exact_data = ""; |
443 |
|
} |
444 |
|
|
445 |
|
if ($idisplay) { |
446 |
|
my $ldel=$page_line_delimiter; |
447 |
|
my @index_data; |
448 |
|
if ($cache->{index_data}->{$field}->[$page]) { |
449 |
|
|
450 |
|
@index_data = @{$cache->{index_data}->{$field}->[$page]}; |
451 |
|
} |
452 |
|
if ($x->{append}) { |
453 |
|
if (@index_data) { |
454 |
|
$index_data[$#index_data] .= $idisplay; |
455 |
|
} else { |
456 |
|
push @index_data, $idisplay; |
457 |
|
} |
458 |
|
} else { |
459 |
|
push @index_data, $idisplay; |
460 |
|
} |
461 |
|
$idisplay = ""; |
462 |
|
@{$cache->{index_data}->{$field}->[$page]} = @index_data; |
463 |
} |
} |
464 |
} |
} |
465 |
} |
} |
466 |
|
|
467 |
|
if (! $iterate_by_page) { |
468 |
|
my $idel = $x->{index_delimiter}; |
469 |
|
# fill data in index |
470 |
|
foreach my $tmp (@index_data) { |
471 |
|
my $i = $d = $tmp; |
472 |
|
if ($idel && $tmp =~ m/$idel/) { |
473 |
|
($i,$d) = split(/$idel/,$tmp); |
474 |
|
} |
475 |
|
$index->insert($field, $i, $d, $path); |
476 |
|
} |
477 |
|
@index_data = (); |
478 |
|
} |
479 |
} |
} |
480 |
|
|
481 |
# now try to parse variables from configuration file |
# now try to parse variables from configuration file |
482 |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
483 |
|
|
484 |
|
$x = unroll_x($x); |
485 |
|
|
486 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
487 |
my $val = $cfg->val($database, x($x->{content})); |
my $val = $cfg->val($database, x($x->{content})); |
488 |
|
|
489 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
# FIXME index_lookup is not supported! |
490 |
$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"); |
|
491 |
|
|
492 |
if ($val) { |
if ($val) { |
493 |
$display_data .= $delimiter.$val if ($d); |
$display_data .= $delimiter.$val if ($d); |
494 |
$swish_data .= $val if ($s); |
$swish_data .= " ".$val if ($s); |
495 |
$index->insert($field, $val, $path) if ($i); |
$index->insert($field, $val, $val, $path) if ($i); |
496 |
} |
} |
497 |
|
|
498 |
|
if ($iterate_by_page) { |
499 |
|
# FIXME data from config tag will appear just |
500 |
|
# on first page!!! |
501 |
|
my $page = 0; |
502 |
|
if ($display_data) { |
503 |
|
push @{$cache->{display_data}->{$field}->[$page]}, { data => $display_data }; |
504 |
|
$display_data = ""; |
505 |
|
} |
506 |
|
if ($swish_data) { |
507 |
|
push @{$cache->{swish_data}->{$field}->[$page]}, { data => $swish_data }; |
508 |
|
$swish_data = ""; |
509 |
|
} |
510 |
|
if ($swish_exact_data) { |
511 |
|
push @{$cache->{swish_exact_data}->{$field}->[$page]}, { data => $swish_exact_data }; |
512 |
|
$swish_exact_data = ""; |
513 |
|
} |
514 |
|
} |
515 |
} |
} |
516 |
|
|
517 |
|
# save data page-by-page |
518 |
|
foreach my $field (@page_fields) { |
519 |
|
my $nr_pages = $page_max{$field} || next; |
520 |
|
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
521 |
|
#print STDERR Dumper($cache->{display_data}); |
522 |
|
my $seen; # used for distinct |
523 |
|
for (my $page=0; $page <= $nr_pages; $page++) { |
524 |
|
my $display_data; |
525 |
|
my $delimiter = ''; |
526 |
|
foreach my $element (@{ $cache->{display_data}->{$field}->[$page] }) { |
527 |
|
my $data = $element->{data}; |
528 |
|
die "BUG! no data in element?" unless ($data); |
529 |
|
|
530 |
|
if ($distinct) { |
531 |
|
next if ($cache->{distinct}->{$field}->{ $data }); |
532 |
|
$cache->{distinct}->{$field}->{ $data } = 1; |
533 |
|
} |
534 |
|
|
535 |
if ($display_data) { |
if ($cache->{format}->{$field}) { |
536 |
|
my $tmp = mkformat($cache->{format}->{$field},$data); |
537 |
|
$display_data .= $delimiter . $tmp if ($tmp); |
538 |
|
} else { |
539 |
|
$display_data .= $delimiter . $data; |
540 |
|
} |
541 |
|
$delimiter = $element->{delimiter} if ($element->{delimiter}); |
542 |
|
} |
543 |
|
|
544 |
if ($field eq "headline") { |
if ($display_data) { # default |
545 |
$xml .= xmlify("headline", $display_data); |
if ($field eq "headline") { |
546 |
} else { |
$xml .= xmlify("headline", $display_data); |
547 |
|
} else { |
548 |
|
|
549 |
# find field name (signular, plural) |
# fallback to empty field name if needed |
550 |
my $field_name = ""; |
$html .= get_field_name($config,$field,$field_usage{$field}) || ''; |
551 |
if ($config->{indexer}->{$field}->{name_singular} && $field_usage{$field} == 1) { |
$html .= "#-#".$display_data."###\n"; |
552 |
$field_name = $config->{indexer}->{$field}->{name_singular}."#-#"; |
} |
553 |
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
} |
554 |
$field_name = $config->{indexer}->{$field}->{name_plural}."#-#"; |
|
555 |
} elsif ($config->{indexer}->{$field}->{name}) { |
my $swish_data = join(" ",map { $_->{data} } @{ $cache->{swish_data}->{$field}->[$page] }); |
556 |
$field_name = $config->{indexer}->{$field}->{name}."#-#"; |
if ($swish_data) { |
557 |
} else { |
# remove extra spaces |
558 |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
$swish_data =~ s/ +/ /g; |
559 |
|
$swish_data =~ s/ +$//g; |
560 |
|
|
561 |
|
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
562 |
|
} |
563 |
|
|
564 |
|
my $swish_exact_data = join(" ", map { $_->{data} } @{ $cache->{swish_exact_data}->{$field}->[$page] }); |
565 |
|
if ($swish_exact_data) { |
566 |
|
$swish_exact_data =~ s/ +/ /g; |
567 |
|
$swish_exact_data =~ s/ +$//g; |
568 |
|
|
569 |
|
# add delimiters before and after word. |
570 |
|
# That is required to produce exact match |
571 |
|
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
572 |
} |
} |
573 |
if ($field_name) { |
|
574 |
$html .= x($field_name); |
my $idel = $cache->{index_delimiter}->{$field}; |
575 |
|
foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) { |
576 |
|
my $i = $tmp; |
577 |
|
my $d = $tmp; |
578 |
|
if ($idel && $tmp =~ m/$idel/) { |
579 |
|
($i,$d) = split(/$idel/,$tmp); |
580 |
|
} |
581 |
|
$index->insert($field, $i, $d, $path); |
582 |
|
#print STDERR "index [$idel] $field: $i --> $d [$path]\n"; |
583 |
} |
} |
|
$html .= $display_data."###\n"; |
|
584 |
} |
} |
585 |
|
|
586 |
} |
} |
587 |
if ($swish_data) { |
|
588 |
# remove extra spaces |
if (! $iterate_by_page) { |
589 |
$swish_data =~ s/ +/ /g; |
if ($display_data) { |
590 |
$swish_data =~ s/ +$//g; |
if ($field eq "headline") { |
591 |
|
$xml .= xmlify("headline", $display_data); |
592 |
|
} else { |
593 |
|
|
594 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
# fallback to empty field name if needed |
595 |
} |
$html .= get_field_name($config,$field,$field_usage{$field}) || ''; |
596 |
|
$html .= "#-#".$display_data."###\n"; |
597 |
|
} |
598 |
|
} |
599 |
|
if ($swish_data) { |
600 |
|
# remove extra spaces |
601 |
|
$swish_data =~ s/ +/ /g; |
602 |
|
$swish_data =~ s/ +$//g; |
603 |
|
|
604 |
|
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
605 |
|
} |
606 |
|
|
607 |
|
if ($swish_exact_data) { |
608 |
|
$swish_exact_data =~ s/ +/ /g; |
609 |
|
$swish_exact_data =~ s/ +$//g; |
610 |
|
|
611 |
|
# add delimiters before and after word. |
612 |
|
# That is required to produce exact match |
613 |
|
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
614 |
|
} |
615 |
|
} |
616 |
} |
} |
617 |
|
|
618 |
# dump formatted output in <html> |
# dump formatted output in <html> |
619 |
if ($html) { |
if ($html) { |
620 |
$xml .= xmlify("html",$html); |
#$xml .= xmlify("html",$html); |
621 |
|
$xml .= "<html><![CDATA[ $html ]]></html>"; |
622 |
} |
} |
623 |
|
|
624 |
if ($xml) { |
if ($xml) { |
645 |
$cfg_global->val('global', 'dbi_passwd') || '', |
$cfg_global->val('global', 'dbi_passwd') || '', |
646 |
); |
); |
647 |
|
|
648 |
|
my $show_progress = $cfg_global->val('global', 'show_progress'); |
649 |
|
|
650 |
|
my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter'); |
651 |
|
if ($my_unac_filter) { |
652 |
|
print STDERR "using $my_unac_filter to filter characters for search\n"; |
653 |
|
require $my_unac_filter; |
654 |
|
} else { |
655 |
|
print STDERR "### fallback to default my_unac_string!\n"; |
656 |
|
eval q{ |
657 |
|
sub main::my_unac_string($$) { |
658 |
|
my ($charset, $string) = (@_); |
659 |
|
return $string; |
660 |
|
} |
661 |
|
}; |
662 |
|
} |
663 |
|
|
664 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
665 |
|
|
666 |
|
# save database name in global variable path for later |
667 |
|
# (need for index filter creation) |
668 |
|
$path = $database; |
669 |
|
|
670 |
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"; |
671 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
672 |
|
|
673 |
print STDERR "reading ./import_xml/$type.xml\n"; |
# create new lookup file |
674 |
|
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
675 |
|
if ($lookup_file) { |
676 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
677 |
|
if (! -e $lookup_file) { |
678 |
|
open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!"; |
679 |
|
close(LOOKUP); |
680 |
|
} |
681 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
682 |
|
print STDERR "creating lookup file '$lookup_file'\n"; |
683 |
|
# delete memory cache for lookup file |
684 |
|
delete $cache->{lhash}; |
685 |
|
} |
686 |
|
|
687 |
|
# open existing lookup file |
688 |
|
$lookup_file = $cfg -> val($database, 'lookup_open'); # optional |
689 |
|
if ($lookup_file) { |
690 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644; |
691 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644; |
692 |
|
print STDERR "opening lookup file '$lookup_file'\n"; |
693 |
|
} |
694 |
|
|
695 |
|
my $import_xml_type = $cfg->val($database, 'import_xml_file') || $type; |
696 |
|
my $import_xml_file = "./import_xml/$import_xml_type.xml"; |
697 |
|
|
698 |
|
if (! -r $import_xml_file) { |
699 |
|
print STDERR "ERROR: file $import_xml_file not readable skipping!\n"; |
700 |
|
next; |
701 |
|
} |
702 |
|
|
703 |
|
print STDERR "reading $import_xml_file\n"; |
704 |
|
|
705 |
# extract just type basic |
# extract just type basic |
706 |
my $type_base = $type; |
my $type_base = $type; |
707 |
$type_base =~ s/_.+$//g; |
$type_base =~ s/_.+$//g; |
708 |
|
|
709 |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config' ], forcecontent => 1); |
my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type_base} || die "can't find which tag to use for type $type"; |
710 |
|
$config=XMLin($import_xml_file, ForceArray => [ $tag, 'config', 'format' ], ForceContent => 1 ); |
711 |
|
|
712 |
|
# check for broken XML::Simple |
713 |
|
if ( $broken_cdata ) { |
714 |
|
map { |
715 |
|
$config->{format}->{$_}->{content} =~ s#>$##; |
716 |
|
} keys %{ $config->{format} }; |
717 |
|
} |
718 |
|
|
719 |
|
# helper for progress bar |
720 |
|
sub fmt_time { |
721 |
|
my $t = shift || 0; |
722 |
|
my $out = ""; |
723 |
|
|
724 |
|
my ($ss,$mm,$hh) = gmtime($t); |
725 |
|
$out .= "${hh}h" if ($hh); |
726 |
|
$out .= sprintf("%02d:%02d", $mm,$ss); |
727 |
|
$out .= " " if ($hh == 0); |
728 |
|
return $out; |
729 |
|
} |
730 |
|
|
731 |
# output current progress indicator |
# output current progress indicator |
732 |
my $last_p = 0; |
my $last_p = 0; |
733 |
|
my $start_t = time(); |
734 |
sub progress { |
sub progress { |
735 |
#return if (! $opts{q}); # FIXME |
return if (! $show_progress); |
736 |
my $current = shift; |
my $current = shift; |
737 |
my $total = shift || 1; |
my $total = shift || 1; |
738 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
739 |
if ($p != $last_p) { |
if ($p < $last_p || $current == 1) { |
740 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
$start_t = time(); |
741 |
|
$last_p = 0; |
742 |
|
} elsif ($p != $last_p) { |
743 |
|
my $rate = ($current / (time() - $start_t || 1)); |
744 |
|
my $eta = ($total-$current) / ($rate || 1); |
745 |
|
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta)); |
746 |
$last_p = $p; |
$last_p = $p; |
747 |
} |
} |
748 |
} |
} |
749 |
|
|
750 |
|
my $fake_dir = 1; |
751 |
|
my $fake_pos = 0; |
752 |
|
my $last_fake_t = time(); |
753 |
|
sub fakeprogress { |
754 |
|
return if (! $show_progress); |
755 |
|
my $current = shift @_; |
756 |
|
|
757 |
|
my @ind = ('-','\\','|','/','-','\\','|','/'); |
758 |
|
|
759 |
|
if ($current < $fake_pos) { |
760 |
|
$start_t = time(); |
761 |
|
$last_fake_t = 0; |
762 |
|
$fake_dir = 1; |
763 |
|
$fake_pos = 0; |
764 |
|
} |
765 |
|
|
766 |
|
if (time()-$last_fake_t >= 1) { |
767 |
|
$last_fake_t = time(); |
768 |
|
$fake_pos += $fake_dir; |
769 |
|
$fake_dir = -$fake_dir if ($fake_pos > 38); |
770 |
|
} |
771 |
|
|
772 |
|
if ($current % 10 == 0) { |
773 |
|
my $rate = ($current / (time() - $start_t || 1)); |
774 |
|
printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate); |
775 |
|
} |
776 |
|
} |
777 |
|
|
778 |
# now read database |
# now read database |
779 |
print STDERR "using: $type...\n"; |
|
780 |
|
# erase cache for tags by order in this database |
781 |
|
delete $cache->{tags_by_order}; |
782 |
|
|
783 |
if ($type_base eq "isis") { |
if ($type_base eq "isis") { |
784 |
|
|
785 |
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!"; |
786 |
|
|
787 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
788 |
my $db = OpenIsis::open( $isis_db ); |
my $db = new Biblio::Isis( isisdb => $isis_db ); |
789 |
|
|
790 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
if (! $db) { |
791 |
|
print STDERR "FATAL: can't read ISIS database: $isis_db, skipping...\n"; |
792 |
|
next; |
793 |
|
} |
794 |
|
|
795 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
my $max_rowid = $db->count if ($db); |
796 |
|
|
797 |
my $path = $database; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
798 |
|
|
799 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
800 |
my $row = OpenIsis::read( $db, $row_id ); |
my $row = $db->to_hash( $row_id ); |
801 |
if ($row && $row->{mfn}) { |
if ($row) { |
802 |
|
|
803 |
|
$row->{mfn} = $row_id; |
804 |
|
$row->{record} = $db->{record}; |
805 |
|
|
806 |
progress($row->{mfn}, $max_rowid); |
progress($row->{mfn}, $max_rowid); |
807 |
|
|
808 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
819 |
print STDERR "\n"; |
print STDERR "\n"; |
820 |
|
|
821 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
822 |
use Spreadsheet::ParseExcel; |
require Spreadsheet::ParseExcel; |
823 |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
require Spreadsheet::ParseExcel::Utility; |
824 |
|
import Spreadsheet::ParseExcel::Utility qw(int2col); |
825 |
|
|
826 |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
827 |
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!"; |
828 |
|
|
829 |
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml"; |
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml"; |
830 |
my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml"; |
my $start_row = x($config->{start_row}) || die "no start_row in $type.xml"; |
831 |
|
$start_row--; |
832 |
|
|
833 |
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'"; |
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'"; |
834 |
|
|
850 |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
851 |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
852 |
if ($cell) { |
if ($cell) { |
853 |
$row->{int2col($iC)} = $cell->Value; |
# this conversion is a cludge. |
854 |
|
# Files from Excell could have |
855 |
|
# characters which don't fit into |
856 |
|
# destination encoding. |
857 |
|
$row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value; |
858 |
} |
} |
859 |
} |
} |
860 |
|
|
878 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
879 |
} |
} |
880 |
} |
} |
881 |
|
|
882 |
|
print STDERR "\n"; |
883 |
|
|
884 |
} elsif ($type_base eq "marc") { |
} elsif ($type_base eq "marc") { |
885 |
## XXX |
|
886 |
use MARC; |
require MARC::File::USMARC; |
887 |
|
|
888 |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
889 |
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!"; |
890 |
|
|
891 |
# optional argument is format |
# optional argument is format |
892 |
my $format = x($config->{format}) || 'usmarc'; |
warn "marc_format is no longer used!" if ($config->{marc_format}); |
893 |
|
print STDERR "Reading MARC file '$marc_file'\n"; |
894 |
|
|
895 |
my %id_stored; # to aviod duplicates |
my $marc = MARC::File::USMARC->in( $marc_file ); |
896 |
|
|
897 |
print STDERR "Reading MARC file '$marc_file'\n"; |
if (! $marc) { |
898 |
|
print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n"; |
899 |
|
next; |
900 |
|
} |
901 |
|
|
902 |
|
# count records in MARC file |
903 |
|
sub marc_count { |
904 |
|
my $filename = shift || die; |
905 |
|
my $file = MARC::File::USMARC->in($filename) || return; |
906 |
|
my $count = 0; |
907 |
|
while ($file->skip()) { |
908 |
|
$count++; |
909 |
|
} |
910 |
|
return $count; |
911 |
|
} |
912 |
|
|
913 |
|
my $count = marc_count($marc_file) || warn "no records in '$marc_file'?"; |
914 |
|
|
915 |
my $marc = new MARC; |
my $i = 1; |
916 |
my $nr = $marc->openmarc({ |
|
917 |
file=>$marc_file, format=>$format |
while( my $rec = $marc->next() ) { |
918 |
}) || die "Can't open MARC file '$marc_file'"; |
|
919 |
|
progress($i,$count); |
|
my $i=0; # record nr. |
|
|
my $inc=1; |
|
|
my $max_i=1000; |
|
|
|
|
|
my $rec; |
|
|
|
|
|
while ($marc->nextmarc(1)) { |
|
|
|
|
|
# XXX |
|
|
progress($i, $max_i); |
|
|
$i += $inc; |
|
|
$inc = -$inc if ($i > $max_i || $i < 0); |
|
920 |
|
|
921 |
my $swishpath = $database."#".$i; |
my $swishpath = $database."#".$i; |
922 |
|
|
923 |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) { |
924 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
925 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
926 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
927 |
print "Content-Length: ".(length($xml)+1)."\n"; |
print "Content-Length: ".(length($xml)+1)."\n"; |
928 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
929 |
} |
} |
930 |
|
|
931 |
|
$i++; |
932 |
|
} |
933 |
|
|
934 |
|
print STDERR "\n"; |
935 |
|
|
936 |
|
} elsif ($type_base eq "feed") { |
937 |
|
|
938 |
|
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
939 |
|
my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!"; |
940 |
|
|
941 |
|
print STDERR "Reading feed from program '$prog'\n"; |
942 |
|
|
943 |
|
open(FEED,"feeds/$prog |") || die "can't start $prog: $!"; |
944 |
|
|
945 |
|
my $i=1; # record nr. |
946 |
|
|
947 |
|
my $data; |
948 |
|
my $line=1; |
949 |
|
|
950 |
|
while (<FEED>) { |
951 |
|
chomp; |
952 |
|
|
953 |
|
if (/^$/) { |
954 |
|
my $swishpath = $database."#".$i++; |
955 |
|
|
956 |
|
if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) { |
957 |
|
$xml = $cp2utf->convert($xml); |
958 |
|
use bytes; # as opposed to chars |
959 |
|
print "Path-Name: $swishpath\n"; |
960 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
961 |
|
print "Document-Type: XML\n\n$xml\n"; |
962 |
|
} |
963 |
|
$line = 1; |
964 |
|
$data = {}; |
965 |
|
next; |
966 |
|
} |
967 |
|
|
968 |
|
$line = $1 if (s/^(\d+):\s*//); |
969 |
|
$data->{$line++} = $_; |
970 |
|
|
971 |
|
fakeprogress($i); |
972 |
|
|
973 |
|
} |
974 |
|
# close lookup |
975 |
|
untie %lhash if (%lhash); |
976 |
|
|
977 |
|
} elsif ($type_base eq "dbf") { |
978 |
|
|
979 |
|
my $dbf_file = $cfg -> val($database, 'dbf_file') || die "$database doesn't have 'dbf_file' defined!"; |
980 |
|
my $dbf_codepage = $cfg -> val($database, 'dbf_codepage') || die "$database doesn't have 'dbf_codepage' defined!"; |
981 |
|
my $dbf_mapping = $cfg -> val($database, 'dbf_mapping') || die "$database doesn't have 'dbf_mapping' defined!"; |
982 |
|
|
983 |
|
$import2cp = Text::Iconv->new($dbf_codepage,$codepage); |
984 |
|
require XBase; |
985 |
|
my $db = new XBase $dbf_file; |
986 |
|
|
987 |
|
if (! $db) { |
988 |
|
print STDERR "ERROR: can't read DBF database: $dbf_file, skipping...\n"; |
989 |
|
next; |
990 |
|
} |
991 |
|
|
992 |
|
my $max_rowid = $db->last_record; |
993 |
|
|
994 |
|
print STDERR "Reading database: $dbf_file [$max_rowid rows]\n"; |
995 |
|
|
996 |
|
my %dbf2iso; |
997 |
|
foreach my $m (split(/[\n\r]+/,$dbf_mapping)) { |
998 |
|
my ($col,$fld) = split(/\s+/,$m,2); |
999 |
|
$dbf2iso{$col} = $fld; |
1000 |
|
} |
1001 |
|
|
1002 |
|
#print STDERR "## dbf2iso: ",Dumper(\%dbf2iso),"\n## /dbf2iso\n"; |
1003 |
|
|
1004 |
|
# bad, bad... |
1005 |
|
require "to_hash.pm"; |
1006 |
|
|
1007 |
|
foreach my $row_id (0 .. $max_rowid) { |
1008 |
|
my $dbf_row = $db->get_record_as_hash($row_id); |
1009 |
|
if ($dbf_row) { |
1010 |
|
|
1011 |
|
#print STDERR "## dbf_row: ",Dumper($dbf_row),"\n## /dbf_row\n"; |
1012 |
|
# apply mapping from config file |
1013 |
|
# all unspecified records will get _ in |
1014 |
|
# front of them - _DELETE will be __DELETE |
1015 |
|
my $rec; |
1016 |
|
map { |
1017 |
|
my $new_fld = $dbf2iso{$_} || '_'.$_; |
1018 |
|
my $data = $dbf_row->{$_}; |
1019 |
|
push @{ $rec->{$new_fld} }, $data if ($data && $data !~ /^(?:\s+|\$a\.|)$/); |
1020 |
|
} keys %{$dbf_row}; |
1021 |
|
#print STDERR "## rec: ",Dumper($rec),"\n## /rec\n"; |
1022 |
|
my $row = to_hash($row_id+1, $rec); |
1023 |
|
|
1024 |
|
$row->{mfn} = $row_id+1; |
1025 |
|
$row->{record} = $rec; |
1026 |
|
|
1027 |
|
#print STDERR "## row: ",Dumper($row),"\n## /row\n"; |
1028 |
|
progress($row->{mfn}, $max_rowid); |
1029 |
|
|
1030 |
|
my $swishpath = $path."#".int($row->{mfn}); |
1031 |
|
|
1032 |
|
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
1033 |
|
$xml = $cp2utf->convert($xml); |
1034 |
|
use bytes; # as opposed to chars |
1035 |
|
print "Path-Name: $swishpath\n"; |
1036 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
1037 |
|
print "Document-Type: XML\n\n$xml\n"; |
1038 |
|
} |
1039 |
|
} |
1040 |
} |
} |
1041 |
|
print STDERR "\n"; |
1042 |
} |
} |
1043 |
} |
} |
1044 |
|
|
1051 |
|
|
1052 |
=head1 NAME |
=head1 NAME |
1053 |
|
|
1054 |
isis2xml.pl - read isis file and dump XML |
all2xml.pl - read various file formats and dump XML for SWISH-E |
1055 |
|
|
1056 |
|
=head1 SYNOPSYS |
1057 |
|
|
1058 |
|
$ all2xml.pl [test.conf] |
1059 |
|
|
1060 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
1061 |
|
|
1062 |
This command will read ISIS data file using OpenIsis perl module and |
This command will read ISIS data file using Biblio::Isis perl module, MARC |
1063 |
create XML file for usage with I<SWISH-E> |
records using MARC::File module and optionally Micro$oft Excel files to |
1064 |
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, |
1065 |
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 |
1066 |
script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |
1067 |
|
|
1068 |
|
If no configuration file is specified, it will use default one called |
1069 |
|
C<all2xml.conf>. |
1070 |
|
|
1071 |
|
=head1 BUGS |
1072 |
|
|
1073 |
|
Documentation is really lacking. However, in true Open Source spirit, source |
1074 |
|
is best documentation. I even made considerable effort to comment parts |
1075 |
|
which are not intuitively clear, so... |
1076 |
|
|
1077 |
=head1 AUTHOR |
=head1 AUTHOR |
1078 |
|
|