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; |
#use GDBM_File; |
12 |
use Fcntl; # for O_RDWR |
use Fcntl; # for O_RDWR |
13 |
use TDB_File; |
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' |
'feed' => 'feed', |
67 |
|
'dbf' => 'isis', # special case, re-use isis import_xml |
68 |
); |
); |
69 |
|
|
70 |
my $cache; # for cacheing |
my $cache; # for cacheing |
75 |
# if you are tight on memory, turn this off |
# if you are tight on memory, turn this off |
76 |
my $use_lhash_cache = 1; |
my $use_lhash_cache = 1; |
77 |
|
|
78 |
|
my $last_field_name; # cache to prevent repeated fields |
79 |
|
|
80 |
sub data2xml { |
sub data2xml { |
81 |
|
|
82 |
use xmlify; |
use xmlify; |
107 |
} |
} |
108 |
|
|
109 |
my @sorted_tags; |
my @sorted_tags; |
110 |
if ($cache->{tags_by_order}->{$type}) { |
if ($cache->{tags_by_order}) { |
111 |
@sorted_tags = @{$cache->{tags_by_order}->{$type}}; |
@sorted_tags = @{$cache->{tags_by_order}}; |
112 |
} else { |
} else { |
113 |
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
114 |
$cache->{tags_by_order}->{$type} = \@sorted_tags; |
$cache->{tags_by_order} = \@sorted_tags; |
115 |
|
} |
116 |
|
|
117 |
|
if (! @sorted_tags) { |
118 |
|
print STDERR "WARNING: no tags for this type found in import_xml file!\n"; |
119 |
} |
} |
120 |
|
|
121 |
# lookup key |
# lookup key |
122 |
my $lookup_key; |
my $lookup_key; |
123 |
|
|
124 |
# cache for field in pages |
# cache for field in pages |
125 |
$cache->{display_data} = (); |
delete $cache->{display_data}; |
126 |
$cache->{swish_data} = (); |
delete $cache->{swish_data}; |
127 |
$cache->{swish_exact_data} = (); |
delete $cache->{swish_exact_data}; |
128 |
|
delete $cache->{index_data}; |
129 |
|
delete $cache->{index_delimiter}; |
130 |
my @page_fields; # names of fields |
my @page_fields; # names of fields |
131 |
|
|
132 |
|
|
146 |
} else { |
} else { |
147 |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
148 |
} |
} |
149 |
|
|
150 |
if ($field_name) { |
if ($field_name) { |
151 |
return x($field_name); |
$field_name = x($field_name); |
152 |
|
if (! $last_field_name) { |
153 |
|
$last_field_name = $field_name; |
154 |
|
return $last_field_name; |
155 |
|
} elsif ($field_name ne $last_field_name) { |
156 |
|
$last_field_name = $field_name; |
157 |
|
return $last_field_name; |
158 |
|
} |
159 |
} |
} |
160 |
} |
} |
161 |
|
|
162 |
|
|
163 |
|
# init variables for different types |
164 |
|
sub init_visible_type($) { |
165 |
|
my $type = shift; |
166 |
|
|
167 |
|
# swish, swish_exact, display, index, index_lookup |
168 |
|
# swish and display defaults |
169 |
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
170 |
|
if (lc($type) eq "display") { |
171 |
|
$s = 0; |
172 |
|
} elsif (lc($type) eq "swish") { |
173 |
|
$d = 0; |
174 |
|
} elsif (lc($type) eq "index") { |
175 |
|
($s,$se,$d,$i) = (0,1,0,1); |
176 |
|
} elsif (lc($type) eq "swish_exact") { |
177 |
|
($s,$se,$d,$i) = (0,1,0,0); |
178 |
|
} elsif (lc($type) =~ /^lookup/) { |
179 |
|
($s,$se,$d,$i,$il) = (0,1,0,0,1); |
180 |
|
} elsif ($type) { |
181 |
|
print STDERR "WARNING: unknown type: $type\n"; |
182 |
|
} |
183 |
|
return ($s,$se,$d,$i,$il); |
184 |
|
} |
185 |
|
|
186 |
|
|
187 |
|
# convert |
188 |
|
# |
189 |
|
# <tag> |
190 |
|
# <delimiter>, </delimiter> |
191 |
|
# <value>200a</value> |
192 |
|
# </tag> |
193 |
|
# |
194 |
|
# to |
195 |
|
# |
196 |
|
# <tag delimiter=", ">200a</tag> |
197 |
|
# |
198 |
|
# but without loosing spaces in delimiter (becasue |
199 |
|
# new XML::Simple strips spaces in attribute values |
200 |
|
# as defined in XML specification) |
201 |
|
# |
202 |
|
sub unroll_x($) { |
203 |
|
my $x = shift; |
204 |
|
|
205 |
|
if (defined $x->{value}) { |
206 |
|
my ($v,$d) = ($x->{value}->{content}, $x->{delimiter}->{content}); |
207 |
|
delete $x->{value}; |
208 |
|
delete $x->{delimiter}; |
209 |
|
$x->{content} = $v; |
210 |
|
$x->{delimiter} = $d; |
211 |
|
} |
212 |
|
return $x; |
213 |
|
} |
214 |
|
|
215 |
# begin real work: go field by field |
# begin real work: go field by field |
216 |
foreach my $field (@sorted_tags) { |
foreach my $field (@sorted_tags) { |
217 |
|
|
221 |
my $swish_data = ""; |
my $swish_data = ""; |
222 |
my $swish_exact_data = ""; |
my $swish_exact_data = ""; |
223 |
my $display_data = ""; |
my $display_data = ""; |
224 |
|
my @index_data; |
225 |
my $line_delimiter; |
my $line_delimiter; |
226 |
|
|
227 |
my ($swish,$display); |
my ($swish,$display); |
234 |
my %page_max = (); |
my %page_max = (); |
235 |
# default line_delimiter if using |
# default line_delimiter if using |
236 |
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
237 |
|
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter}; |
238 |
|
|
239 |
|
my $format_name = $config->{indexer}->{$field}->{format_name}; |
240 |
|
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter}; |
241 |
|
if ($format_name && $format_delimiter) { |
242 |
|
$cache->{format}->{$field}->{format_name} = $format_name; |
243 |
|
$cache->{format}->{$field}->{format_delimiter} = $format_delimiter; |
244 |
|
} |
245 |
|
|
246 |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
247 |
|
|
248 |
|
$x = unroll_x($x); |
249 |
|
|
250 |
my $format = x($x->{content}); |
my $format = x($x->{content}); |
251 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
252 |
|
|
253 |
my $repeat_off = 0; # init repeatable offset |
my $repeat_off = 0; # init repeatable offset |
254 |
|
|
255 |
# swish, swish_exact, display, index, index_lookup |
my ($s,$se,$d,$i,$il) = init_visible_type($x->{type}); |
|
# swish and display defaults |
|
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
|
|
$s = 0 if (lc($x->{type}) eq "display"); |
|
|
$d = 0 if (lc($x->{type}) eq "swish"); |
|
|
$se = 1 if (lc($x->{type}) eq "swish_exact"); |
|
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
|
|
$il = 1 if (lc($x->{type}) =~ /^lookup/); |
|
|
|
|
256 |
|
|
257 |
# what will separate last line from this one? |
# what will separate last line from this one? |
258 |
if ($display_data && $x->{append} && $x->{append} eq "1") { |
if ($display_data && $x->{append}) { |
259 |
$line_delimiter = ' '; |
$line_delimiter = $delimiter; |
260 |
} elsif ($display_data) { |
} elsif ($display_data) { |
261 |
$line_delimiter = '<br/>'; |
$line_delimiter = '<br/>'; |
262 |
} |
} |
264 |
# init vars so that we go into while... |
# init vars so that we go into while... |
265 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
266 |
|
|
267 |
# placeholder for all repeatable entries for index |
sub mkformat($$) { |
|
my @index_data; |
|
|
|
|
|
sub mkformat { |
|
268 |
my $x = shift || die "mkformat needs tag reference"; |
my $x = shift || die "mkformat needs tag reference"; |
269 |
my $data = shift || return; |
my $data = shift || return; |
270 |
my $format_name = x($x->{format_name}) || return $data; |
my $format_name = x($x->{format_name}) || return $data; |
282 |
if (($#data+1) == $nr) { |
if (($#data+1) == $nr) { |
283 |
return sprintf($fmt,@data); |
return sprintf($fmt,@data); |
284 |
} else { |
} else { |
285 |
print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
#print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
286 |
return $data; |
return $data; |
287 |
} |
} |
288 |
} else { |
} else { |
311 |
$display = $new_display; |
$display = $new_display; |
312 |
$cache->{lhash}->{$display} = $new_display; |
$cache->{lhash}->{$display} = $new_display; |
313 |
} else { |
} else { |
314 |
print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
# print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
315 |
$display = ""; |
$display = ""; |
316 |
$cache->{lhash}->{$display} = $null; |
$cache->{lhash}->{$display} = $null; |
317 |
} |
} |
332 |
} |
} |
333 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
334 |
if ($swish) { |
if ($swish) { |
335 |
|
my $tmp = $swish; |
336 |
if ($filter && ($s || $se)) { |
if ($filter && ($s || $se)) { |
337 |
no strict 'refs'; |
no strict 'refs'; |
338 |
my $tmp = join(" ",&$filter($swish)) if ($s || $se); |
$tmp = join(" ",&$filter($tmp)) if ($s || $se); |
|
$swish_data .= $tmp if ($s); |
|
|
$swish_exact_data .= $tmp if ($se); |
|
|
|
|
|
} else { |
|
|
$swish_data .= $swish if ($s); |
|
|
$swish_exact_data .= $swish if ($se); |
|
339 |
} |
} |
340 |
|
|
341 |
|
$swish_data .= $tmp if ($s && $tmp); |
342 |
|
$swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se); |
343 |
} |
} |
344 |
|
|
345 |
# type="display" ; field for display |
# type="display" ; field for display |
346 |
if ($d && $display) { |
if ($d && $display) { |
347 |
|
my $ldel = $delimiter; |
348 |
if ($line_delimiter && $display_data) { |
if ($line_delimiter && $display_data) { |
349 |
$display_data .= $line_delimiter; |
$ldel = $line_delimiter; |
350 |
} |
} |
351 |
if ($filter) { |
if ($filter) { |
352 |
no strict 'refs'; |
no strict 'refs'; |
353 |
if ($display_data) { |
my @arr; |
354 |
$display_data .= $delimiter.join($delimiter,mkformat($x,&$filter($display))); |
foreach my $tmp (&$filter($display)) { |
355 |
} else { |
my $tmp2 = mkformat($x,$tmp); |
356 |
$display_data = join($delimiter,mkformat($x,&$filter($display))); |
push @arr,$tmp2 if ($tmp2); |
357 |
} |
} |
358 |
|
$display_data .= $ldel if ($display_data && @arr); |
359 |
|
$display_data .= join($delimiter,@arr); |
360 |
} else { |
} else { |
361 |
if ($display_data) { |
$display_data .= $ldel if ($display_data); |
362 |
$display_data .= $delimiter.mkformat($x,$display); |
my $tmp = mkformat($x,$display); |
363 |
} else { |
$display_data .= $tmp if ($tmp); |
|
$display_data = mkformat($x,$display); |
|
|
} |
|
364 |
} |
} |
365 |
} |
} |
366 |
|
|
367 |
# type="index" ; insert into index |
# type="index" ; insert into index |
368 |
|
my $idisplay; |
369 |
if ($i && $display) { |
if ($i && $display) { |
370 |
|
$idisplay = $display; |
371 |
if ($filter) { |
if ($filter) { |
372 |
no strict 'refs'; |
no strict 'refs'; |
373 |
$display = &$filter($display); |
$idisplay = &$filter($idisplay); |
|
} |
|
|
if ($x->{append} && @index_data) { |
|
|
$index_data[$#index_data].=$display; |
|
|
} else { |
|
|
push @index_data, $display; |
|
374 |
} |
} |
375 |
|
push @index_data, $idisplay if ($idisplay && !$iterate_by_page); |
376 |
} |
} |
377 |
|
|
378 |
# store fields in lookup |
# store fields in lookup |
381 |
if ($lookup_key) { |
if ($lookup_key) { |
382 |
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
383 |
} else { |
} else { |
384 |
$lookup_key = $display; |
if ($filter) { |
385 |
|
no strict 'refs'; |
386 |
|
$lookup_key = &$filter($display); |
387 |
|
} else { |
388 |
|
$lookup_key = $display; |
389 |
|
} |
390 |
} |
} |
391 |
} elsif (lc($x->{type}) eq "lookup_val") { |
} elsif (lc($x->{type}) eq "lookup_val") { |
392 |
if ($lookup_key) { |
if ($lookup_key) { |
393 |
$lhash{$lookup_key} = $display; |
if ($filter) { |
394 |
|
no strict 'refs'; |
395 |
|
$lhash{$lookup_key} = &$filter($display); |
396 |
|
} else { |
397 |
|
$lhash{$lookup_key} = $display; |
398 |
|
} |
399 |
} else { |
} else { |
400 |
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
401 |
} |
} |
408 |
sub iterate_fld($$$$$$) { |
sub iterate_fld($$$$$$) { |
409 |
my ($cache,$what,$field,$page,$data,$append) = @_; |
my ($cache,$what,$field,$page,$data,$append) = @_; |
410 |
return if (!$data); |
return if (!$data); |
411 |
my $line_delimiter = $page_line_delimiter; |
|
412 |
$line_delimiter = " " if ($append); |
my $ldel = $page_line_delimiter; |
413 |
|
$ldel = " " if ($append); |
414 |
|
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
415 |
if (! $cache->{$what}->{$field}->[$page]) { |
if (! $cache->{$what}->{$field}->[$page]) { |
416 |
$cache->{$what}->{$field}->[$page] = $data; |
$cache->{$what}->{$field}->[$page] = $data; |
417 |
} else { |
} else { |
418 |
$cache->{$what}->{$field}->[$page] .= $line_delimiter.$data; |
$cache->{$what}->{$field}->[$page] .= $ldel.$data; |
419 |
} |
} |
420 |
} |
} |
421 |
|
|
422 |
if ($display_data) { |
if ($display_data) { |
|
print STDERR "line delimiter: ",Dumper($line_delimiter) if ($line_delimiter); |
|
423 |
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append}); |
424 |
} |
} |
425 |
$display_data = ""; |
$display_data = ""; |
431 |
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append}); |
432 |
$swish_exact_data = ""; |
$swish_exact_data = ""; |
433 |
} |
} |
434 |
|
|
435 |
|
if ($idisplay) { |
436 |
|
my $ldel=$page_line_delimiter; |
437 |
|
my @index_data; |
438 |
|
if ($cache->{index_data}->{$field}->[$page]) { |
439 |
|
|
440 |
|
@index_data = @{$cache->{index_data}->{$field}->[$page]}; |
441 |
|
} |
442 |
|
if ($x->{append}) { |
443 |
|
if (@index_data) { |
444 |
|
$index_data[$#index_data] .= $idisplay; |
445 |
|
} else { |
446 |
|
push @index_data, $idisplay; |
447 |
|
} |
448 |
|
} else { |
449 |
|
push @index_data, $idisplay; |
450 |
|
} |
451 |
|
$idisplay = ""; |
452 |
|
@{$cache->{index_data}->{$field}->[$page]} = @index_data; |
453 |
|
} |
454 |
} |
} |
455 |
} |
} |
456 |
|
|
457 |
# fill data in index |
if (! $iterate_by_page) { |
458 |
foreach my $d (@index_data) { |
my $idel = $x->{index_delimiter}; |
459 |
$index->insert($field, $d, $path); |
# fill data in index |
460 |
|
foreach my $tmp (@index_data) { |
461 |
|
my $i = $d = $tmp; |
462 |
|
if ($idel && $tmp =~ m/$idel/) { |
463 |
|
($i,$d) = split(/$idel/,$tmp); |
464 |
|
} |
465 |
|
$index->insert($field, $i, $d, $path); |
466 |
|
} |
467 |
|
@index_data = (); |
468 |
} |
} |
469 |
} |
} |
470 |
|
|
471 |
# now try to parse variables from configuration file |
# now try to parse variables from configuration file |
472 |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
473 |
|
|
474 |
|
$x = unroll_x($x); |
475 |
|
|
476 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
477 |
my $val = $cfg->val($database, x($x->{content})); |
my $val = $cfg->val($database, x($x->{content})); |
478 |
|
|
479 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
# FIXME index_lookup is not supported! |
480 |
$s = 0 if (lc($x->{type}) eq "display"); |
my ($s,$se,$d,$i,$il) = init_visible_type($x->{type}); |
|
$d = 0 if (lc($x->{type}) eq "swish"); |
|
|
# no support for swish exact in config. |
|
|
# IMHO, it's useless |
|
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
|
481 |
|
|
482 |
if ($val) { |
if ($val) { |
483 |
$display_data .= $delimiter.$val if ($d); |
$display_data .= $delimiter.$val if ($d); |
484 |
$swish_data .= $val if ($s); |
$swish_data .= " ".$val if ($s); |
485 |
$index->insert($field, $val, $path) if ($i); |
$index->insert($field, $val, $path) if ($i); |
486 |
} |
} |
487 |
|
|
510 |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
511 |
#print STDERR Dumper($cache->{display_data}); |
#print STDERR Dumper($cache->{display_data}); |
512 |
for (my $page=0; $page <= $nr_pages; $page++) { |
for (my $page=0; $page <= $nr_pages; $page++) { |
513 |
|
my $display_data; |
514 |
my $display_data = $cache->{display_data}->{$field}->[$page]; |
if ($cache->{format}->{$field}) { |
515 |
|
my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]); |
516 |
|
$display_data=$tmp if ($tmp); |
517 |
|
} else { |
518 |
|
$display_data = $cache->{display_data}->{$field}->[$page]; |
519 |
|
} |
520 |
if ($display_data) { # default |
if ($display_data) { # default |
521 |
if ($field eq "headline") { |
if ($field eq "headline") { |
522 |
$xml .= xmlify("headline", $display_data); |
$xml .= xmlify("headline", $display_data); |
534 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
535 |
$swish_data =~ s/ +$//g; |
$swish_data =~ s/ +$//g; |
536 |
|
|
537 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
538 |
} |
} |
539 |
|
|
540 |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
544 |
|
|
545 |
# add delimiters before and after word. |
# add delimiters before and after word. |
546 |
# That is required to produce exact match |
# That is required to produce exact match |
547 |
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
548 |
|
} |
549 |
|
|
550 |
|
my $idel = $cache->{index_delimiter}->{$field}; |
551 |
|
foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) { |
552 |
|
my $i = $tmp; |
553 |
|
my $d = $tmp; |
554 |
|
if ($idel && $tmp =~ m/$idel/) { |
555 |
|
($i,$d) = split(/$idel/,$tmp); |
556 |
|
} |
557 |
|
$index->insert($field, $i, $d, $path); |
558 |
|
#print STDERR "index [$idel] $field: $i --> $d [$path]\n"; |
559 |
} |
} |
560 |
} |
} |
561 |
|
|
577 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
578 |
$swish_data =~ s/ +$//g; |
$swish_data =~ s/ +$//g; |
579 |
|
|
580 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
581 |
} |
} |
582 |
|
|
583 |
if ($swish_exact_data) { |
if ($swish_exact_data) { |
586 |
|
|
587 |
# add delimiters before and after word. |
# add delimiters before and after word. |
588 |
# That is required to produce exact match |
# That is required to produce exact match |
589 |
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
590 |
} |
} |
591 |
} |
} |
592 |
} |
} |
623 |
|
|
624 |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
625 |
|
|
626 |
my $unac_filter = $cfg_global->val('global', 'unac_filter'); |
my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter'); |
627 |
if ($unac_filter) { |
if ($my_unac_filter) { |
628 |
require $unac_filter; |
print STDERR "using $my_unac_filter to filter characters for search\n"; |
629 |
|
require $my_unac_filter; |
630 |
|
} else { |
631 |
|
print STDERR "### fallback to default my_unac_string!\n"; |
632 |
|
eval q{ |
633 |
|
sub main::my_unac_string($$) { |
634 |
|
my ($charset, $string) = (@_); |
635 |
|
return $string; |
636 |
|
} |
637 |
|
}; |
638 |
} |
} |
639 |
|
|
640 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
641 |
|
|
642 |
|
# save database name in global variable path for later |
643 |
|
# (need for index filter creation) |
644 |
|
$path = $database; |
645 |
|
|
646 |
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"; |
647 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
648 |
|
|
650 |
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
651 |
if ($lookup_file) { |
if ($lookup_file) { |
652 |
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
653 |
|
if (! -e $lookup_file) { |
654 |
|
open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!"; |
655 |
|
close(LOOKUP); |
656 |
|
} |
657 |
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
658 |
print STDERR "creating lookup file '$lookup_file'\n"; |
print STDERR "creating lookup file '$lookup_file'\n"; |
659 |
|
# delete memory cache for lookup file |
660 |
|
delete $cache->{lhash}; |
661 |
} |
} |
662 |
|
|
663 |
# open existing lookup file |
# open existing lookup file |
668 |
print STDERR "opening lookup file '$lookup_file'\n"; |
print STDERR "opening lookup file '$lookup_file'\n"; |
669 |
} |
} |
670 |
|
|
671 |
print STDERR "reading ./import_xml/$type.xml\n"; |
my $import_xml_file = "./import_xml/$type.xml"; |
672 |
|
|
673 |
|
if (! -r $import_xml_file) { |
674 |
|
print STDERR "ERROR: file $import_xml_file not readable skipping!\n"; |
675 |
|
next; |
676 |
|
} |
677 |
|
|
678 |
|
print STDERR "reading $import_xml_file\n"; |
679 |
|
|
680 |
# extract just type basic |
# extract just type basic |
681 |
my $type_base = $type; |
my $type_base = $type; |
682 |
$type_base =~ s/_.+$//g; |
$type_base =~ s/_.+$//g; |
683 |
|
|
684 |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1); |
$config=XMLin($import_xml_file, ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 ); |
685 |
|
|
686 |
|
# helper for progress bar |
687 |
|
sub fmt_time { |
688 |
|
my $t = shift || 0; |
689 |
|
my $out = ""; |
690 |
|
|
691 |
|
my ($ss,$mm,$hh) = gmtime($t); |
692 |
|
$out .= "${hh}h" if ($hh); |
693 |
|
$out .= sprintf("%02d:%02d", $mm,$ss); |
694 |
|
$out .= " " if ($hh == 0); |
695 |
|
return $out; |
696 |
|
} |
697 |
|
|
698 |
# output current progress indicator |
# output current progress indicator |
699 |
my $last_p = 0; |
my $last_p = 0; |
700 |
|
my $start_t = time(); |
701 |
sub progress { |
sub progress { |
702 |
return if (! $show_progress); |
return if (! $show_progress); |
703 |
my $current = shift; |
my $current = shift; |
704 |
my $total = shift || 1; |
my $total = shift || 1; |
705 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
706 |
if ($p != $last_p) { |
if ($p < $last_p || $current == 1) { |
707 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
$start_t = time(); |
708 |
|
$last_p = 0; |
709 |
|
} elsif ($p != $last_p) { |
710 |
|
my $rate = ($current / (time() - $start_t || 1)); |
711 |
|
my $eta = ($total-$current) / ($rate || 1); |
712 |
|
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta)); |
713 |
$last_p = $p; |
$last_p = $p; |
714 |
} |
} |
715 |
} |
} |
716 |
|
|
717 |
my $fake_dir = 1; |
my $fake_dir = 1; |
718 |
|
my $fake_pos = 0; |
719 |
|
my $last_fake_t = time(); |
720 |
sub fakeprogress { |
sub fakeprogress { |
721 |
return if (! $show_progress); |
return if (! $show_progress); |
722 |
my $current = shift @_; |
my $current = shift @_; |
723 |
|
|
724 |
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
my @ind = ('-','\\','|','/','-','\\','|','/'); |
725 |
|
|
726 |
|
if ($current < $fake_pos) { |
727 |
|
$start_t = time(); |
728 |
|
$last_fake_t = 0; |
729 |
|
$fake_dir = 1; |
730 |
|
$fake_pos = 0; |
731 |
|
} |
732 |
|
|
733 |
|
if (time()-$last_fake_t >= 1) { |
734 |
|
$last_fake_t = time(); |
735 |
|
$fake_pos += $fake_dir; |
736 |
|
$fake_dir = -$fake_dir if ($fake_pos > 38); |
737 |
|
} |
738 |
|
|
739 |
$last_p += $fake_dir; |
if ($current % 10 == 0) { |
740 |
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
my $rate = ($current / (time() - $start_t || 1)); |
741 |
if ($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]); |
|
742 |
} |
} |
743 |
} |
} |
744 |
|
|
745 |
# now read database |
# now read database |
746 |
print STDERR "using: $type...\n"; |
print STDERR "using: $type...\n"; |
747 |
|
|
748 |
|
# erase cache for tags by order in this database |
749 |
|
delete $cache->{tags_by_order}; |
750 |
|
|
751 |
if ($type_base eq "isis") { |
if ($type_base eq "isis") { |
752 |
|
|
753 |
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!"; |
754 |
|
|
755 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
756 |
my $db = OpenIsis::open( $isis_db ); |
my $db = new Biblio::Isis( isisdb => $isis_db ); |
|
|
|
|
# check if .txt database for OpenIsis is zero length, |
|
|
# if so, erase it and re-open database |
|
|
sub check_txt_db { |
|
|
my $isis_db = shift || die "need isis database name"; |
|
|
my $reopen = 0; |
|
|
|
|
|
if (-e $isis_db.".TXT") { |
|
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
|
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
|
|
$reopen++; |
|
|
} |
|
|
if (-e $isis_db.".PTR") { |
|
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
|
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
|
|
$reopen++; |
|
|
} |
|
|
return OpenIsis::open( $isis_db ) if ($reopen); |
|
|
} |
|
|
|
|
|
# EOF error |
|
|
if ($db == -1) { |
|
|
$db = check_txt_db($isis_db); |
|
|
if ($db == -1) { |
|
|
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
|
|
next; |
|
|
} |
|
|
} |
|
757 |
|
|
758 |
# OpenIsis::ERR_BADF |
if (! $db) { |
759 |
if ($db == -4) { |
print STDERR "FATAL: can't read ISIS database: $isis_db, skipping...\n"; |
|
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"; |
|
760 |
next; |
next; |
761 |
} |
} |
762 |
|
|
763 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
my $max_rowid = $db->count if ($db); |
|
|
|
|
# if 0 records, try to rease isis .txt database |
|
|
if ($max_rowid == 0) { |
|
|
# force removal of database |
|
|
$db = check_txt_db($isis_db); |
|
|
$max_rowid = OpenIsis::maxRowid( $db ); |
|
|
} |
|
764 |
|
|
765 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
766 |
|
|
|
my $path = $database; |
|
|
|
|
767 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
768 |
my $row = OpenIsis::read( $db, $row_id ); |
my $row = $db->to_hash( $row_id ); |
769 |
if ($row && $row->{mfn}) { |
if ($row) { |
770 |
|
|
771 |
|
$row->{mfn} = $row_id; |
772 |
|
$row->{record} = $db->{record}; |
773 |
|
|
774 |
progress($row->{mfn}, $max_rowid); |
progress($row->{mfn}, $max_rowid); |
775 |
|
|
776 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
784 |
} |
} |
785 |
} |
} |
786 |
} |
} |
|
# for this to work with current version of OpenIsis (0.9.0) |
|
|
# you might need my patch from |
|
|
# http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
|
|
OpenIsis::close($db); |
|
787 |
print STDERR "\n"; |
print STDERR "\n"; |
788 |
|
|
789 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
790 |
use Spreadsheet::ParseExcel; |
require Spreadsheet::ParseExcel; |
791 |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
require Spreadsheet::ParseExcel::Utility; |
792 |
|
import Spreadsheet::ParseExcel::Utility qw(int2col); |
793 |
|
|
794 |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
795 |
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!"; |
817 |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
818 |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
819 |
if ($cell) { |
if ($cell) { |
820 |
$row->{int2col($iC)} = $cell->Value; |
# this conversion is a cludge. |
821 |
|
# Files from Excell could have |
822 |
|
# characters which don't fit into |
823 |
|
# destination encoding. |
824 |
|
$row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value; |
825 |
} |
} |
826 |
} |
} |
827 |
|
|
845 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
846 |
} |
} |
847 |
} |
} |
848 |
|
|
849 |
|
print STDERR "\n"; |
850 |
|
|
851 |
} elsif ($type_base eq "marc") { |
} elsif ($type_base eq "marc") { |
852 |
|
|
853 |
use MARC; |
require MARC::File::USMARC; |
854 |
|
|
855 |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
856 |
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!"; |
857 |
|
|
858 |
# optional argument is format |
# optional argument is format |
859 |
my $format = x($config->{format}) || 'usmarc'; |
warn "marc_format is no longer used!" if ($config->{marc_format}); |
|
|
|
860 |
print STDERR "Reading MARC file '$marc_file'\n"; |
print STDERR "Reading MARC file '$marc_file'\n"; |
861 |
|
|
862 |
my $marc = new MARC; |
my $marc = MARC::File::USMARC->in( $marc_file ); |
863 |
my $nr = $marc->openmarc({ |
|
864 |
file=>$marc_file, format=>$format |
if (! $marc) { |
865 |
}) || die "Can't open MARC file '$marc_file'"; |
print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n"; |
866 |
|
next; |
867 |
|
} |
868 |
|
|
869 |
|
# count records in MARC file |
870 |
|
sub marc_count { |
871 |
|
my $filename = shift || die; |
872 |
|
my $file = MARC::File::USMARC->in($filename) || return; |
873 |
|
my $count = 0; |
874 |
|
while ($file->skip()) { |
875 |
|
$count++; |
876 |
|
} |
877 |
|
return $count; |
878 |
|
} |
879 |
|
|
880 |
my $i=0; # record nr. |
my $count = marc_count($marc_file) || warn "no records in '$marc_file'?"; |
881 |
|
|
882 |
my $rec; |
my $i = 1; |
883 |
|
|
884 |
while ($marc->nextmarc(1)) { |
while( my $rec = $marc->next() ) { |
885 |
|
|
886 |
# XXX |
progress($i,$count); |
|
fakeprogress($i++); |
|
887 |
|
|
888 |
my $swishpath = $database."#".$i; |
my $swishpath = $database."#".$i; |
889 |
|
|
890 |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) { |
891 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
892 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
893 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
894 |
print "Content-Length: ".(length($xml)+1)."\n"; |
print "Content-Length: ".(length($xml)+1)."\n"; |
895 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
896 |
} |
} |
897 |
|
|
898 |
|
$i++; |
899 |
} |
} |
900 |
|
|
901 |
|
print STDERR "\n"; |
902 |
|
|
903 |
} elsif ($type_base eq "feed") { |
} elsif ($type_base eq "feed") { |
904 |
|
|
905 |
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
940 |
} |
} |
941 |
# close lookup |
# close lookup |
942 |
untie %lhash if (%lhash); |
untie %lhash if (%lhash); |
943 |
|
|
944 |
|
} elsif ($type_base eq "dbf") { |
945 |
|
|
946 |
|
my $dbf_file = $cfg -> val($database, 'dbf_file') || die "$database doesn't have 'dbf_file' defined!"; |
947 |
|
my $dbf_codepage = $cfg -> val($database, 'dbf_codepage') || die "$database doesn't have 'dbf_codepage' defined!"; |
948 |
|
my $dbf_mapping = $cfg -> val($database, 'dbf_mapping') || die "$database doesn't have 'dbf_mapping' defined!"; |
949 |
|
|
950 |
|
$import2cp = Text::Iconv->new($dbf_codepage,$codepage); |
951 |
|
require XBase; |
952 |
|
my $db = new XBase $dbf_file; |
953 |
|
|
954 |
|
if (! $db) { |
955 |
|
print STDERR "ERROR: can't read DBF database: $dbf_file, skipping...\n"; |
956 |
|
next; |
957 |
|
} |
958 |
|
|
959 |
|
my $max_rowid = $db->last_record; |
960 |
|
|
961 |
|
print STDERR "Reading database: $dbf_file [$max_rowid rows]\n"; |
962 |
|
|
963 |
|
my %dbf2iso; |
964 |
|
foreach my $m (split(/[\n\r]+/,$dbf_mapping)) { |
965 |
|
my ($col,$fld) = split(/\s+/,$m,2); |
966 |
|
$dbf2iso{$col} = $fld; |
967 |
|
} |
968 |
|
|
969 |
|
#print STDERR "## dbf2iso: ",Dumper(\%dbf2iso),"\n## /dbf2iso\n"; |
970 |
|
|
971 |
|
# bad, bad... |
972 |
|
require "to_hash.pm"; |
973 |
|
|
974 |
|
foreach my $row_id (0 .. $max_rowid) { |
975 |
|
my $dbf_row = $db->get_record_as_hash($row_id); |
976 |
|
if ($dbf_row) { |
977 |
|
|
978 |
|
#print STDERR "## dbf_row: ",Dumper($dbf_row),"\n## /dbf_row\n"; |
979 |
|
# apply mapping from config file |
980 |
|
# all unspecified records will get _ in |
981 |
|
# front of them - _DELETE will be __DELETE |
982 |
|
my $rec; |
983 |
|
map { |
984 |
|
my $new_fld = $dbf2iso{$_} || '_'.$_; |
985 |
|
my $data = $dbf_row->{$_}; |
986 |
|
push @{ $rec->{$new_fld} }, $data if ($data && $data !~ /^(?:\s+|\$a\.|)$/); |
987 |
|
} keys %{$dbf_row}; |
988 |
|
#print STDERR "## rec: ",Dumper($rec),"\n## /rec\n"; |
989 |
|
my $row = to_hash($row_id+1, $rec); |
990 |
|
|
991 |
|
$row->{mfn} = $row_id+1; |
992 |
|
$row->{record} = $rec; |
993 |
|
|
994 |
|
#print STDERR "## row: ",Dumper($row),"\n## /row\n"; |
995 |
|
progress($row->{mfn}, $max_rowid); |
996 |
|
|
997 |
|
my $swishpath = $path."#".int($row->{mfn}); |
998 |
|
|
999 |
|
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
1000 |
|
$xml = $cp2utf->convert($xml); |
1001 |
|
use bytes; # as opposed to chars |
1002 |
|
print "Path-Name: $swishpath\n"; |
1003 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
1004 |
|
print "Document-Type: XML\n\n$xml\n"; |
1005 |
|
} |
1006 |
|
} |
1007 |
|
} |
1008 |
|
print STDERR "\n"; |
1009 |
} |
} |
1010 |
} |
} |
1011 |
|
|
1020 |
|
|
1021 |
all2xml.pl - read various file formats and dump XML for SWISH-E |
all2xml.pl - read various file formats and dump XML for SWISH-E |
1022 |
|
|
1023 |
|
=head1 SYNOPSYS |
1024 |
|
|
1025 |
|
$ all2xml.pl [test.conf] |
1026 |
|
|
1027 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
1028 |
|
|
1029 |
This command will read ISIS data file using OpenIsis perl module, MARC |
This command will read ISIS data file using Biblio::Isis perl module, MARC |
1030 |
records using MARC module and optionally Micro$oft Excel files to |
records using MARC::File module and optionally Micro$oft Excel files to |
1031 |
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, |
1032 |
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 |
1033 |
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. |
1034 |
|
|
1035 |
|
If no configuration file is specified, it will use default one called |
1036 |
|
C<all2xml.conf>. |
1037 |
|
|
1038 |
=head1 BUGS |
=head1 BUGS |
1039 |
|
|
1040 |
Documentation is really lacking. However, in true Open Source spirit, source |
Documentation is really lacking. However, in true Open Source spirit, source |