/[webpac]/branches/ffzg/all2xml.pl
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Contents of /branches/ffzg/all2xml.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 633 - (show annotations)
Sun Jan 16 18:57:15 2005 UTC (14 years, 9 months ago) by dpavlin
File MIME type: text/plain
File size: 26267 byte(s)
update branches to trunk

1 #!/usr/bin/perl -w
2
3 use strict;
4 use Biblio::Isis;
5 use Getopt::Std;
6 use Data::Dumper;
7 use XML::Simple;
8 use Text::Iconv;
9 use Config::IniFiles;
10 use Encode;
11 #use GDBM_File;
12 use Fcntl; # for O_RDWR
13 use TDB_File;
14
15 $|=1;
16
17 my $config_file = $0;
18 $config_file =~ s/\.pl$/.conf/;
19 $config_file = $ARGV[0] if ($ARGV[0] && -f $ARGV[0]);
20 die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file);
21
22 my $config;
23
24 #use index_DBI; # default DBI module for index
25 use index_DBI_cache; # faster DBI module using memory cache
26 my $index;
27
28 my %opts;
29
30 # usage:
31 # -d directory name
32 # -m multiple directories
33 # -q quiet
34 # -s run swish
35
36 getopts('d:m:qs', \%opts);
37
38 my $path; # this is name of database
39
40 Text::Iconv->raise_error(0); # Conversion errors don't raise exceptions
41
42 # this is encoding of all files on disk, including import_xml/*.xml file and
43 # filter/*.pm files! It will be used to store strings in perl internally!
44 my $codepage = 'ISO-8859-2';
45
46 my $utf2cp = Text::Iconv->new('UTF-8',$codepage);
47 # this function will convert data from XML files to local encoding
48 sub x {
49 return $utf2cp->convert($_[0]);
50 }
51
52 # decode isis/excel or other import codepage
53 my $import2cp;
54
55 # outgoing xml must be in UTF-8
56 my $cp2utf = Text::Iconv->new($codepage,'UTF-8');
57
58 # mapping between data type and tag which specify
59 # format in XML file
60 my %type2tag = (
61 'isis' => 'isis',
62 '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 {
78
79 use xmlify;
80
81 my $type = shift @_;
82 my $row = shift @_;
83 my $add_xml = shift @_;
84 # needed to read values from configuration file
85 my $cfg = shift @_;
86 my $database = shift @_;
87
88 my $xml;
89
90 use parse_format;
91
92 my $html = ""; # html formatted display output
93
94 my %field_usage; # counter for usage of each field
95
96 # sort subrouting using order="" attribute
97 sub by_order {
98 my $va = $config->{indexer}->{$a}->{order} ||
99 $config->{indexer}->{$a};
100 my $vb = $config->{indexer}->{$b}->{order} ||
101 $config->{indexer}->{$b};
102
103 return $va <=> $vb;
104 }
105
106 my @sorted_tags;
107 if ($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 # 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
143 if ($field_name) {
144 $field_name = x($field_name);
145 if (! $last_field_name) {
146 $last_field_name = $field_name;
147 return $last_field_name;
148 } elsif ($field_name ne $last_field_name) {
149 $last_field_name = $field_name;
150 return $last_field_name;
151 }
152 }
153 }
154
155
156 # init variables for different types
157 sub init_visible_type($) {
158 my $type = shift;
159
160 # swish, swish_exact, display, index, index_lookup
161 # swish and display defaults
162 my ($s,$se,$d,$i,$il) = (1,0,1,0,0);
163 if (lc($type) eq "display") {
164 $s = 0;
165 } elsif (lc($type) eq "swish") {
166 $d = 0;
167 } elsif (lc($type) eq "index") {
168 ($s,$se,$d,$i) = (0,1,0,1);
169 } elsif (lc($type) eq "swish_exact") {
170 ($s,$se,$d,$i) = (0,1,0,0);
171 } elsif (lc($type) =~ /^lookup/) {
172 ($s,$se,$d,$i,$il) = (0,1,0,0,1);
173 }
174 return ($s,$se,$d,$i,$il);
175 }
176
177
178 # convert
179 #
180 # <tag>
181 # <delimiter>, </delimiter>
182 # <value>200a</value>
183 # </tag>
184 #
185 # to
186 #
187 # <tag delimiter=", ">200a</tag>
188 #
189 # but without loosing spaces in delimiter (becasue
190 # new XML::Simple strips spaces in attribute values
191 # as defined in XML specification)
192 #
193 sub unroll_x($) {
194 my $x = shift;
195
196 if (defined $x->{value}) {
197 my ($v,$d) = ($x->{value}->{content}, $x->{delimiter}->{content});
198 delete $x->{value};
199 delete $x->{delimiter};
200 $x->{content} = $v;
201 $x->{delimiter} = $d;
202 }
203 return $x;
204 }
205
206 # begin real work: go field by field
207 foreach my $field (@sorted_tags) {
208
209 $field=x($field);
210 $field_usage{$field}++;
211
212 my $swish_data = "";
213 my $swish_exact_data = "";
214 my $display_data = "";
215 my @index_data;
216 my $line_delimiter;
217
218 my ($swish,$display);
219
220 my $tag = $type2tag{$type} || die "can't find which tag to use for type $type";
221
222 # is this field page-by-page?
223 my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page};
224 push @page_fields,$field if ($iterate_by_page);
225 my %page_max = ();
226 # default line_delimiter if using
227 my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>';
228 $cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter};
229
230 my $format_name = $config->{indexer}->{$field}->{format_name};
231 my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter};
232 if ($format_name && $format_delimiter) {
233 $cache->{format}->{$field}->{format_name} = $format_name;
234 $cache->{format}->{$field}->{format_delimiter} = $format_delimiter;
235 }
236
237 foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) {
238
239 $x = unroll_x($x);
240
241 my $format = x($x->{content});
242 my $delimiter = x($x->{delimiter}) || ' ';
243
244 my $repeat_off = 0; # init repeatable offset
245
246 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
247
248 # what will separate last line from this one?
249 if ($display_data && $x->{append}) {
250 $line_delimiter = $delimiter;
251 } elsif ($display_data) {
252 $line_delimiter = '<br/>';
253 }
254
255 # init vars so that we go into while...
256 ($swish,$display) = (1,1);
257
258 # placeholder for all repeatable entries for index
259
260 sub mkformat($$) {
261 my $x = shift || die "mkformat needs tag reference";
262 my $data = shift || return;
263 my $format_name = x($x->{format_name}) || return $data;
264 my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!";
265 my $format_delimiter = x($x->{format_delimiter});
266 my @data;
267 if ($format_delimiter) {
268 @data = split(/$format_delimiter/,$data);
269 } else {
270 push @data,$data;
271 }
272
273 if ($fmt) {
274 my $nr = scalar $fmt =~ s/%s/%s/g;
275 if (($#data+1) == $nr) {
276 return sprintf($fmt,@data);
277 } else {
278 #print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n";
279 return $data;
280 }
281 } else {
282 print STDERR "usage of link '$format_name' without defined format (<link> tag)\n";
283 }
284 }
285
286 # while because of repeatable fields
287 while ($swish || $display) {
288 my $page = $repeat_off;
289 $page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0));
290 ($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp);
291 if ($repeat_off > 1000) {
292 print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n";
293 last;
294 }
295
296 # is this field is lookup?
297 if ($display && $x->{lookup}) {
298 my $null = "<!-- null -->";
299 if ($use_lhash_cache) {
300 if (!defined($cache->{lhash}->{$display})) {
301 my $new_display = $lhash{$display};
302 if (defined($new_display)) {
303 #print STDERR "lookup cache store '$display' = '$new_display'\n";
304 $display = $new_display;
305 $cache->{lhash}->{$display} = $new_display;
306 } else {
307 # print STDERR "WARNING: lookup for '$display' didn't find anything.\n";
308 $display = "";
309 $cache->{lhash}->{$display} = $null;
310 }
311 } else {
312 $display = $cache->{lhash}->{$display};
313 }
314 } else {
315 $display = $lhash{$display} || $null;
316 }
317 }
318
319 # filter="name" ; filter this field through
320 # filter/[name].pm
321 my $filter = $x->{filter};
322 if ($filter && !$cache->{filter_loaded}->{$filter}) {
323 require "filter/".$filter.".pm";
324 $cache->{filter_loaded}->{$filter}++;
325 }
326 # type="swish" ; field for swish
327 if ($swish) {
328 my $tmp = $swish;
329 if ($filter && ($s || $se)) {
330 no strict 'refs';
331 $tmp = join(" ",&$filter($tmp)) if ($s || $se);
332 }
333
334 $swish_data .= $tmp if ($s && $tmp);
335 $swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se);
336 }
337
338 # type="display" ; field for display
339 if ($d && $display) {
340 my $ldel = $delimiter;
341 if ($line_delimiter && $display_data) {
342 $ldel = $line_delimiter;
343 }
344 if ($filter) {
345 no strict 'refs';
346 my @arr;
347 foreach my $tmp (&$filter($display)) {
348 my $tmp2 = mkformat($x,$tmp);
349 push @arr,$tmp2 if ($tmp2);
350 }
351 $display_data .= $ldel if ($display_data && @arr);
352 $display_data .= join($delimiter,@arr);
353 } else {
354 $display_data .= $ldel if ($display_data);
355 my $tmp = mkformat($x,$display);
356 $display_data .= $tmp if ($tmp);
357 }
358 }
359
360 # type="index" ; insert into index
361 my $idisplay;
362 if ($i && $display) {
363 $idisplay = $display;
364 if ($filter) {
365 no strict 'refs';
366 $idisplay = &$filter($idisplay);
367 }
368 push @index_data, $idisplay if ($idisplay && !$iterate_by_page);
369 }
370
371 # store fields in lookup
372 if ($il && $display) {
373 if (lc($x->{type}) eq "lookup_key") {
374 if ($lookup_key) {
375 print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)";
376 } else {
377 if ($filter) {
378 no strict 'refs';
379 $lookup_key = &$filter($display);
380 } else {
381 $lookup_key = $display;
382 }
383 }
384 } elsif (lc($x->{type}) eq "lookup_val") {
385 if ($lookup_key) {
386 if ($filter) {
387 no strict 'refs';
388 $lhash{$lookup_key} = &$filter($display);
389 } else {
390 $lhash{$lookup_key} = $display;
391 }
392 } else {
393 print STDERR "WARNING: no lookup_key defined for '$display'?";
394 }
395 }
396
397 }
398
399 # store data for page-by-page repeatable fields
400 if ($iterate_by_page) {
401 sub iterate_fld($$$$$$) {
402 my ($cache,$what,$field,$page,$data,$append) = @_;
403 return if (!$data);
404
405 my $ldel = $page_line_delimiter;
406 $ldel = " " if ($append);
407 #print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
408 if (! $cache->{$what}->{$field}->[$page]) {
409 $cache->{$what}->{$field}->[$page] = $data;
410 } else {
411 $cache->{$what}->{$field}->[$page] .= $ldel.$data;
412 }
413 }
414
415 if ($display_data) {
416 iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append});
417 }
418 $display_data = "";
419 if ($swish_data) {
420 iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append});
421 $swish_data = "";
422 }
423 if ($swish_exact_data) {
424 iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append});
425 $swish_exact_data = "";
426 }
427
428 if ($idisplay) {
429 my $ldel=$page_line_delimiter;
430 my @index_data;
431 if ($cache->{index_data}->{$field}->[$page]) {
432
433 @index_data = @{$cache->{index_data}->{$field}->[$page]};
434 }
435 if ($x->{append}) {
436 if (@index_data) {
437 $index_data[$#index_data] .= $idisplay;
438 } else {
439 push @index_data, $idisplay;
440 }
441 } else {
442 push @index_data, $idisplay;
443 }
444 $idisplay = "";
445 @{$cache->{index_data}->{$field}->[$page]} = @index_data;
446 }
447 }
448 }
449
450 if (! $iterate_by_page) {
451 my $idel = $x->{index_delimiter};
452 # fill data in index
453 foreach my $tmp (@index_data) {
454 my $i = $d = $tmp;
455 if ($idel && $tmp =~ m/$idel/) {
456 ($i,$d) = split(/$idel/,$tmp);
457 }
458 $index->insert($field, $i, $d, $path);
459 }
460 @index_data = ();
461 }
462 }
463
464 # now try to parse variables from configuration file
465 foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) {
466
467 $x = unroll_x($x);
468
469 my $delimiter = x($x->{delimiter}) || ' ';
470 my $val = $cfg->val($database, x($x->{content}));
471
472 # FIXME index_lookup is not supported!
473 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
474
475 if ($val) {
476 $display_data .= $delimiter.$val if ($d);
477 $swish_data .= " ".$val if ($s);
478 $index->insert($field, $val, $path) if ($i);
479 }
480
481 if ($iterate_by_page) {
482 # FIXME data from config tag will appear just
483 # on first page!!!
484 my $page = 0;
485 if ($display_data) {
486 $cache->{display_data}->{$field}->[$page] = $display_data;
487 $display_data = "";
488 }
489 if ($swish_data) {
490 $cache->{swish_data}->{$field}->[$page] = $swish_data;
491 $swish_data = "";
492 }
493 if ($swish_exact_data) {
494 $cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data;
495 $swish_exact_data = "";
496 }
497 }
498 }
499
500 # save data page-by-page
501 foreach my $field (@page_fields) {
502 my $nr_pages = $page_max{$field} || next;
503 #print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
504 #print STDERR Dumper($cache->{display_data});
505 for (my $page=0; $page <= $nr_pages; $page++) {
506 my $display_data;
507 if ($cache->{format}->{$field}) {
508 my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]);
509 $display_data=$tmp if ($tmp);
510 } else {
511 $display_data = $cache->{display_data}->{$field}->[$page];
512 }
513 if ($display_data) { # default
514 if ($field eq "headline") {
515 $xml .= xmlify("headline", $display_data);
516 } else {
517
518 # fallback to empty field name if needed
519 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
520 $html .= "#-#".$display_data."###\n";
521 }
522 }
523
524 my $swish_data = $cache->{swish_data}->{$field}->[$page];
525 if ($swish_data) {
526 # remove extra spaces
527 $swish_data =~ s/ +/ /g;
528 $swish_data =~ s/ +$//g;
529
530 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
531 }
532
533 my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
534 if ($swish_exact_data) {
535 $swish_exact_data =~ s/ +/ /g;
536 $swish_exact_data =~ s/ +$//g;
537
538 # add delimiters before and after word.
539 # That is required to produce exact match
540 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
541 }
542
543 my $idel = $cache->{index_delimiter}->{$field};
544 foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) {
545 my $i = $tmp;
546 my $d = $tmp;
547 if ($idel && $tmp =~ m/$idel/) {
548 ($i,$d) = split(/$idel/,$tmp);
549 }
550 $index->insert($field, $i, $d, $path);
551 #print STDERR "index [$idel] $field: $i --> $d [$path]\n";
552 }
553 }
554
555 }
556
557 if (! $iterate_by_page) {
558 if ($display_data) {
559 if ($field eq "headline") {
560 $xml .= xmlify("headline", $display_data);
561 } else {
562
563 # fallback to empty field name if needed
564 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
565 $html .= "#-#".$display_data."###\n";
566 }
567 }
568 if ($swish_data) {
569 # remove extra spaces
570 $swish_data =~ s/ +/ /g;
571 $swish_data =~ s/ +$//g;
572
573 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
574 }
575
576 if ($swish_exact_data) {
577 $swish_exact_data =~ s/ +/ /g;
578 $swish_exact_data =~ s/ +$//g;
579
580 # add delimiters before and after word.
581 # That is required to produce exact match
582 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
583 }
584 }
585 }
586
587 # dump formatted output in <html>
588 if ($html) {
589 #$xml .= xmlify("html",$html);
590 $xml .= "<html><![CDATA[ $html ]]></html>";
591 }
592
593 if ($xml) {
594 $xml .= $add_xml if ($add_xml);
595 return "<xml>\n$xml</xml>\n";
596 } else {
597 return;
598 }
599 }
600
601 ##########################################################################
602
603 # read configuration for this script
604 my $cfg = new Config::IniFiles( -file => $config_file );
605
606 # read global.conf configuration
607 my $cfg_global = new Config::IniFiles( -file => 'global.conf' );
608
609 # open index
610 $index = new index_DBI(
611 $cfg_global->val('global', 'dbi_dbd'),
612 $cfg_global->val('global', 'dbi_dsn'),
613 $cfg_global->val('global', 'dbi_user'),
614 $cfg_global->val('global', 'dbi_passwd') || '',
615 );
616
617 my $show_progress = $cfg_global->val('global', 'show_progress');
618
619 my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter');
620 if ($my_unac_filter) {
621 print STDERR "using $my_unac_filter to filter characters for search\n";
622 require $my_unac_filter;
623 } else {
624 print STDERR "### fallback to default my_unac_string!\n";
625 eval q{
626 sub main::my_unac_string($$) {
627 my ($charset, $string) = (@_);
628 return $string;
629 }
630 };
631 }
632
633 foreach my $database ($cfg->Sections) {
634
635 my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
636 my $add_xml = $cfg -> val($database, 'xml'); # optional
637
638 # create new lookup file
639 my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
640 if ($lookup_file) {
641 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
642 if (! -e $lookup_file) {
643 open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!";
644 close(LOOKUP);
645 }
646 tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
647 print STDERR "creating lookup file '$lookup_file'\n";
648 # delete memory cache for lookup file
649 delete $cache->{lhash};
650 }
651
652 # open existing lookup file
653 $lookup_file = $cfg -> val($database, 'lookup_open'); # optional
654 if ($lookup_file) {
655 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644;
656 tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644;
657 print STDERR "opening lookup file '$lookup_file'\n";
658 }
659
660 print STDERR "reading ./import_xml/$type.xml\n";
661
662 # extract just type basic
663 my $type_base = $type;
664 $type_base =~ s/_.+$//g;
665
666 $config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 );
667
668 # helper for progress bar
669 sub fmt_time {
670 my $t = shift || 0;
671 my $out = "";
672
673 my ($ss,$mm,$hh) = gmtime($t);
674 $out .= "${hh}h" if ($hh);
675 $out .= sprintf("%02d:%02d", $mm,$ss);
676 $out .= " " if ($hh == 0);
677 return $out;
678 }
679
680 # output current progress indicator
681 my $last_p = 0;
682 my $start_t = time();
683 sub progress {
684 return if (! $show_progress);
685 my $current = shift;
686 my $total = shift || 1;
687 my $p = int($current * 100 / $total);
688 if ($p < $last_p || $current == 1) {
689 $start_t = time();
690 $last_p = 0;
691 } elsif ($p != $last_p) {
692 my $rate = ($current / (time() - $start_t || 1));
693 my $eta = ($total-$current) / ($rate || 1);
694 printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta));
695 $last_p = $p;
696 }
697 }
698
699 my $fake_dir = 1;
700 my $fake_pos = 0;
701 my $last_fake_t = time();
702 sub fakeprogress {
703 return if (! $show_progress);
704 my $current = shift @_;
705
706 my @ind = ('-','\\','|','/','-','\\','|','/');
707
708 if ($current < $fake_pos) {
709 $start_t = time();
710 $last_fake_t = 0;
711 $fake_dir = 1;
712 $fake_pos = 0;
713 }
714
715 if (time()-$last_fake_t >= 1) {
716 $last_fake_t = time();
717 $fake_pos += $fake_dir;
718 $fake_dir = -$fake_dir if ($fake_pos > 38);
719 }
720
721 if ($current % 10 == 0) {
722 my $rate = ($current / (time() - $start_t || 1));
723 printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate);
724 }
725 }
726
727 # now read database
728 print STDERR "using: $type...\n";
729
730 # erase cache for tags by order in this database
731 delete $cache->{tags_by_order};
732
733 if ($type_base eq "isis") {
734
735 my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
736
737 $import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
738 my $db = new Biblio::Isis( isisdb => $isis_db );
739
740 my $max_rowid = $db->count || die "can't find maxmfn";
741
742 print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
743
744 my $path = $database;
745
746 for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
747 my $row = $db->to_hash( $row_id );
748 if ($row) {
749
750 $row->{mfn} = $row_id;
751 $row->{record} = $db->{record};
752
753 progress($row->{mfn}, $max_rowid);
754
755 my $swishpath = $path."#".int($row->{mfn});
756
757 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
758 $xml = $cp2utf->convert($xml);
759 use bytes; # as opposed to chars
760 print "Path-Name: $swishpath\n";
761 print "Content-Length: ".(length($xml)+1)."\n";
762 print "Document-Type: XML\n\n$xml\n";
763 }
764 }
765 }
766 print STDERR "\n";
767
768 } elsif ($type_base eq "excel") {
769 require Spreadsheet::ParseExcel;
770 require Spreadsheet::ParseExcel::Utility;
771 import Spreadsheet::ParseExcel::Utility qw(int2col);
772
773 $import2cp = Text::Iconv->new($config->{excel_codepage},$codepage);
774 my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
775
776 my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
777 my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml";
778
779 my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
780
781 my $sheet_nr = 0;
782 foreach my $oWks (@{$oBook->{Worksheet}}) {
783 #print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n";
784 last if ($oWks->{Name} eq $sheet);
785 $sheet_nr++;
786 }
787
788 my $oWorksheet = $oBook->{Worksheet}[$sheet_nr];
789
790 print STDERR "using sheet: ",$oWorksheet->{Name},"\n";
791 defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file";
792 my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow};
793
794 for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) {
795 my $row;
796 for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) {
797 my $cell = $oWorksheet->{Cells}[$iR][$iC];
798 if ($cell) {
799 # this conversion is a cludge.
800 # Files from Excell could have
801 # characters which don't fit into
802 # destination encoding.
803 $row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value;
804 }
805 }
806
807 progress($iR, $end_row);
808
809 # print "row[$iR/$end_row] ";
810 # foreach (keys %{$row}) {
811 # print "$_: ",$row->{$_},"\t";
812 # }
813 # print "\n";
814
815 my $swishpath = $database."#".$iR;
816
817 next if (! $row);
818
819 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
820 $xml = $cp2utf->convert($xml);
821 use bytes; # as opposed to chars
822 print "Path-Name: $swishpath\n";
823 print "Content-Length: ".(length($xml)+1)."\n";
824 print "Document-Type: XML\n\n$xml\n";
825 }
826 }
827 } elsif ($type_base eq "marc") {
828
829 require MARC::File::USMARC;
830
831 $import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
832 my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
833
834 # optional argument is format
835 warn "marc_format is no longer used!" if ($config->{marc_format});
836 print STDERR "Reading MARC file '$marc_file'\n";
837
838 my $marc = MARC::File::USMARC->in( $marc_file )
839 || die "Can't open MARC file '$marc_file': ".$MARC::File::ERROR;
840
841 # count records in MARC file
842 sub marc_count {
843 my $filename = shift || die;
844 my $file = MARC::File::USMARC->in($filename) || die $MARC::File::ERROR;
845 my $count = 0;
846 while ($file->skip()) {
847 $count++;
848 }
849 return $count;
850 }
851
852 my $count = marc_count($marc_file) || warn "no records in '$marc_file'?";
853
854 my $i = 0;
855
856 while( my $rec = $marc->next() ) {
857
858 progress($i++,$count);
859
860 my $swishpath = $database."#".$i;
861
862 if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) {
863 $xml = $cp2utf->convert($xml);
864 use bytes; # as opposed to chars
865 print "Path-Name: $swishpath\n";
866 print "Content-Length: ".(length($xml)+1)."\n";
867 print "Document-Type: XML\n\n$xml\n";
868 }
869 }
870
871 print STDERR "\n";
872
873 } elsif ($type_base eq "feed") {
874
875 $import2cp = Text::Iconv->new($config->{feed_codepage},$codepage);
876 my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!";
877
878 print STDERR "Reading feed from program '$prog'\n";
879
880 open(FEED,"feeds/$prog |") || die "can't start $prog: $!";
881
882 my $i=1; # record nr.
883
884 my $data;
885 my $line=1;
886
887 while (<FEED>) {
888 chomp;
889
890 if (/^$/) {
891 my $swishpath = $database."#".$i++;
892
893 if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) {
894 $xml = $cp2utf->convert($xml);
895 use bytes; # as opposed to chars
896 print "Path-Name: $swishpath\n";
897 print "Content-Length: ".(length($xml)+1)."\n";
898 print "Document-Type: XML\n\n$xml\n";
899 }
900 $line = 1;
901 $data = {};
902 next;
903 }
904
905 $line = $1 if (s/^(\d+):\s*//);
906 $data->{$line++} = $_;
907
908 fakeprogress($i);
909
910 }
911 # close lookup
912 untie %lhash if (%lhash);
913 }
914 }
915
916 # call this to commit index
917 $index->close;
918
919 1;
920 __END__
921 ##########################################################################
922
923 =head1 NAME
924
925 all2xml.pl - read various file formats and dump XML for SWISH-E
926
927 =head1 SYNOPSYS
928
929 $ all2xml.pl [test.conf]
930
931 =head1 DESCRIPTION
932
933 This command will read ISIS data file using IsisDB perl module, MARC
934 records using MARC module and optionally Micro$oft Excel files to
935 create one XML file for usage with I<SWISH-E> indexer. Dispite it's name,
936 this script B<isn't general xml generator> from isis files (isis allready
937 has something like that). Output of this script is tailor-made for SWISH-E.
938
939 If no configuration file is specified, it will use default one called
940 C<all2xml.conf>.
941
942 =head1 BUGS
943
944 Documentation is really lacking. However, in true Open Source spirit, source
945 is best documentation. I even made considerable effort to comment parts
946 which are not intuitively clear, so...
947
948 =head1 AUTHOR
949
950 Dobrica Pavlinusic <dpavlin@rot13.org>
951
952 =head1 COPYRIGHT
953
954 GNU Public License (GPL) v2 or later
955
956 =head1 SEE ALSO
957
958 SWISH-E web site at http://www.swish-e.org
959
960 =cut

Properties

Name Value
cvs2svn:cvs-rev 1.64
svn:executable *

  ViewVC Help
Powered by ViewVC 1.1.26