/[webpac]/branches/unesco/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/unesco/all2xml.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 731 - (show annotations)
Thu Apr 13 19:47:32 2006 UTC (13 years, 7 months ago) by dpavlin
File MIME type: text/plain
File size: 29849 byte(s)
updated 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 use Carp;
15
16 $|=1;
17
18 my $config_file = $0;
19 $config_file =~ s/\.pl$/.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);
22
23 my $config;
24
25 #use index_DBI; # default DBI module for index
26 #use index_DBI_cache; # faster DBI module using memory cache
27 use index_DBI_filter; # filter support for indexes
28 my $index;
29
30 my %opts;
31
32 # usage:
33 # -d directory name
34 # -m multiple directories
35 # -q quiet
36 # -s run swish
37
38 getopts('d:m:qs', \%opts);
39
40 my $path; # this is name of database
41
42 Text::Iconv->raise_error(0); # Conversion errors don't raise exceptions
43
44 # this is encoding of all files on disk, including import_xml/*.xml file and
45 # filter/*.pm files! It will be used to store strings in perl internally!
46 my $codepage = 'ISO-8859-2';
47
48 my $utf2cp = Text::Iconv->new('UTF-8',$codepage);
49 # this function will convert data from XML files to local encoding
50 sub x {
51 return $utf2cp->convert($_[0]);
52 }
53
54 # decode isis/excel or other import codepage
55 my $import2cp;
56
57 # outgoing xml must be in UTF-8
58 my $cp2utf = Text::Iconv->new($codepage,'UTF-8');
59
60 # mapping between data type and tag which specify
61 # format in XML file
62 my %type2tag = (
63 'isis' => 'isis',
64 'excel' => 'column',
65 '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 sub data2xml {
80
81 use xmlify;
82
83 my $type = shift @_;
84 my $row = shift @_;
85 my $add_xml = shift @_;
86 # needed to read values from configuration file
87 my $cfg = shift @_;
88 my $database = shift @_;
89
90 my $xml;
91
92 use parse_format;
93
94 my $html = ""; # html formatted display output
95
96 my %field_usage; # counter for usage of each field
97
98 # sort subrouting using order="" attribute
99 sub by_order {
100 my $va = $config->{indexer}->{$a}->{order} ||
101 $config->{indexer}->{$a};
102 my $vb = $config->{indexer}->{$b}->{order} ||
103 $config->{indexer}->{$b};
104
105 return $va <=> $vb;
106 }
107
108 my @sorted_tags;
109 if ($cache->{tags_by_order}) {
110 @sorted_tags = @{$cache->{tags_by_order}};
111 } else {
112 @sorted_tags = sort by_order keys %{$config->{indexer}};
113 $cache->{tags_by_order} = \@sorted_tags;
114 }
115
116 if (! @sorted_tags) {
117 print STDERR "WARNING: no tags for this type found in import_xml file!\n";
118 }
119
120 # lookup key
121 my $lookup_key;
122
123 # cache for field in pages
124 delete $cache->{display_data};
125 delete $cache->{swish_data};
126 delete $cache->{swish_exact_data};
127 delete $cache->{index_data};
128 delete $cache->{index_delimiter};
129 delete $cache->{distinct};
130 my @page_fields; # names of fields
131
132
133 # subs used to produce output
134
135 sub get_field_name($$$) {
136 my ($config,$field,$field_usage) = @_;
137
138 # find field name (signular, plural)
139 my $field_name = "";
140 if ($config->{indexer}->{$field}->{name_singular} && $field_usage == 1) {
141 $field_name = $config->{indexer}->{$field}->{name_singular};
142 } elsif ($config->{indexer}->{$field}->{name_plural}) {
143 $field_name = $config->{indexer}->{$field}->{name_plural};
144 } elsif ($config->{indexer}->{$field}->{name}) {
145 $field_name = $config->{indexer}->{$field}->{name};
146 } else {
147 print STDERR "WARNING: field '$field' doesn't have 'name' attribute!";
148 }
149
150 if ($field_name) {
151 $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
216 foreach my $field (@sorted_tags) {
217
218 $field=x($field);
219 $field_usage{$field}++;
220
221 my $swish_data = "";
222 my $swish_exact_data = "";
223 my $display_data = "";
224 my @index_data;
225 my $line_delimiter;
226
227 my ($swish,$display);
228
229 my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type} || die "can't find which tag to use for type $type";
230
231 # is this field page-by-page?
232 my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page};
233 push @page_fields,$field if ($iterate_by_page);
234 my %page_max = ();
235 # default line_delimiter if using
236 my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>';
237 $cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter};
238 my $distinct = $config->{indexer}->{$field}->{distinct};
239 if ($distinct && !$iterate_by_page) {
240 warn "WARNING: distinct is currently not supported without iterate_by_page!\n";
241 $distinct = 0;
242 }
243
244 my $format_name = $config->{indexer}->{$field}->{format_name};
245 my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter};
246 if ($format_name && $format_delimiter) {
247 $cache->{format}->{$field}->{format_name} = $format_name;
248 $cache->{format}->{$field}->{format_delimiter} = $format_delimiter;
249 }
250
251 foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) {
252
253 $x = unroll_x($x);
254
255 my $format = x($x->{content});
256 my $delimiter = x($x->{delimiter}) || ' ';
257
258 my $repeat_off = 0; # init repeatable offset
259
260 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
261
262 # what will separate last line from this one?
263 if ($display_data && $x->{append}) {
264 $line_delimiter = $delimiter;
265 } elsif ($display_data) {
266 $line_delimiter = '<br/>';
267 }
268
269 # init vars so that we go into while...
270 ($swish,$display) = (1,1);
271
272 sub mkformat($$) {
273 my $x = shift || die "mkformat needs tag reference";
274 my $data = shift || return;
275 my $format_name = x($x->{format_name}) || return $data;
276 my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!";
277 my $format_delimiter = x($x->{format_delimiter});
278 my @data;
279 if ($format_delimiter) {
280 @data = split(/$format_delimiter/,$data);
281 } else {
282 push @data,$data;
283 }
284
285 if ($fmt) {
286 my $nr = scalar $fmt =~ s/%s/%s/g;
287 if (($#data+1) == $nr) {
288 return sprintf($fmt,@data);
289 } else {
290 #print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n";
291 return $data;
292 }
293 } else {
294 print STDERR "usage of link '$format_name' without defined format (<link> tag)\n";
295 }
296 }
297
298 # while because of repeatable fields
299 while ($swish || $display) {
300 my $page = $repeat_off;
301 $page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0));
302 ($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp);
303 if ($repeat_off > 1000) {
304 print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n";
305 last;
306 }
307
308 # is this field is lookup?
309 if ($display && $x->{lookup}) {
310 my $null = "<!-- null -->";
311 if ($use_lhash_cache) {
312 if (!defined($cache->{lhash}->{$display})) {
313 my $new_display = $lhash{$display};
314 if (defined($new_display)) {
315 #print STDERR "lookup cache store '$display' = '$new_display'\n";
316 $display = $new_display;
317 $cache->{lhash}->{$display} = $new_display;
318 } else {
319 # print STDERR "WARNING: lookup for '$display' didn't find anything.\n";
320 $display = "";
321 $cache->{lhash}->{$display} = $null;
322 }
323 } else {
324 $display = $cache->{lhash}->{$display};
325 }
326 } else {
327 $display = $lhash{$display} || $null;
328 }
329 }
330
331 # filter="name" ; filter this field through
332 # filter/[name].pm
333 my $filter = $x->{filter};
334 if ($filter && !$cache->{filter_loaded}->{$filter}) {
335 require "filter/".$filter.".pm";
336 $cache->{filter_loaded}->{$filter}++;
337 }
338 # type="swish" ; field for swish
339 if ($swish) {
340 my $tmp = $swish;
341 if ($filter && ($s || $se)) {
342 no strict 'refs';
343 $tmp = join(" ",&$filter($tmp)) if ($s || $se);
344 }
345
346 $swish_data .= $tmp if ($s && $tmp);
347 $swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se);
348 }
349
350 # type="display" ; field for display
351 if ($d && $display) {
352 my $ldel = $delimiter;
353 if ($line_delimiter && $display_data) {
354 $ldel = $line_delimiter;
355 }
356 if ($filter) {
357 no strict 'refs';
358 my @arr;
359 foreach my $tmp (&$filter($display)) {
360 my $tmp2 = mkformat($x,$tmp);
361 push @arr,$tmp2 if ($tmp2);
362 }
363 $display_data .= $ldel if ($display_data && @arr);
364 $display_data .= join($delimiter,@arr);
365 } else {
366 $display_data .= $ldel if ($display_data);
367 my $tmp = mkformat($x,$display);
368 $display_data .= $tmp if ($tmp);
369 }
370 }
371
372 # type="index" ; insert into index
373 my $idisplay;
374 if ($i && $display) {
375 $idisplay = $display;
376 if ($filter) {
377 no strict 'refs';
378 $idisplay = &$filter($idisplay);
379 }
380 push @index_data, $idisplay if ($idisplay && !$iterate_by_page);
381 }
382
383 # store fields in lookup
384 if ($il && $display) {
385 if (lc($x->{type}) eq "lookup_key") {
386 if ($lookup_key) {
387 print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)";
388 } else {
389 if ($filter) {
390 no strict 'refs';
391 $lookup_key = &$filter($display);
392 } else {
393 $lookup_key = $display;
394 }
395 }
396 } elsif (lc($x->{type}) eq "lookup_val") {
397 if ($lookup_key) {
398 if ($filter) {
399 no strict 'refs';
400 $lhash{$lookup_key} = &$filter($display);
401 } else {
402 $lhash{$lookup_key} = $display;
403 }
404 } else {
405 print STDERR "WARNING: no lookup_key defined for '$display'?";
406 }
407 }
408
409 }
410
411 # store data for page-by-page repeatable fields
412 if ($iterate_by_page) {
413 sub iterate_fld($$$$$$) {
414 my ($cache,$what,$field,$page,$data,$append) = @_;
415 return if (!$data);
416
417 my $ldel = $page_line_delimiter;
418 $ldel = " " if ($append);
419 #print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
420 if (! $cache->{$what}->{$field}->[$page]) {
421 push @{$cache->{$what}->{$field}->[$page]}, {
422 data => $data,
423 delimiter => $ldel,
424 };
425 }
426 }
427
428 if ($display_data) {
429 iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append});
430 }
431 $display_data = "";
432 if ($swish_data) {
433 iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append});
434 $swish_data = "";
435 }
436 if ($swish_exact_data) {
437 iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append});
438 $swish_exact_data = "";
439 }
440
441 if ($idisplay) {
442 my $ldel=$page_line_delimiter;
443 my @index_data;
444 if ($cache->{index_data}->{$field}->[$page]) {
445
446 @index_data = @{$cache->{index_data}->{$field}->[$page]};
447 }
448 if ($x->{append}) {
449 if (@index_data) {
450 $index_data[$#index_data] .= $idisplay;
451 } else {
452 push @index_data, $idisplay;
453 }
454 } else {
455 push @index_data, $idisplay;
456 }
457 $idisplay = "";
458 @{$cache->{index_data}->{$field}->[$page]} = @index_data;
459 }
460 }
461 }
462
463 if (! $iterate_by_page) {
464 my $idel = $x->{index_delimiter};
465 # fill data in index
466 foreach my $tmp (@index_data) {
467 my $i = $d = $tmp;
468 if ($idel && $tmp =~ m/$idel/) {
469 ($i,$d) = split(/$idel/,$tmp);
470 }
471 $index->insert($field, $i, $d, $path);
472 }
473 @index_data = ();
474 }
475 }
476
477 # now try to parse variables from configuration file
478 foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) {
479
480 $x = unroll_x($x);
481
482 my $delimiter = x($x->{delimiter}) || ' ';
483 my $val = $cfg->val($database, x($x->{content}));
484
485 # FIXME index_lookup is not supported!
486 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
487
488 if ($val) {
489 $display_data .= $delimiter.$val if ($d);
490 $swish_data .= " ".$val if ($s);
491 $index->insert($field, $val, $path) if ($i);
492 }
493
494 if ($iterate_by_page) {
495 # FIXME data from config tag will appear just
496 # on first page!!!
497 my $page = 0;
498 if ($display_data) {
499 push @{$cache->{display_data}->{$field}->[$page]}, { data => $display_data };
500 $display_data = "";
501 }
502 if ($swish_data) {
503 push @{$cache->{swish_data}->{$field}->[$page]}, { data => $swish_data };
504 $swish_data = "";
505 }
506 if ($swish_exact_data) {
507 push @{$cache->{swish_exact_data}->{$field}->[$page]}, { data => $swish_exact_data };
508 $swish_exact_data = "";
509 }
510 }
511 }
512
513 # save data page-by-page
514 foreach my $field (@page_fields) {
515 my $nr_pages = $page_max{$field} || next;
516 #print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
517 #print STDERR Dumper($cache->{display_data});
518 my $seen; # used for distinct
519 for (my $page=0; $page <= $nr_pages; $page++) {
520 my $display_data;
521 my $delimiter = '';
522 foreach my $element (@{ $cache->{display_data}->{$field}->[$page] }) {
523 my $data = $element->{data};
524 die "BUG! no data in element?" unless ($data);
525
526 if ($distinct) {
527 next if ($cache->{distinct}->{$field}->{ $data });
528 $cache->{distinct}->{$field}->{ $data } = 1;
529 }
530
531 if ($cache->{format}->{$field}) {
532 my $tmp = mkformat($cache->{format}->{$field},$data);
533 $display_data .= $delimiter . $tmp if ($tmp);
534 } else {
535 $display_data .= $delimiter . $data;
536 }
537 $delimiter = $element->{delimiter} if ($element->{delimiter});
538 }
539
540 if ($display_data) { # default
541 if ($field eq "headline") {
542 $xml .= xmlify("headline", $display_data);
543 } else {
544
545 # fallback to empty field name if needed
546 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
547 $html .= "#-#".$display_data."###\n";
548 }
549 }
550
551 my $swish_data = join(" ",map { $_->{data} } @{ $cache->{swish_data}->{$field}->[$page] });
552 if ($swish_data) {
553 # remove extra spaces
554 $swish_data =~ s/ +/ /g;
555 $swish_data =~ s/ +$//g;
556
557 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
558 }
559
560 my $swish_exact_data = join(" ", map { $_->{data} } @{ $cache->{swish_exact_data}->{$field}->[$page] });
561 if ($swish_exact_data) {
562 $swish_exact_data =~ s/ +/ /g;
563 $swish_exact_data =~ s/ +$//g;
564
565 # add delimiters before and after word.
566 # That is required to produce exact match
567 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
568 }
569
570 my $idel = $cache->{index_delimiter}->{$field};
571 foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) {
572 my $i = $tmp;
573 my $d = $tmp;
574 if ($idel && $tmp =~ m/$idel/) {
575 ($i,$d) = split(/$idel/,$tmp);
576 }
577 $index->insert($field, $i, $d, $path);
578 #print STDERR "index [$idel] $field: $i --> $d [$path]\n";
579 }
580 }
581
582 }
583
584 if (! $iterate_by_page) {
585 if ($display_data) {
586 if ($field eq "headline") {
587 $xml .= xmlify("headline", $display_data);
588 } else {
589
590 # fallback to empty field name if needed
591 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
592 $html .= "#-#".$display_data."###\n";
593 }
594 }
595 if ($swish_data) {
596 # remove extra spaces
597 $swish_data =~ s/ +/ /g;
598 $swish_data =~ s/ +$//g;
599
600 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
601 }
602
603 if ($swish_exact_data) {
604 $swish_exact_data =~ s/ +/ /g;
605 $swish_exact_data =~ s/ +$//g;
606
607 # add delimiters before and after word.
608 # That is required to produce exact match
609 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
610 }
611 }
612 }
613
614 # dump formatted output in <html>
615 if ($html) {
616 #$xml .= xmlify("html",$html);
617 $xml .= "<html><![CDATA[ $html ]]></html>";
618 }
619
620 if ($xml) {
621 $xml .= $add_xml if ($add_xml);
622 return "<xml>\n$xml</xml>\n";
623 } else {
624 return;
625 }
626 }
627
628 ##########################################################################
629
630 # read configuration for this script
631 my $cfg = new Config::IniFiles( -file => $config_file );
632
633 # read global.conf configuration
634 my $cfg_global = new Config::IniFiles( -file => 'global.conf' );
635
636 # open index
637 $index = new index_DBI(
638 $cfg_global->val('global', 'dbi_dbd'),
639 $cfg_global->val('global', 'dbi_dsn'),
640 $cfg_global->val('global', 'dbi_user'),
641 $cfg_global->val('global', 'dbi_passwd') || '',
642 );
643
644 my $show_progress = $cfg_global->val('global', 'show_progress');
645
646 my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter');
647 if ($my_unac_filter) {
648 print STDERR "using $my_unac_filter to filter characters for search\n";
649 require $my_unac_filter;
650 } else {
651 print STDERR "### fallback to default my_unac_string!\n";
652 eval q{
653 sub main::my_unac_string($$) {
654 my ($charset, $string) = (@_);
655 return $string;
656 }
657 };
658 }
659
660 foreach my $database ($cfg->Sections) {
661
662 # save database name in global variable path for later
663 # (need for index filter creation)
664 $path = $database;
665
666 my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
667 my $add_xml = $cfg -> val($database, 'xml'); # optional
668
669 # create new lookup file
670 my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
671 if ($lookup_file) {
672 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
673 if (! -e $lookup_file) {
674 open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!";
675 close(LOOKUP);
676 }
677 tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
678 print STDERR "creating lookup file '$lookup_file'\n";
679 # delete memory cache for lookup file
680 delete $cache->{lhash};
681 }
682
683 # open existing lookup file
684 $lookup_file = $cfg -> val($database, 'lookup_open'); # optional
685 if ($lookup_file) {
686 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644;
687 tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644;
688 print STDERR "opening lookup file '$lookup_file'\n";
689 }
690
691 my $import_xml_type = $cfg->val($database, 'import_xml_file') || $type;
692 my $import_xml_file = "./import_xml/$import_xml_type.xml";
693
694 if (! -r $import_xml_file) {
695 print STDERR "ERROR: file $import_xml_file not readable skipping!\n";
696 next;
697 }
698
699 print STDERR "reading $import_xml_file\n";
700
701 # extract just type basic
702 my $type_base = $type;
703 $type_base =~ s/_.+$//g;
704
705 my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type_base} || die "can't find which tag to use for type $type";
706 $config=XMLin($import_xml_file, ForceArray => [ $tag, 'config', 'format' ], ForceContent => 1 );
707
708 # helper for progress bar
709 sub fmt_time {
710 my $t = shift || 0;
711 my $out = "";
712
713 my ($ss,$mm,$hh) = gmtime($t);
714 $out .= "${hh}h" if ($hh);
715 $out .= sprintf("%02d:%02d", $mm,$ss);
716 $out .= " " if ($hh == 0);
717 return $out;
718 }
719
720 # output current progress indicator
721 my $last_p = 0;
722 my $start_t = time();
723 sub progress {
724 return if (! $show_progress);
725 my $current = shift;
726 my $total = shift || 1;
727 my $p = int($current * 100 / $total);
728 if ($p < $last_p || $current == 1) {
729 $start_t = time();
730 $last_p = 0;
731 } elsif ($p != $last_p) {
732 my $rate = ($current / (time() - $start_t || 1));
733 my $eta = ($total-$current) / ($rate || 1);
734 printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta));
735 $last_p = $p;
736 }
737 }
738
739 my $fake_dir = 1;
740 my $fake_pos = 0;
741 my $last_fake_t = time();
742 sub fakeprogress {
743 return if (! $show_progress);
744 my $current = shift @_;
745
746 my @ind = ('-','\\','|','/','-','\\','|','/');
747
748 if ($current < $fake_pos) {
749 $start_t = time();
750 $last_fake_t = 0;
751 $fake_dir = 1;
752 $fake_pos = 0;
753 }
754
755 if (time()-$last_fake_t >= 1) {
756 $last_fake_t = time();
757 $fake_pos += $fake_dir;
758 $fake_dir = -$fake_dir if ($fake_pos > 38);
759 }
760
761 if ($current % 10 == 0) {
762 my $rate = ($current / (time() - $start_t || 1));
763 printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate);
764 }
765 }
766
767 # now read database
768
769 # erase cache for tags by order in this database
770 delete $cache->{tags_by_order};
771
772 if ($type_base eq "isis") {
773
774 my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
775
776 $import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
777 my $db = new Biblio::Isis( isisdb => $isis_db );
778
779 if (! $db) {
780 print STDERR "FATAL: can't read ISIS database: $isis_db, skipping...\n";
781 next;
782 }
783
784 my $max_rowid = $db->count if ($db);
785
786 print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
787
788 for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
789 my $row = $db->to_hash( $row_id );
790 if ($row) {
791
792 $row->{mfn} = $row_id;
793 $row->{record} = $db->{record};
794
795 progress($row->{mfn}, $max_rowid);
796
797 my $swishpath = $path."#".int($row->{mfn});
798
799 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
800 $xml = $cp2utf->convert($xml);
801 use bytes; # as opposed to chars
802 print "Path-Name: $swishpath\n";
803 print "Content-Length: ".(length($xml)+1)."\n";
804 print "Document-Type: XML\n\n$xml\n";
805 }
806 }
807 }
808 print STDERR "\n";
809
810 } elsif ($type_base eq "excel") {
811 require Spreadsheet::ParseExcel;
812 require Spreadsheet::ParseExcel::Utility;
813 import Spreadsheet::ParseExcel::Utility qw(int2col);
814
815 $import2cp = Text::Iconv->new($config->{excel_codepage},$codepage);
816 my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
817
818 my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
819 my $start_row = x($config->{start_row}) || die "no start_row in $type.xml";
820 $start_row--;
821
822 my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
823
824 my $sheet_nr = 0;
825 foreach my $oWks (@{$oBook->{Worksheet}}) {
826 #print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n";
827 last if ($oWks->{Name} eq $sheet);
828 $sheet_nr++;
829 }
830
831 my $oWorksheet = $oBook->{Worksheet}[$sheet_nr];
832
833 print STDERR "using sheet: ",$oWorksheet->{Name},"\n";
834 defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file";
835 my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow};
836
837 for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) {
838 my $row;
839 for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) {
840 my $cell = $oWorksheet->{Cells}[$iR][$iC];
841 if ($cell) {
842 # this conversion is a cludge.
843 # Files from Excell could have
844 # characters which don't fit into
845 # destination encoding.
846 $row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value;
847 }
848 }
849
850 progress($iR, $end_row);
851
852 # print "row[$iR/$end_row] ";
853 # foreach (keys %{$row}) {
854 # print "$_: ",$row->{$_},"\t";
855 # }
856 # print "\n";
857
858 my $swishpath = $database."#".$iR;
859
860 next if (! $row);
861
862 if (my $xml = data2xml($type_base,$row,$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 "marc") {
874
875 require MARC::File::USMARC;
876
877 $import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
878 my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
879
880 # optional argument is format
881 warn "marc_format is no longer used!" if ($config->{marc_format});
882 print STDERR "Reading MARC file '$marc_file'\n";
883
884 my $marc = MARC::File::USMARC->in( $marc_file );
885
886 if (! $marc) {
887 print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n";
888 next;
889 }
890
891 # count records in MARC file
892 sub marc_count {
893 my $filename = shift || die;
894 my $file = MARC::File::USMARC->in($filename) || return;
895 my $count = 0;
896 while ($file->skip()) {
897 $count++;
898 }
899 return $count;
900 }
901
902 my $count = marc_count($marc_file) || warn "no records in '$marc_file'?";
903
904 my $i = 1;
905
906 while( my $rec = $marc->next() ) {
907
908 progress($i,$count);
909
910 my $swishpath = $database."#".$i;
911
912 if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) {
913 $xml = $cp2utf->convert($xml);
914 use bytes; # as opposed to chars
915 print "Path-Name: $swishpath\n";
916 print "Content-Length: ".(length($xml)+1)."\n";
917 print "Document-Type: XML\n\n$xml\n";
918 }
919
920 $i++;
921 }
922
923 print STDERR "\n";
924
925 } elsif ($type_base eq "feed") {
926
927 $import2cp = Text::Iconv->new($config->{feed_codepage},$codepage);
928 my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!";
929
930 print STDERR "Reading feed from program '$prog'\n";
931
932 open(FEED,"feeds/$prog |") || die "can't start $prog: $!";
933
934 my $i=1; # record nr.
935
936 my $data;
937 my $line=1;
938
939 while (<FEED>) {
940 chomp;
941
942 if (/^$/) {
943 my $swishpath = $database."#".$i++;
944
945 if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) {
946 $xml = $cp2utf->convert($xml);
947 use bytes; # as opposed to chars
948 print "Path-Name: $swishpath\n";
949 print "Content-Length: ".(length($xml)+1)."\n";
950 print "Document-Type: XML\n\n$xml\n";
951 }
952 $line = 1;
953 $data = {};
954 next;
955 }
956
957 $line = $1 if (s/^(\d+):\s*//);
958 $data->{$line++} = $_;
959
960 fakeprogress($i);
961
962 }
963 # close lookup
964 untie %lhash if (%lhash);
965
966 } elsif ($type_base eq "dbf") {
967
968 my $dbf_file = $cfg -> val($database, 'dbf_file') || die "$database doesn't have 'dbf_file' defined!";
969 my $dbf_codepage = $cfg -> val($database, 'dbf_codepage') || die "$database doesn't have 'dbf_codepage' defined!";
970 my $dbf_mapping = $cfg -> val($database, 'dbf_mapping') || die "$database doesn't have 'dbf_mapping' defined!";
971
972 $import2cp = Text::Iconv->new($dbf_codepage,$codepage);
973 require XBase;
974 my $db = new XBase $dbf_file;
975
976 if (! $db) {
977 print STDERR "ERROR: can't read DBF database: $dbf_file, skipping...\n";
978 next;
979 }
980
981 my $max_rowid = $db->last_record;
982
983 print STDERR "Reading database: $dbf_file [$max_rowid rows]\n";
984
985 my %dbf2iso;
986 foreach my $m (split(/[\n\r]+/,$dbf_mapping)) {
987 my ($col,$fld) = split(/\s+/,$m,2);
988 $dbf2iso{$col} = $fld;
989 }
990
991 #print STDERR "## dbf2iso: ",Dumper(\%dbf2iso),"\n## /dbf2iso\n";
992
993 # bad, bad...
994 require "to_hash.pm";
995
996 foreach my $row_id (0 .. $max_rowid) {
997 my $dbf_row = $db->get_record_as_hash($row_id);
998 if ($dbf_row) {
999
1000 #print STDERR "## dbf_row: ",Dumper($dbf_row),"\n## /dbf_row\n";
1001 # apply mapping from config file
1002 # all unspecified records will get _ in
1003 # front of them - _DELETE will be __DELETE
1004 my $rec;
1005 map {
1006 my $new_fld = $dbf2iso{$_} || '_'.$_;
1007 my $data = $dbf_row->{$_};
1008 push @{ $rec->{$new_fld} }, $data if ($data && $data !~ /^(?:\s+|\$a\.|)$/);
1009 } keys %{$dbf_row};
1010 #print STDERR "## rec: ",Dumper($rec),"\n## /rec\n";
1011 my $row = to_hash($row_id+1, $rec);
1012
1013 $row->{mfn} = $row_id+1;
1014 $row->{record} = $rec;
1015
1016 #print STDERR "## row: ",Dumper($row),"\n## /row\n";
1017 progress($row->{mfn}, $max_rowid);
1018
1019 my $swishpath = $path."#".int($row->{mfn});
1020
1021 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
1022 $xml = $cp2utf->convert($xml);
1023 use bytes; # as opposed to chars
1024 print "Path-Name: $swishpath\n";
1025 print "Content-Length: ".(length($xml)+1)."\n";
1026 print "Document-Type: XML\n\n$xml\n";
1027 }
1028 }
1029 }
1030 print STDERR "\n";
1031 }
1032 }
1033
1034 # call this to commit index
1035 $index->close;
1036
1037 1;
1038 __END__
1039 ##########################################################################
1040
1041 =head1 NAME
1042
1043 all2xml.pl - read various file formats and dump XML for SWISH-E
1044
1045 =head1 SYNOPSYS
1046
1047 $ all2xml.pl [test.conf]
1048
1049 =head1 DESCRIPTION
1050
1051 This command will read ISIS data file using Biblio::Isis perl module, MARC
1052 records using MARC::File module and optionally Micro$oft Excel files to
1053 create one XML file for usage with I<SWISH-E> indexer. Dispite it's name,
1054 this script B<isn't general xml generator> from isis files (isis allready
1055 has something like that). Output of this script is tailor-made for SWISH-E.
1056
1057 If no configuration file is specified, it will use default one called
1058 C<all2xml.conf>.
1059
1060 =head1 BUGS
1061
1062 Documentation is really lacking. However, in true Open Source spirit, source
1063 is best documentation. I even made considerable effort to comment parts
1064 which are not intuitively clear, so...
1065
1066 =head1 AUTHOR
1067
1068 Dobrica Pavlinusic <dpavlin@rot13.org>
1069
1070 =head1 COPYRIGHT
1071
1072 GNU Public License (GPL) v2 or later
1073
1074 =head1 SEE ALSO
1075
1076 SWISH-E web site at http://www.swish-e.org
1077
1078 =cut

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.26