/[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 458 - (show annotations)
Tue Sep 21 16:53:44 2004 UTC (19 years, 6 months ago) by dpavlin
File MIME type: text/plain
File size: 26671 byte(s)
update branches with bugfixes from HEAD

1 #!/usr/bin/perl -w
2
3 use strict;
4 use OpenIsis;
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 if ($field_name) {
143 if (! $last_field_name) {
144 $last_field_name = x($field_name);
145 return $last_field_name;
146 } elsif ($field_name ne $last_field_name) {
147 $last_field_name = x($field_name);
148 return $last_field_name;
149 }
150 }
151 }
152
153
154 # init variables for different types
155 sub init_visible_type($) {
156 my $type = shift;
157
158 # swish, swish_exact, display, index, index_lookup
159 # swish and display defaults
160 my ($s,$se,$d,$i,$il) = (1,0,1,0,0);
161 if (lc($type) eq "display") {
162 $s = 0;
163 } elsif (lc($type) eq "swish") {
164 $d = 0;
165 } elsif (lc($type) eq "index") {
166 ($s,$se,$d,$i) = (0,1,0,1);
167 } elsif (lc($type) eq "swish_exact") {
168 ($s,$se,$d,$i) = (0,1,0,0);
169 } elsif (lc($type) =~ /^lookup/) {
170 ($s,$se,$d,$i,$il) = (0,1,0,0,1);
171 }
172 return ($s,$se,$d,$i,$il);
173 }
174
175
176 # convert
177 #
178 # <tag>
179 # <delimiter>, </delimiter>
180 # <value>200a</value>
181 # </tag>
182 #
183 # to
184 #
185 # <tag delimiter=", ">200a</tag>
186 #
187 # but without loosing spaces in delimiter (becasue
188 # new XML::Simple strips spaces in attribute values
189 # as defined in XML specification)
190 #
191 sub unroll_x($) {
192 my $x = shift;
193
194 if (defined $x->{value}) {
195 my ($v,$d) = ($x->{value}->{content}, $x->{delimiter}->{content});
196 delete $x->{value};
197 delete $x->{delimiter};
198 $x->{content} = $v;
199 $x->{delimiter} = $d;
200 }
201 return $x;
202 }
203
204 # begin real work: go field by field
205 foreach my $field (@sorted_tags) {
206
207 $field=x($field);
208 $field_usage{$field}++;
209
210 my $swish_data = "";
211 my $swish_exact_data = "";
212 my $display_data = "";
213 my @index_data;
214 my $line_delimiter;
215
216 my ($swish,$display);
217
218 my $tag = $type2tag{$type} || die "can't find which tag to use for type $type";
219
220 # is this field page-by-page?
221 my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page};
222 push @page_fields,$field if ($iterate_by_page);
223 my %page_max = ();
224 # default line_delimiter if using
225 my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>';
226 $cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter};
227
228 my $format_name = $config->{indexer}->{$field}->{format_name};
229 my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter};
230 if ($format_name && $format_delimiter) {
231 $cache->{format}->{$field}->{format_name} = $format_name;
232 $cache->{format}->{$field}->{format_delimiter} = $format_delimiter;
233 }
234
235 foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) {
236
237 $x = unroll_x($x);
238
239 my $format = x($x->{content});
240 my $delimiter = x($x->{delimiter}) || ' ';
241
242 my $repeat_off = 0; # init repeatable offset
243
244 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
245
246 # what will separate last line from this one?
247 if ($display_data && $x->{append}) {
248 $line_delimiter = $delimiter;
249 } elsif ($display_data) {
250 $line_delimiter = '<br/>';
251 }
252
253 # init vars so that we go into while...
254 ($swish,$display) = (1,1);
255
256 # placeholder for all repeatable entries for index
257
258 sub mkformat($$) {
259 my $x = shift || die "mkformat needs tag reference";
260 my $data = shift || return;
261 my $format_name = x($x->{format_name}) || return $data;
262 my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!";
263 my $format_delimiter = x($x->{format_delimiter});
264 my @data;
265 if ($format_delimiter) {
266 @data = split(/$format_delimiter/,$data);
267 } else {
268 push @data,$data;
269 }
270
271 if ($fmt) {
272 my $nr = scalar $fmt =~ s/%s/%s/g;
273 if (($#data+1) == $nr) {
274 return sprintf($fmt,@data);
275 } else {
276 #print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n";
277 return $data;
278 }
279 } else {
280 print STDERR "usage of link '$format_name' without defined format (<link> tag)\n";
281 }
282 }
283
284 # while because of repeatable fields
285 while ($swish || $display) {
286 my $page = $repeat_off;
287 $page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0));
288 ($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp);
289 if ($repeat_off > 1000) {
290 print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n";
291 last;
292 }
293
294 # is this field is lookup?
295 if ($display && $x->{lookup}) {
296 my $null = "<!-- null -->";
297 if ($use_lhash_cache) {
298 if (!defined($cache->{lhash}->{$display})) {
299 my $new_display = $lhash{$display};
300 if (defined($new_display)) {
301 #print STDERR "lookup cache store '$display' = '$new_display'\n";
302 $display = $new_display;
303 $cache->{lhash}->{$display} = $new_display;
304 } else {
305 # print STDERR "WARNING: lookup for '$display' didn't find anything.\n";
306 $display = "";
307 $cache->{lhash}->{$display} = $null;
308 }
309 } else {
310 $display = $cache->{lhash}->{$display};
311 }
312 } else {
313 $display = $lhash{$display} || $null;
314 }
315 }
316
317 # filter="name" ; filter this field through
318 # filter/[name].pm
319 my $filter = $x->{filter};
320 if ($filter && !$cache->{filter_loaded}->{$filter}) {
321 require "filter/".$filter.".pm";
322 $cache->{filter_loaded}->{$filter}++;
323 }
324 # type="swish" ; field for swish
325 if ($swish) {
326 my $tmp = $swish;
327 if ($filter && ($s || $se)) {
328 no strict 'refs';
329 $tmp = join(" ",&$filter($tmp)) if ($s || $se);
330 }
331
332 $swish_data .= $tmp if ($s && $tmp);
333 $swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se);
334 }
335
336 # type="display" ; field for display
337 if ($d && $display) {
338 my $ldel = $delimiter;
339 if ($line_delimiter && $display_data) {
340 $ldel = $line_delimiter;
341 }
342 if ($filter) {
343 no strict 'refs';
344 my @arr;
345 foreach my $tmp (&$filter($display)) {
346 my $tmp2 = mkformat($x,$tmp);
347 push @arr,$tmp2 if ($tmp2);
348 }
349 $display_data .= $ldel if ($display_data && @arr);
350 $display_data .= join($delimiter,@arr);
351 } else {
352 $display_data .= $ldel if ($display_data);
353 my $tmp = mkformat($x,$display);
354 $display_data .= $tmp if ($tmp);
355 }
356 }
357
358 # type="index" ; insert into index
359 my $idisplay;
360 if ($i && $display) {
361 $idisplay = $display;
362 if ($filter) {
363 no strict 'refs';
364 $idisplay = &$filter($idisplay);
365 }
366 push @index_data, $idisplay if ($idisplay && !$iterate_by_page);
367 }
368
369 # store fields in lookup
370 if ($il && $display) {
371 if (lc($x->{type}) eq "lookup_key") {
372 if ($lookup_key) {
373 print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)";
374 } else {
375 if ($filter) {
376 no strict 'refs';
377 $lookup_key = &$filter($display);
378 } else {
379 $lookup_key = $display;
380 }
381 }
382 } elsif (lc($x->{type}) eq "lookup_val") {
383 if ($lookup_key) {
384 if ($filter) {
385 no strict 'refs';
386 $lhash{$lookup_key} = &$filter($display);
387 } else {
388 $lhash{$lookup_key} = $display;
389 }
390 } else {
391 print STDERR "WARNING: no lookup_key defined for '$display'?";
392 }
393 }
394
395 }
396
397 # store data for page-by-page repeatable fields
398 if ($iterate_by_page) {
399 sub iterate_fld($$$$$$) {
400 my ($cache,$what,$field,$page,$data,$append) = @_;
401 return if (!$data);
402
403 my $ldel = $page_line_delimiter;
404 $ldel = " " if ($append);
405 #print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
406 if (! $cache->{$what}->{$field}->[$page]) {
407 $cache->{$what}->{$field}->[$page] = $data;
408 } else {
409 $cache->{$what}->{$field}->[$page] .= $ldel.$data;
410 }
411 }
412
413 if ($display_data) {
414 iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append});
415 }
416 $display_data = "";
417 if ($swish_data) {
418 iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append});
419 $swish_data = "";
420 }
421 if ($swish_exact_data) {
422 iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append});
423 $swish_exact_data = "";
424 }
425
426 if ($idisplay) {
427 my $ldel=$page_line_delimiter;
428 my @index_data;
429 if ($cache->{index_data}->{$field}->[$page]) {
430
431 @index_data = @{$cache->{index_data}->{$field}->[$page]};
432 }
433 if ($x->{append}) {
434 if (@index_data) {
435 $index_data[$#index_data] .= $idisplay;
436 } else {
437 push @index_data, $idisplay;
438 }
439 } else {
440 push @index_data, $idisplay;
441 }
442 $idisplay = "";
443 @{$cache->{index_data}->{$field}->[$page]} = @index_data;
444 }
445 }
446 }
447
448 if (! $iterate_by_page) {
449 my $idel = $x->{index_delimiter};
450 # fill data in index
451 foreach my $tmp (@index_data) {
452 my $i = $d = $tmp;
453 if ($idel && $tmp =~ m/$idel/) {
454 ($i,$d) = split(/$idel/,$tmp);
455 }
456 $index->insert($field, $i, $d, $path);
457 }
458 @index_data = ();
459 }
460 }
461
462 # now try to parse variables from configuration file
463 foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) {
464
465 $x = unroll_x($x);
466
467 my $delimiter = x($x->{delimiter}) || ' ';
468 my $val = $cfg->val($database, x($x->{content}));
469
470 # FIXME index_lookup is not supported!
471 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
472
473 if ($val) {
474 $display_data .= $delimiter.$val if ($d);
475 $swish_data .= " ".$val if ($s);
476 $index->insert($field, $val, $path) if ($i);
477 }
478
479 if ($iterate_by_page) {
480 # FIXME data from config tag will appear just
481 # on first page!!!
482 my $page = 0;
483 if ($display_data) {
484 $cache->{display_data}->{$field}->[$page] = $display_data;
485 $display_data = "";
486 }
487 if ($swish_data) {
488 $cache->{swish_data}->{$field}->[$page] = $swish_data;
489 $swish_data = "";
490 }
491 if ($swish_exact_data) {
492 $cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data;
493 $swish_exact_data = "";
494 }
495 }
496 }
497
498 # save data page-by-page
499 foreach my $field (@page_fields) {
500 my $nr_pages = $page_max{$field} || next;
501 #print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
502 #print STDERR Dumper($cache->{display_data});
503 for (my $page=0; $page <= $nr_pages; $page++) {
504 my $display_data;
505 if ($cache->{format}->{$field}) {
506 my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]);
507 $display_data=$tmp if ($tmp);
508 } else {
509 $display_data = $cache->{display_data}->{$field}->[$page];
510 }
511 if ($display_data) { # default
512 if ($field eq "headline") {
513 $xml .= xmlify("headline", $display_data);
514 } else {
515
516 # fallback to empty field name if needed
517 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
518 $html .= "#-#".$display_data."###\n";
519 }
520 }
521
522 my $swish_data = $cache->{swish_data}->{$field}->[$page];
523 if ($swish_data) {
524 # remove extra spaces
525 $swish_data =~ s/ +/ /g;
526 $swish_data =~ s/ +$//g;
527
528 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
529 }
530
531 my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
532 if ($swish_exact_data) {
533 $swish_exact_data =~ s/ +/ /g;
534 $swish_exact_data =~ s/ +$//g;
535
536 # add delimiters before and after word.
537 # That is required to produce exact match
538 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
539 }
540
541 my $idel = $cache->{index_delimiter}->{$field};
542 foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) {
543 my $i = $tmp;
544 my $d = $tmp;
545 if ($idel && $tmp =~ m/$idel/) {
546 ($i,$d) = split(/$idel/,$tmp);
547 }
548 $index->insert($field, $i, $d, $path);
549 #print STDERR "index [$idel] $field: $i --> $d [$path]\n";
550 }
551 }
552
553 }
554
555 if (! $iterate_by_page) {
556 if ($display_data) {
557 if ($field eq "headline") {
558 $xml .= xmlify("headline", $display_data);
559 } else {
560
561 # fallback to empty field name if needed
562 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
563 $html .= "#-#".$display_data."###\n";
564 }
565 }
566 if ($swish_data) {
567 # remove extra spaces
568 $swish_data =~ s/ +/ /g;
569 $swish_data =~ s/ +$//g;
570
571 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
572 }
573
574 if ($swish_exact_data) {
575 $swish_exact_data =~ s/ +/ /g;
576 $swish_exact_data =~ s/ +$//g;
577
578 # add delimiters before and after word.
579 # That is required to produce exact match
580 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
581 }
582 }
583 }
584
585 # dump formatted output in <html>
586 if ($html) {
587 #$xml .= xmlify("html",$html);
588 $xml .= "<html><![CDATA[ $html ]]></html>";
589 }
590
591 if ($xml) {
592 $xml .= $add_xml if ($add_xml);
593 return "<xml>\n$xml</xml>\n";
594 } else {
595 return;
596 }
597 }
598
599 ##########################################################################
600
601 # read configuration for this script
602 my $cfg = new Config::IniFiles( -file => $config_file );
603
604 # read global.conf configuration
605 my $cfg_global = new Config::IniFiles( -file => 'global.conf' );
606
607 # open index
608 $index = new index_DBI(
609 $cfg_global->val('global', 'dbi_dbd'),
610 $cfg_global->val('global', 'dbi_dsn'),
611 $cfg_global->val('global', 'dbi_user'),
612 $cfg_global->val('global', 'dbi_passwd') || '',
613 );
614
615 my $show_progress = $cfg_global->val('global', 'show_progress');
616
617 my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter');
618 if ($my_unac_filter) {
619 print STDERR "using $my_unac_filter to filter characters for search\n";
620 require $my_unac_filter;
621 } else {
622 print STDERR "### fallback to default my_unac_string!\n";
623 eval q{
624 sub main::my_unac_string($$) {
625 my ($charset, $string) = (@_);
626 return $string;
627 }
628 };
629 }
630
631 foreach my $database ($cfg->Sections) {
632
633 my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
634 my $add_xml = $cfg -> val($database, 'xml'); # optional
635
636 # create new lookup file
637 my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
638 if ($lookup_file) {
639 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
640 if (! -e $lookup_file) {
641 open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!";
642 close(LOOKUP);
643 }
644 tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
645 print STDERR "creating lookup file '$lookup_file'\n";
646 # delete memory cache for lookup file
647 delete $cache->{lhash};
648 }
649
650 # open existing lookup file
651 $lookup_file = $cfg -> val($database, 'lookup_open'); # optional
652 if ($lookup_file) {
653 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644;
654 tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644;
655 print STDERR "opening lookup file '$lookup_file'\n";
656 }
657
658 print STDERR "reading ./import_xml/$type.xml\n";
659
660 # extract just type basic
661 my $type_base = $type;
662 $type_base =~ s/_.+$//g;
663
664 $config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 );
665
666 # output current progress indicator
667 my $last_p = 0;
668 sub progress {
669 return if (! $show_progress);
670 my $current = shift;
671 my $total = shift || 1;
672 my $p = int($current * 100 / $total);
673 if ($p != $last_p) {
674 printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p );
675 $last_p = $p;
676 }
677 }
678
679 my $fake_dir = 1;
680 sub fakeprogress {
681 return if (! $show_progress);
682 my $current = shift @_;
683
684 my @ind = ('-','\\','|','/','-','\\','|','/', '-');
685
686 $last_p += $fake_dir;
687 $fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0);
688 if ($last_p % 10 == 0) {
689 printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]);
690 }
691 }
692
693 # now read database
694 print STDERR "using: $type...\n";
695
696 # erase cache for tags by order in this database
697 delete $cache->{tags_by_order};
698
699 if ($type_base eq "isis") {
700
701 my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
702
703 $import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
704 my $db = OpenIsis::open( $isis_db );
705
706 # check if .txt database for OpenIsis is zero length,
707 # if so, erase it and re-open database
708 sub check_txt_db {
709 my $isis_db = shift || die "need isis database name";
710 my $reopen = 0;
711
712 if (-e $isis_db.".TXT") {
713 print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n";
714 unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!";
715 $reopen++;
716 }
717 if (-e $isis_db.".PTR") {
718 print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n";
719 unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!";
720 $reopen++;
721 }
722 return OpenIsis::open( $isis_db ) if ($reopen);
723 }
724
725 # EOF error
726 if ($db == -1) {
727 $db = check_txt_db($isis_db);
728 if ($db == -1) {
729 print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n";
730 next;
731 }
732 }
733
734 # OpenIsis::ERR_BADF
735 if ($db == -4) {
736 print STDERR "FATAL: OpenIsis can't find file $isis_db\n";
737 next;
738 # OpenIsis::ERR_IO
739 } elsif ($db == -5) {
740 print STDERR "FATAL: OpenIsis can't access file $isis_db\n";
741 next;
742 } elsif ($db < 0) {
743 print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n";
744 next;
745 }
746
747 my $max_rowid = OpenIsis::maxRowid( $db );
748
749 # if 0 records, try to rease isis .txt database
750 if ($max_rowid == 0) {
751 # force removal of database
752 $db = check_txt_db($isis_db);
753 $max_rowid = OpenIsis::maxRowid( $db );
754 }
755
756 print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
757
758 my $path = $database;
759
760 for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
761 my $row = OpenIsis::read( $db, $row_id );
762 if ($row && $row->{mfn}) {
763
764 progress($row->{mfn}, $max_rowid);
765
766 my $swishpath = $path."#".int($row->{mfn});
767
768 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
769 $xml = $cp2utf->convert($xml);
770 use bytes; # as opposed to chars
771 print "Path-Name: $swishpath\n";
772 print "Content-Length: ".(length($xml)+1)."\n";
773 print "Document-Type: XML\n\n$xml\n";
774 }
775 }
776 }
777 # for this to work with current version of OpenIsis (0.9.0)
778 # you might need my patch from
779 # http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff
780 OpenIsis::close($db);
781 print STDERR "\n";
782
783 } elsif ($type_base eq "excel") {
784 require Spreadsheet::ParseExcel;
785 require Spreadsheet::ParseExcel::Utility;
786 import Spreadsheet::ParseExcel::Utility qw(int2col);
787
788 $import2cp = Text::Iconv->new($config->{excel_codepage},$codepage);
789 my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
790
791 my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
792 my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml";
793
794 my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
795
796 my $sheet_nr = 0;
797 foreach my $oWks (@{$oBook->{Worksheet}}) {
798 #print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n";
799 last if ($oWks->{Name} eq $sheet);
800 $sheet_nr++;
801 }
802
803 my $oWorksheet = $oBook->{Worksheet}[$sheet_nr];
804
805 print STDERR "using sheet: ",$oWorksheet->{Name},"\n";
806 defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file";
807 my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow};
808
809 for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) {
810 my $row;
811 for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) {
812 my $cell = $oWorksheet->{Cells}[$iR][$iC];
813 if ($cell) {
814 $row->{int2col($iC)} = $cell->Value;
815 }
816 }
817
818 progress($iR, $end_row);
819
820 # print "row[$iR/$end_row] ";
821 # foreach (keys %{$row}) {
822 # print "$_: ",$row->{$_},"\t";
823 # }
824 # print "\n";
825
826 my $swishpath = $database."#".$iR;
827
828 next if (! $row);
829
830 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
831 $xml = $cp2utf->convert($xml);
832 use bytes; # as opposed to chars
833 print "Path-Name: $swishpath\n";
834 print "Content-Length: ".(length($xml)+1)."\n";
835 print "Document-Type: XML\n\n$xml\n";
836 }
837 }
838 } elsif ($type_base eq "marc") {
839
840 require MARC;
841
842 $import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
843 my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
844
845 # optional argument is format
846 my $format = x($config->{marc_format}) || 'usmarc';
847
848 print STDERR "Reading MARC file '$marc_file'\n";
849
850 my $marc = new MARC;
851 my $nr = $marc->openmarc({
852 file=>$marc_file, format=>$format
853 }) || die "Can't open MARC file '$marc_file' with format '$format'";
854
855 # read MARC file in memory
856 $marc->nextmarc(-1);
857
858 my $max_rec = $marc->marc_count();
859
860 for(my $i=1; $i<=$max_rec; $i++) {
861
862 progress($i,$max_rec);
863
864 # store value for marc_sf.pm
865 $main::cache->{marc_record} = $i;
866
867 my $swishpath = $database."#".$i;
868
869 if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) {
870 $xml = $cp2utf->convert($xml);
871 use bytes; # as opposed to chars
872 print "Path-Name: $swishpath\n";
873 print "Content-Length: ".(length($xml)+1)."\n";
874 print "Document-Type: XML\n\n$xml\n";
875 }
876 }
877
878 print STDERR "\n";
879
880 } elsif ($type_base eq "feed") {
881
882 $import2cp = Text::Iconv->new($config->{feed_codepage},$codepage);
883 my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!";
884
885 print STDERR "Reading feed from program '$prog'\n";
886
887 open(FEED,"feeds/$prog |") || die "can't start $prog: $!";
888
889 my $i=1; # record nr.
890
891 my $data;
892 my $line=1;
893
894 while (<FEED>) {
895 chomp;
896
897 if (/^$/) {
898 my $swishpath = $database."#".$i++;
899
900 if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) {
901 $xml = $cp2utf->convert($xml);
902 use bytes; # as opposed to chars
903 print "Path-Name: $swishpath\n";
904 print "Content-Length: ".(length($xml)+1)."\n";
905 print "Document-Type: XML\n\n$xml\n";
906 }
907 $line = 1;
908 $data = {};
909 next;
910 }
911
912 $line = $1 if (s/^(\d+):\s*//);
913 $data->{$line++} = $_;
914
915 fakeprogress($i);
916
917 }
918 # close lookup
919 untie %lhash if (%lhash);
920 }
921 }
922
923 # call this to commit index
924 $index->close;
925
926 1;
927 __END__
928 ##########################################################################
929
930 =head1 NAME
931
932 all2xml.pl - read various file formats and dump XML for SWISH-E
933
934 =head1 SYNOPSYS
935
936 $ all2xml.pl [test.conf]
937
938 =head1 DESCRIPTION
939
940 This command will read ISIS data file using OpenIsis perl module, MARC
941 records using MARC module and optionally Micro$oft Excel files to
942 create one XML file for usage with I<SWISH-E> indexer. Dispite it's name,
943 this script B<isn't general xml generator> from isis files (isis allready
944 has something like that). Output of this script is tailor-made for SWISH-E.
945
946 If no configuration file is specified, it will use default one called
947 C<all2xml.conf>.
948
949 =head1 BUGS
950
951 Documentation is really lacking. However, in true Open Source spirit, source
952 is best documentation. I even made considerable effort to comment parts
953 which are not intuitively clear, so...
954
955 =head1 AUTHOR
956
957 Dobrica Pavlinusic <dpavlin@rot13.org>
958
959 =head1 COPYRIGHT
960
961 GNU Public License (GPL) v2 or later
962
963 =head1 SEE ALSO
964
965 SWISH-E web site at http://www.swish-e.org
966
967 =cut

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.26