/[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 224 - (show annotations)
Sun Feb 8 20:16:54 2004 UTC (20 years, 1 month ago) by dpavlin
Original Path: trunk/all2xml.pl
File MIME type: text/plain
File size: 25490 byte(s)
important bug fix for bug introduced in 1.57: it might eat your data
if you are not using filter. This one was hard do find...

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::Unaccent 1.02; # 1.01 won't compile on my platform,
9 use Text::Iconv;
10 use Config::IniFiles;
11 use Encode;
12 #use GDBM_File;
13 use Fcntl; # for O_RDWR
14 use TDB_File;
15
16 $|=1;
17
18 my $config_file = $0;
19 $config_file =~ s/\.pl$/.conf/;
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 # begin real work: go field by field
155 foreach my $field (@sorted_tags) {
156
157 $field=x($field);
158 $field_usage{$field}++;
159
160 my $swish_data = "";
161 my $swish_exact_data = "";
162 my $display_data = "";
163 my @index_data;
164 my $line_delimiter;
165
166 my ($swish,$display);
167
168 my $tag = $type2tag{$type} || die "can't find which tag to use for type $type";
169
170 # is this field page-by-page?
171 my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page};
172 push @page_fields,$field if ($iterate_by_page);
173 my %page_max = ();
174 # default line_delimiter if using
175 my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>';
176 $cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter};
177
178 my $format_name = $config->{indexer}->{$field}->{format_name};
179 my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter};
180 if ($format_name && $format_delimiter) {
181 $cache->{format}->{$field}->{format_name} = $format_name;
182 $cache->{format}->{$field}->{format_delimiter} = $format_delimiter;
183 }
184
185 foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) {
186
187 my $format = x($x->{content});
188 my $delimiter = x($x->{delimiter}) || ' ';
189
190 my $repeat_off = 0; # init repeatable offset
191
192 # swish, swish_exact, display, index, index_lookup
193 # swish and display defaults
194 my ($s,$se,$d,$i,$il) = (1,0,1,0,0);
195 $s = 0 if (lc($x->{type}) eq "display");
196 $d = 0 if (lc($x->{type}) eq "swish");
197 ($s,$se,$d,$i) = (0,1,0,1) if (lc($x->{type}) eq "index");
198 ($s,$se,$d,$i) = (0,1,0,0) if (lc($x->{type}) eq "swish_exact");
199 ($s,$se,$d,$i,$il) = (0,1,0,0,1) if (lc($x->{type}) =~ /^lookup/);
200
201 # what will separate last line from this one?
202 if ($display_data && $x->{append}) {
203 $line_delimiter = ' ';
204 } elsif ($display_data) {
205 $line_delimiter = '<br/>';
206 }
207
208 # init vars so that we go into while...
209 ($swish,$display) = (1,1);
210
211 # placeholder for all repeatable entries for index
212
213 sub chk_eval($) {
214 my $data = shift;
215 if ($data =~ s/eval{([^}]+)}//) {
216 if (eval "$1") {
217 return $data;
218 } else {
219 return undef;
220 }
221 } else {
222 return $data;
223 }
224 }
225
226 sub mkformat($$) {
227 my $x = shift || die "mkformat needs tag reference";
228 my $data = shift || return;
229 my $format_name = x($x->{format_name}) || return chk_eval($data);
230 my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!";
231 my $format_delimiter = x($x->{format_delimiter});
232 my @data;
233 if ($format_delimiter) {
234 @data = split(/$format_delimiter/,$data);
235 } else {
236 push @data,$data;
237 }
238
239 if ($fmt) {
240 my $nr = scalar $fmt =~ s/%s/%s/g;
241 if (($#data+1) == $nr) {
242 return chk_eval(sprintf($fmt,@data));
243 } else {
244 #print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n";
245 return chk_eval($data);
246 }
247 } else {
248 print STDERR "usage of link '$format_name' without defined format (<link> tag)\n";
249 }
250 }
251
252 # while because of repeatable fields
253 while ($swish || $display) {
254 my $page = $repeat_off;
255 $page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0));
256 ($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp);
257 if ($repeat_off > 1000) {
258 print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n";
259 last;
260 }
261
262 # is this field is lookup?
263 if ($display && $x->{lookup}) {
264 my $null = "<!-- null -->";
265 if ($use_lhash_cache) {
266 if (!defined($cache->{lhash}->{$display})) {
267 my $new_display = $lhash{$display};
268 if (defined($new_display)) {
269 #print STDERR "lookup cache store '$display' = '$new_display'\n";
270 $display = $new_display;
271 $cache->{lhash}->{$display} = $new_display;
272 } else {
273 print STDERR "WARNING: lookup for '$display' didn't find anything.\n";
274 $display = "";
275 $cache->{lhash}->{$display} = $null;
276 }
277 } else {
278 $display = $cache->{lhash}->{$display};
279 }
280 } else {
281 $display = $lhash{$display} || $null;
282 }
283 }
284
285 # filter="name" ; filter this field through
286 # filter/[name].pm
287 my $filter = $x->{filter};
288 if ($filter && !$cache->{filter_loaded}->{$filter}) {
289 require "filter/".$filter.".pm";
290 $cache->{filter_loaded}->{$filter}++;
291 }
292 # type="swish" ; field for swish
293 if ($swish) {
294 if ($filter && ($s || $se)) {
295 no strict 'refs';
296 my $tmp = join(" ",&$filter($swish)) if ($s || $se);
297 $swish_data .= $tmp if ($s);
298 $swish_exact_data .= "xxbxx $tmp xxexx " if ($se && $tmp ne "");
299
300 } else {
301 $swish_data .= $swish if ($s);
302 $swish_exact_data .= "xxbxx $swish xxexx " if ($se && $swish ne "");
303 }
304 }
305
306 # type="display" ; field for display
307 if ($d && $display) {
308 my $ldel = $delimiter;
309 if ($line_delimiter && $display_data) {
310 $ldel = $line_delimiter;
311 }
312 if ($filter) {
313 no strict 'refs';
314 my @arr;
315 foreach my $tmp (&$filter($display)) {
316 my $tmp2 = mkformat($x,$tmp);
317 push @arr,$tmp2 if ($tmp2);
318 }
319 $display_data .= $ldel if ($display_data && @arr);
320 $display_data .= join($delimiter,@arr);
321 } else {
322 $display_data .= $ldel if ($display_data);
323 my $tmp = mkformat($x,$display);
324 $display_data .= $tmp if ($tmp);
325 }
326 }
327
328 # type="index" ; insert into index
329 my $idisplay;
330 if ($i && $display) {
331 $idisplay = $display;
332 if ($filter) {
333 no strict 'refs';
334 $idisplay = &$filter($idisplay);
335 }
336 push @index_data, $idisplay if (! $iterate_by_page);
337 }
338
339 # store fields in lookup
340 if ($il && $display) {
341 if (lc($x->{type}) eq "lookup_key") {
342 if ($lookup_key) {
343 print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)";
344 } else {
345 $lookup_key = $display;
346 }
347 } elsif (lc($x->{type}) eq "lookup_val") {
348 if ($lookup_key) {
349 $lhash{$lookup_key} = $display;
350 } else {
351 print STDERR "WARNING: no lookup_key defined for '$display'?";
352 }
353 }
354
355 }
356
357 # store data for page-by-page repeatable fields
358 if ($iterate_by_page) {
359 sub iterate_fld($$$$$$) {
360 my ($cache,$what,$field,$page,$data,$append) = @_;
361 return if (!$data);
362
363 my $ldel = $page_line_delimiter;
364 $ldel = " " if ($append);
365 #print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
366 if (! $cache->{$what}->{$field}->[$page]) {
367 $cache->{$what}->{$field}->[$page] = $data;
368 } else {
369 $cache->{$what}->{$field}->[$page] .= $ldel.$data;
370 }
371 }
372
373 if ($display_data) {
374 iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append});
375 }
376 $display_data = "";
377 if ($swish_data) {
378 iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append});
379 $swish_data = "";
380 }
381 if ($swish_exact_data) {
382 iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append});
383 $swish_exact_data = "";
384 }
385
386 if ($idisplay) {
387 my $ldel=$page_line_delimiter;
388 my @index_data;
389 if ($cache->{index_data}->{$field}->[$page]) {
390
391 @index_data = @{$cache->{index_data}->{$field}->[$page]};
392 }
393 if ($x->{append}) {
394 if (@index_data) {
395 $index_data[$#index_data] .= $idisplay;
396 } else {
397 push @index_data, $idisplay;
398 }
399 } else {
400 push @index_data, $idisplay;
401 }
402 $idisplay = "";
403 @{$cache->{index_data}->{$field}->[$page]} = @index_data;
404 }
405 }
406 }
407
408 if (! $iterate_by_page) {
409 my $idel = $x->{index_delimiter};
410 # fill data in index
411 foreach my $tmp (@index_data) {
412 my $i = $d = $tmp;
413 if ($idel && $tmp =~ m/$idel/) {
414 ($i,$d) = split(/$idel/,$tmp);
415 }
416 $index->insert($field, $i, $d, $path);
417 }
418 @index_data = ();
419 }
420 }
421
422 # now try to parse variables from configuration file
423 foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) {
424
425 my $delimiter = x($x->{delimiter}) || ' ';
426 my $val = $cfg->val($database, x($x->{content}));
427
428 my ($s,$d,$i) = (1,1,0); # swish, display default
429 $s = 0 if (lc($x->{type}) eq "display");
430 $d = 0 if (lc($x->{type}) eq "swish");
431 # no support for swish exact in config.
432 # IMHO, it's useless
433 ($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index");
434
435 if ($val) {
436 $display_data .= $delimiter.$val if ($d);
437 $swish_data .= $val if ($s);
438 $index->insert($field, $val, $path) if ($i);
439 }
440
441 if ($iterate_by_page) {
442 # FIXME data from config tag will appear just
443 # on first page!!!
444 my $page = 0;
445 if ($display_data) {
446 $cache->{display_data}->{$field}->[$page] = $display_data;
447 $display_data = "";
448 }
449 if ($swish_data) {
450 $cache->{swish_data}->{$field}->[$page] = $swish_data;
451 $swish_data = "";
452 }
453 if ($swish_exact_data) {
454 $cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data;
455 $swish_exact_data = "";
456 }
457 }
458 }
459
460 # save data page-by-page
461 foreach my $field (@page_fields) {
462 my $nr_pages = $page_max{$field} || next;
463 #print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
464 #print STDERR Dumper($cache->{display_data});
465 for (my $page=0; $page <= $nr_pages; $page++) {
466 my $display_data;
467 if ($cache->{format}->{$field}) {
468 my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]);
469 $display_data=$tmp if ($tmp);
470 } else {
471 $display_data = $cache->{display_data}->{$field}->[$page];
472 }
473 if ($display_data) { # default
474 if ($field eq "headline") {
475 $xml .= xmlify("headline", $display_data);
476 } else {
477
478 # fallback to empty field name if needed
479 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
480 $html .= "#-#".$display_data."###\n";
481 }
482 }
483
484 my $swish_data = $cache->{swish_data}->{$field}->[$page];
485 if ($swish_data) {
486 # remove extra spaces
487 $swish_data =~ s/ +/ /g;
488 $swish_data =~ s/ +$//g;
489
490 $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));
491 }
492
493 my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
494 if ($swish_exact_data) {
495 $swish_exact_data =~ s/ +/ /g;
496 $swish_exact_data =~ s/ +$//g;
497
498 # add delimiters before and after word.
499 # That is required to produce exact match
500 $xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data));
501 }
502
503 my $idel = $cache->{index_delimiter}->{$field};
504 foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) {
505 my $i = $tmp;
506 my $d = $tmp;
507 if ($idel && $tmp =~ m/$idel/) {
508 ($i,$d) = split(/$idel/,$tmp);
509 }
510 $index->insert($field, $i, $d, $path);
511 #print STDERR "index [$idel] $field: $i --> $d [$path]\n";
512 }
513 }
514
515 }
516
517 if (! $iterate_by_page) {
518 if ($display_data) {
519 if ($field eq "headline") {
520 $xml .= xmlify("headline", $display_data);
521 } else {
522
523 # fallback to empty field name if needed
524 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
525 $html .= "#-#".$display_data."###\n";
526 }
527 }
528 if ($swish_data) {
529 # remove extra spaces
530 $swish_data =~ s/ +/ /g;
531 $swish_data =~ s/ +$//g;
532
533 $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));
534 }
535
536 if ($swish_exact_data) {
537 $swish_exact_data =~ s/ +/ /g;
538 $swish_exact_data =~ s/ +$//g;
539
540 # add delimiters before and after word.
541 # That is required to produce exact match
542 $xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data));
543 }
544 }
545 }
546
547 # dump formatted output in <html>
548 if ($html) {
549 #$xml .= xmlify("html",$html);
550 $xml .= "<html><![CDATA[ $html ]]></html>";
551 }
552
553 if ($xml) {
554 $xml .= $add_xml if ($add_xml);
555 return "<xml>\n$xml</xml>\n";
556 } else {
557 return;
558 }
559 }
560
561 ##########################################################################
562
563 # read configuration for this script
564 my $cfg = new Config::IniFiles( -file => $config_file );
565
566 # read global.conf configuration
567 my $cfg_global = new Config::IniFiles( -file => 'global.conf' );
568
569 # open index
570 $index = new index_DBI(
571 $cfg_global->val('global', 'dbi_dbd'),
572 $cfg_global->val('global', 'dbi_dsn'),
573 $cfg_global->val('global', 'dbi_user'),
574 $cfg_global->val('global', 'dbi_passwd') || '',
575 );
576
577 my $show_progress = $cfg_global->val('global', 'show_progress');
578
579 my $unac_filter = $cfg_global->val('global', 'unac_filter');
580 if ($unac_filter) {
581 require $unac_filter;
582 }
583
584 foreach my $database ($cfg->Sections) {
585
586 my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
587 my $add_xml = $cfg -> val($database, 'xml'); # optional
588
589 # create new lookup file
590 my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
591 if ($lookup_file) {
592 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
593 tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
594 print STDERR "creating lookup file '$lookup_file'\n";
595 }
596
597 # open existing lookup file
598 $lookup_file = $cfg -> val($database, 'lookup_open'); # optional
599 if ($lookup_file) {
600 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644;
601 tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644;
602 print STDERR "opening lookup file '$lookup_file'\n";
603 }
604
605 print STDERR "reading ./import_xml/$type.xml\n";
606
607 # extract just type basic
608 my $type_base = $type;
609 $type_base =~ s/_.+$//g;
610
611 $config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1);
612
613 # output current progress indicator
614 my $last_p = 0;
615 sub progress {
616 return if (! $show_progress);
617 my $current = shift;
618 my $total = shift || 1;
619 my $p = int($current * 100 / $total);
620 if ($p != $last_p) {
621 printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p );
622 $last_p = $p;
623 }
624 }
625
626 my $fake_dir = 1;
627 sub fakeprogress {
628 return if (! $show_progress);
629 my $current = shift @_;
630
631 my @ind = ('-','\\','|','/','-','\\','|','/', '-');
632
633 $last_p += $fake_dir;
634 $fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0);
635 if ($last_p % 10 == 0) {
636 printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]);
637 }
638 }
639
640 # now read database
641 print STDERR "using: $type...\n";
642
643 # erase cache for tags by order in this database
644 delete $cache->{tags_by_order};
645
646 if ($type_base eq "isis") {
647
648 my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
649
650 $import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
651 my $db = OpenIsis::open( $isis_db );
652
653 # check if .txt database for OpenIsis is zero length,
654 # if so, erase it and re-open database
655 sub check_txt_db {
656 my $isis_db = shift || die "need isis database name";
657 my $reopen = 0;
658
659 if (-e $isis_db.".TXT") {
660 print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n";
661 unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!";
662 $reopen++;
663 }
664 if (-e $isis_db.".PTR") {
665 print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n";
666 unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!";
667 $reopen++;
668 }
669 return OpenIsis::open( $isis_db ) if ($reopen);
670 }
671
672 # EOF error
673 if ($db == -1) {
674 $db = check_txt_db($isis_db);
675 if ($db == -1) {
676 print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n";
677 next;
678 }
679 }
680
681 # OpenIsis::ERR_BADF
682 if ($db == -4) {
683 print STDERR "FATAL: OpenIsis can't find file $isis_db\n";
684 next;
685 # OpenIsis::ERR_IO
686 } elsif ($db == -5) {
687 print STDERR "FATAL: OpenIsis can't access file $isis_db\n";
688 next;
689 } elsif ($db < 0) {
690 print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n";
691 next;
692 }
693
694 my $max_rowid = OpenIsis::maxRowid( $db );
695
696 # if 0 records, try to rease isis .txt database
697 if ($max_rowid == 0) {
698 # force removal of database
699 $db = check_txt_db($isis_db);
700 $max_rowid = OpenIsis::maxRowid( $db );
701 }
702
703 print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
704
705 my $path = $database;
706
707 for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
708 my $row = OpenIsis::read( $db, $row_id );
709 if ($row && $row->{mfn}) {
710
711 progress($row->{mfn}, $max_rowid);
712
713 my $swishpath = $path."#".int($row->{mfn});
714
715 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
716 $xml = $cp2utf->convert($xml);
717 use bytes; # as opposed to chars
718 print "Path-Name: $swishpath\n";
719 print "Content-Length: ".(length($xml)+1)."\n";
720 print "Document-Type: XML\n\n$xml\n";
721 }
722 }
723 }
724 # for this to work with current version of OpenIsis (0.9.0)
725 # you might need my patch from
726 # http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff
727 OpenIsis::close($db);
728 print STDERR "\n";
729
730 } elsif ($type_base eq "excel") {
731 use Spreadsheet::ParseExcel;
732 use Spreadsheet::ParseExcel::Utility qw(int2col);
733
734 $import2cp = Text::Iconv->new($config->{excel_codepage},$codepage);
735 my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
736
737 my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
738 my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml";
739
740 my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
741
742 my $sheet_nr = 0;
743 foreach my $oWks (@{$oBook->{Worksheet}}) {
744 #print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n";
745 last if ($oWks->{Name} eq $sheet);
746 $sheet_nr++;
747 }
748
749 my $oWorksheet = $oBook->{Worksheet}[$sheet_nr];
750
751 print STDERR "using sheet: ",$oWorksheet->{Name},"\n";
752 defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file";
753 my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow};
754
755 for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) {
756 my $row;
757 for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) {
758 my $cell = $oWorksheet->{Cells}[$iR][$iC];
759 if ($cell) {
760 $row->{int2col($iC)} = $cell->Value;
761 }
762 }
763
764 progress($iR, $end_row);
765
766 # print "row[$iR/$end_row] ";
767 # foreach (keys %{$row}) {
768 # print "$_: ",$row->{$_},"\t";
769 # }
770 # print "\n";
771
772 my $swishpath = $database."#".$iR;
773
774 next if (! $row);
775
776 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
777 $xml = $cp2utf->convert($xml);
778 use bytes; # as opposed to chars
779 print "Path-Name: $swishpath\n";
780 print "Content-Length: ".(length($xml)+1)."\n";
781 print "Document-Type: XML\n\n$xml\n";
782 }
783 }
784 } elsif ($type_base eq "marc") {
785
786 use MARC;
787
788 $import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
789 my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
790
791 # optional argument is format
792 my $format = x($config->{marc_format}) || 'usmarc';
793
794 print STDERR "Reading MARC file '$marc_file'\n";
795
796 my $marc = new MARC;
797 my $nr = $marc->openmarc({
798 file=>$marc_file, format=>$format
799 }) || die "Can't open MARC file '$marc_file' with format '$format'";
800
801 # read MARC file in memory
802 $marc->nextmarc(-1);
803
804 my $max_rec = $marc->marc_count();
805
806 for(my $i=1; $i<=$max_rec; $i++) {
807
808 progress($i,$max_rec);
809
810 # store value for marc_sf.pm
811 $main::cache->{marc_record} = $i;
812
813 my $swishpath = $database."#".$i;
814
815 if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) {
816 $xml = $cp2utf->convert($xml);
817 use bytes; # as opposed to chars
818 print "Path-Name: $swishpath\n";
819 print "Content-Length: ".(length($xml)+1)."\n";
820 print "Document-Type: XML\n\n$xml\n";
821 }
822 }
823
824 print STDERR "\n";
825
826 } elsif ($type_base eq "feed") {
827
828 $import2cp = Text::Iconv->new($config->{feed_codepage},$codepage);
829 my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!";
830
831 print STDERR "Reading feed from program '$prog'\n";
832
833 open(FEED,"feeds/$prog |") || die "can't start $prog: $!";
834
835 my $i=1; # record nr.
836
837 my $data;
838 my $line=1;
839
840 while (<FEED>) {
841 chomp;
842
843 if (/^$/) {
844 my $swishpath = $database."#".$i++;
845
846 if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) {
847 $xml = $cp2utf->convert($xml);
848 use bytes; # as opposed to chars
849 print "Path-Name: $swishpath\n";
850 print "Content-Length: ".(length($xml)+1)."\n";
851 print "Document-Type: XML\n\n$xml\n";
852 }
853 $line = 1;
854 $data = {};
855 next;
856 }
857
858 $line = $1 if (s/^(\d+):\s*//);
859 $data->{$line++} = $_;
860
861 fakeprogress($i);
862
863 }
864 # close lookup
865 untie %lhash if (%lhash);
866 }
867 }
868
869 # call this to commit index
870 $index->close;
871
872 1;
873 __END__
874 ##########################################################################
875
876 =head1 NAME
877
878 all2xml.pl - read various file formats and dump XML for SWISH-E
879
880 =head1 DESCRIPTION
881
882 This command will read ISIS data file using OpenIsis perl module, MARC
883 records using MARC module and optionally Micro$oft Excel files to
884 create one XML file for usage with I<SWISH-E> indexer. Dispite it's name,
885 this script B<isn't general xml generator> from isis files (isis allready
886 has something like that). Output of this script is tailor-made for SWISH-E.
887
888 =head1 BUGS
889
890 Documentation is really lacking. However, in true Open Source spirit, source
891 is best documentation. I even made considerable effort to comment parts
892 which are not intuitively clear, so...
893
894 =head1 AUTHOR
895
896 Dobrica Pavlinusic <dpavlin@rot13.org>
897
898 =head1 COPYRIGHT
899
900 GNU Public License (GPL) v2 or later
901
902 =head1 SEE ALSO
903
904 SWISH-E web site at http://www.swish-e.org
905
906 =cut

Properties

Name Value
cvs2svn:cvs-rev 1.61
svn:executable *

  ViewVC Help
Powered by ViewVC 1.1.26