/[webpac]/trunk/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 /trunk/all2xml.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 233 - (show annotations)
Fri Mar 5 23:33:19 2004 UTC (20 years ago) by dpavlin
File MIME type: text/plain
File size: 25796 byte(s)
lookup_key and lookup_val types now support filters

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 ($lookup_key) {
343 print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)";
344 } else {
345 if ($filter) {
346 no strict 'refs';
347 $lookup_key = &$filter($display);
348 } else {
349 $lookup_key = $display;
350 }
351 }
352 } elsif (lc($x->{type}) eq "lookup_val") {
353 if ($lookup_key) {
354 if ($filter) {
355 no strict 'refs';
356 $lhash{$lookup_key} = &$filter($display);
357 } else {
358 $lhash{$lookup_key} = $display;
359 }
360 } else {
361 print STDERR "WARNING: no lookup_key defined for '$display'?";
362 }
363 }
364
365 }
366
367 # store data for page-by-page repeatable fields
368 if ($iterate_by_page) {
369 sub iterate_fld($$$$$$) {
370 my ($cache,$what,$field,$page,$data,$append) = @_;
371 return if (!$data);
372
373 my $ldel = $page_line_delimiter;
374 $ldel = " " if ($append);
375 #print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
376 if (! $cache->{$what}->{$field}->[$page]) {
377 $cache->{$what}->{$field}->[$page] = $data;
378 } else {
379 $cache->{$what}->{$field}->[$page] .= $ldel.$data;
380 }
381 }
382
383 if ($display_data) {
384 iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append});
385 }
386 $display_data = "";
387 if ($swish_data) {
388 iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append});
389 $swish_data = "";
390 }
391 if ($swish_exact_data) {
392 iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append});
393 $swish_exact_data = "";
394 }
395
396 if ($idisplay) {
397 my $ldel=$page_line_delimiter;
398 my @index_data;
399 if ($cache->{index_data}->{$field}->[$page]) {
400
401 @index_data = @{$cache->{index_data}->{$field}->[$page]};
402 }
403 if ($x->{append}) {
404 if (@index_data) {
405 $index_data[$#index_data] .= $idisplay;
406 } else {
407 push @index_data, $idisplay;
408 }
409 } else {
410 push @index_data, $idisplay;
411 }
412 $idisplay = "";
413 @{$cache->{index_data}->{$field}->[$page]} = @index_data;
414 }
415 }
416 }
417
418 if (! $iterate_by_page) {
419 my $idel = $x->{index_delimiter};
420 # fill data in index
421 foreach my $tmp (@index_data) {
422 my $i = $d = $tmp;
423 if ($idel && $tmp =~ m/$idel/) {
424 ($i,$d) = split(/$idel/,$tmp);
425 }
426 $index->insert($field, $i, $d, $path);
427 }
428 @index_data = ();
429 }
430 }
431
432 # now try to parse variables from configuration file
433 foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) {
434
435 my $delimiter = x($x->{delimiter}) || ' ';
436 my $val = $cfg->val($database, x($x->{content}));
437
438 my ($s,$d,$i) = (1,1,0); # swish, display default
439 $s = 0 if (lc($x->{type}) eq "display");
440 $d = 0 if (lc($x->{type}) eq "swish");
441 # no support for swish exact in config.
442 # IMHO, it's useless
443 ($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index");
444
445 if ($val) {
446 $display_data .= $delimiter.$val if ($d);
447 $swish_data .= $val if ($s);
448 $index->insert($field, $val, $path) if ($i);
449 }
450
451 if ($iterate_by_page) {
452 # FIXME data from config tag will appear just
453 # on first page!!!
454 my $page = 0;
455 if ($display_data) {
456 $cache->{display_data}->{$field}->[$page] = $display_data;
457 $display_data = "";
458 }
459 if ($swish_data) {
460 $cache->{swish_data}->{$field}->[$page] = $swish_data;
461 $swish_data = "";
462 }
463 if ($swish_exact_data) {
464 $cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data;
465 $swish_exact_data = "";
466 }
467 }
468 }
469
470 # save data page-by-page
471 foreach my $field (@page_fields) {
472 my $nr_pages = $page_max{$field} || next;
473 #print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
474 #print STDERR Dumper($cache->{display_data});
475 for (my $page=0; $page <= $nr_pages; $page++) {
476 my $display_data;
477 if ($cache->{format}->{$field}) {
478 my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]);
479 $display_data=$tmp if ($tmp);
480 } else {
481 $display_data = $cache->{display_data}->{$field}->[$page];
482 }
483 if ($display_data) { # default
484 if ($field eq "headline") {
485 $xml .= xmlify("headline", $display_data);
486 } else {
487
488 # fallback to empty field name if needed
489 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
490 $html .= "#-#".$display_data."###\n";
491 }
492 }
493
494 my $swish_data = $cache->{swish_data}->{$field}->[$page];
495 if ($swish_data) {
496 # remove extra spaces
497 $swish_data =~ s/ +/ /g;
498 $swish_data =~ s/ +$//g;
499
500 $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));
501 }
502
503 my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
504 if ($swish_exact_data) {
505 $swish_exact_data =~ s/ +/ /g;
506 $swish_exact_data =~ s/ +$//g;
507
508 # add delimiters before and after word.
509 # That is required to produce exact match
510 $xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data));
511 }
512
513 my $idel = $cache->{index_delimiter}->{$field};
514 foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) {
515 my $i = $tmp;
516 my $d = $tmp;
517 if ($idel && $tmp =~ m/$idel/) {
518 ($i,$d) = split(/$idel/,$tmp);
519 }
520 $index->insert($field, $i, $d, $path);
521 #print STDERR "index [$idel] $field: $i --> $d [$path]\n";
522 }
523 }
524
525 }
526
527 if (! $iterate_by_page) {
528 if ($display_data) {
529 if ($field eq "headline") {
530 $xml .= xmlify("headline", $display_data);
531 } else {
532
533 # fallback to empty field name if needed
534 $html .= get_field_name($config,$field,$field_usage{$field}) || '';
535 $html .= "#-#".$display_data."###\n";
536 }
537 }
538 if ($swish_data) {
539 # remove extra spaces
540 $swish_data =~ s/ +/ /g;
541 $swish_data =~ s/ +$//g;
542
543 $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));
544 }
545
546 if ($swish_exact_data) {
547 $swish_exact_data =~ s/ +/ /g;
548 $swish_exact_data =~ s/ +$//g;
549
550 # add delimiters before and after word.
551 # That is required to produce exact match
552 $xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data));
553 }
554 }
555 }
556
557 # dump formatted output in <html>
558 if ($html) {
559 #$xml .= xmlify("html",$html);
560 $xml .= "<html><![CDATA[ $html ]]></html>";
561 }
562
563 if ($xml) {
564 $xml .= $add_xml if ($add_xml);
565 return "<xml>\n$xml</xml>\n";
566 } else {
567 return;
568 }
569 }
570
571 ##########################################################################
572
573 # read configuration for this script
574 my $cfg = new Config::IniFiles( -file => $config_file );
575
576 # read global.conf configuration
577 my $cfg_global = new Config::IniFiles( -file => 'global.conf' );
578
579 # open index
580 $index = new index_DBI(
581 $cfg_global->val('global', 'dbi_dbd'),
582 $cfg_global->val('global', 'dbi_dsn'),
583 $cfg_global->val('global', 'dbi_user'),
584 $cfg_global->val('global', 'dbi_passwd') || '',
585 );
586
587 my $show_progress = $cfg_global->val('global', 'show_progress');
588
589 my $unac_filter = $cfg_global->val('global', 'unac_filter');
590 if ($unac_filter) {
591 require $unac_filter;
592 }
593
594 foreach my $database ($cfg->Sections) {
595
596 my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
597 my $add_xml = $cfg -> val($database, 'xml'); # optional
598
599 # create new lookup file
600 my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
601 if ($lookup_file) {
602 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
603 tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
604 print STDERR "creating lookup file '$lookup_file'\n";
605 # delete memory cache for lookup file
606 delete $cache->{lhash};
607 }
608
609 # open existing lookup file
610 $lookup_file = $cfg -> val($database, 'lookup_open'); # optional
611 if ($lookup_file) {
612 #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644;
613 tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644;
614 print STDERR "opening lookup file '$lookup_file'\n";
615 }
616
617 print STDERR "reading ./import_xml/$type.xml\n";
618
619 # extract just type basic
620 my $type_base = $type;
621 $type_base =~ s/_.+$//g;
622
623 $config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1);
624
625 # output current progress indicator
626 my $last_p = 0;
627 sub progress {
628 return if (! $show_progress);
629 my $current = shift;
630 my $total = shift || 1;
631 my $p = int($current * 100 / $total);
632 if ($p != $last_p) {
633 printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p );
634 $last_p = $p;
635 }
636 }
637
638 my $fake_dir = 1;
639 sub fakeprogress {
640 return if (! $show_progress);
641 my $current = shift @_;
642
643 my @ind = ('-','\\','|','/','-','\\','|','/', '-');
644
645 $last_p += $fake_dir;
646 $fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0);
647 if ($last_p % 10 == 0) {
648 printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]);
649 }
650 }
651
652 # now read database
653 print STDERR "using: $type...\n";
654
655 # erase cache for tags by order in this database
656 delete $cache->{tags_by_order};
657
658 if ($type_base eq "isis") {
659
660 my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
661
662 $import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
663 my $db = OpenIsis::open( $isis_db );
664
665 # check if .txt database for OpenIsis is zero length,
666 # if so, erase it and re-open database
667 sub check_txt_db {
668 my $isis_db = shift || die "need isis database name";
669 my $reopen = 0;
670
671 if (-e $isis_db.".TXT") {
672 print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n";
673 unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!";
674 $reopen++;
675 }
676 if (-e $isis_db.".PTR") {
677 print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n";
678 unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!";
679 $reopen++;
680 }
681 return OpenIsis::open( $isis_db ) if ($reopen);
682 }
683
684 # EOF error
685 if ($db == -1) {
686 $db = check_txt_db($isis_db);
687 if ($db == -1) {
688 print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n";
689 next;
690 }
691 }
692
693 # OpenIsis::ERR_BADF
694 if ($db == -4) {
695 print STDERR "FATAL: OpenIsis can't find file $isis_db\n";
696 next;
697 # OpenIsis::ERR_IO
698 } elsif ($db == -5) {
699 print STDERR "FATAL: OpenIsis can't access file $isis_db\n";
700 next;
701 } elsif ($db < 0) {
702 print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n";
703 next;
704 }
705
706 my $max_rowid = OpenIsis::maxRowid( $db );
707
708 # if 0 records, try to rease isis .txt database
709 if ($max_rowid == 0) {
710 # force removal of database
711 $db = check_txt_db($isis_db);
712 $max_rowid = OpenIsis::maxRowid( $db );
713 }
714
715 print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
716
717 my $path = $database;
718
719 for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
720 my $row = OpenIsis::read( $db, $row_id );
721 if ($row && $row->{mfn}) {
722
723 progress($row->{mfn}, $max_rowid);
724
725 my $swishpath = $path."#".int($row->{mfn});
726
727 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
728 $xml = $cp2utf->convert($xml);
729 use bytes; # as opposed to chars
730 print "Path-Name: $swishpath\n";
731 print "Content-Length: ".(length($xml)+1)."\n";
732 print "Document-Type: XML\n\n$xml\n";
733 }
734 }
735 }
736 # for this to work with current version of OpenIsis (0.9.0)
737 # you might need my patch from
738 # http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff
739 OpenIsis::close($db);
740 print STDERR "\n";
741
742 } elsif ($type_base eq "excel") {
743 use Spreadsheet::ParseExcel;
744 use Spreadsheet::ParseExcel::Utility qw(int2col);
745
746 $import2cp = Text::Iconv->new($config->{excel_codepage},$codepage);
747 my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
748
749 my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
750 my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml";
751
752 my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
753
754 my $sheet_nr = 0;
755 foreach my $oWks (@{$oBook->{Worksheet}}) {
756 #print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n";
757 last if ($oWks->{Name} eq $sheet);
758 $sheet_nr++;
759 }
760
761 my $oWorksheet = $oBook->{Worksheet}[$sheet_nr];
762
763 print STDERR "using sheet: ",$oWorksheet->{Name},"\n";
764 defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file";
765 my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow};
766
767 for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) {
768 my $row;
769 for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) {
770 my $cell = $oWorksheet->{Cells}[$iR][$iC];
771 if ($cell) {
772 $row->{int2col($iC)} = $cell->Value;
773 }
774 }
775
776 progress($iR, $end_row);
777
778 # print "row[$iR/$end_row] ";
779 # foreach (keys %{$row}) {
780 # print "$_: ",$row->{$_},"\t";
781 # }
782 # print "\n";
783
784 my $swishpath = $database."#".$iR;
785
786 next if (! $row);
787
788 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
789 $xml = $cp2utf->convert($xml);
790 use bytes; # as opposed to chars
791 print "Path-Name: $swishpath\n";
792 print "Content-Length: ".(length($xml)+1)."\n";
793 print "Document-Type: XML\n\n$xml\n";
794 }
795 }
796 } elsif ($type_base eq "marc") {
797
798 use MARC;
799
800 $import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
801 my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
802
803 # optional argument is format
804 my $format = x($config->{marc_format}) || 'usmarc';
805
806 print STDERR "Reading MARC file '$marc_file'\n";
807
808 my $marc = new MARC;
809 my $nr = $marc->openmarc({
810 file=>$marc_file, format=>$format
811 }) || die "Can't open MARC file '$marc_file' with format '$format'";
812
813 # read MARC file in memory
814 $marc->nextmarc(-1);
815
816 my $max_rec = $marc->marc_count();
817
818 for(my $i=1; $i<=$max_rec; $i++) {
819
820 progress($i,$max_rec);
821
822 # store value for marc_sf.pm
823 $main::cache->{marc_record} = $i;
824
825 my $swishpath = $database."#".$i;
826
827 if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) {
828 $xml = $cp2utf->convert($xml);
829 use bytes; # as opposed to chars
830 print "Path-Name: $swishpath\n";
831 print "Content-Length: ".(length($xml)+1)."\n";
832 print "Document-Type: XML\n\n$xml\n";
833 }
834 }
835
836 print STDERR "\n";
837
838 } elsif ($type_base eq "feed") {
839
840 $import2cp = Text::Iconv->new($config->{feed_codepage},$codepage);
841 my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!";
842
843 print STDERR "Reading feed from program '$prog'\n";
844
845 open(FEED,"feeds/$prog |") || die "can't start $prog: $!";
846
847 my $i=1; # record nr.
848
849 my $data;
850 my $line=1;
851
852 while (<FEED>) {
853 chomp;
854
855 if (/^$/) {
856 my $swishpath = $database."#".$i++;
857
858 if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) {
859 $xml = $cp2utf->convert($xml);
860 use bytes; # as opposed to chars
861 print "Path-Name: $swishpath\n";
862 print "Content-Length: ".(length($xml)+1)."\n";
863 print "Document-Type: XML\n\n$xml\n";
864 }
865 $line = 1;
866 $data = {};
867 next;
868 }
869
870 $line = $1 if (s/^(\d+):\s*//);
871 $data->{$line++} = $_;
872
873 fakeprogress($i);
874
875 }
876 # close lookup
877 untie %lhash if (%lhash);
878 }
879 }
880
881 # call this to commit index
882 $index->close;
883
884 1;
885 __END__
886 ##########################################################################
887
888 =head1 NAME
889
890 all2xml.pl - read various file formats and dump XML for SWISH-E
891
892 =head1 DESCRIPTION
893
894 This command will read ISIS data file using OpenIsis perl module, MARC
895 records using MARC module and optionally Micro$oft Excel files to
896 create one XML file for usage with I<SWISH-E> indexer. Dispite it's name,
897 this script B<isn't general xml generator> from isis files (isis allready
898 has something like that). Output of this script is tailor-made for SWISH-E.
899
900 =head1 BUGS
901
902 Documentation is really lacking. However, in true Open Source spirit, source
903 is best documentation. I even made considerable effort to comment parts
904 which are not intuitively clear, so...
905
906 =head1 AUTHOR
907
908 Dobrica Pavlinusic <dpavlin@rot13.org>
909
910 =head1 COPYRIGHT
911
912 GNU Public License (GPL) v2 or later
913
914 =head1 SEE ALSO
915
916 SWISH-E web site at http://www.swish-e.org
917
918 =cut

Properties

Name Value
cvs2svn:cvs-rev 1.63
svn:executable *

  ViewVC Help
Powered by ViewVC 1.1.26