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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 277 - (show annotations)
Sun Mar 14 12:11:19 2004 UTC (20 years ago) by dpavlin
File MIME type: text/plain
File size: 25529 byte(s)
merge -r263:HEAD trunk/ branches/

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

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.26