9 |
use Text::Iconv; |
use Text::Iconv; |
10 |
use Config::IniFiles; |
use Config::IniFiles; |
11 |
use Encode; |
use Encode; |
12 |
|
#use GDBM_File; |
13 |
|
use Fcntl; # for O_RDWR |
14 |
|
use TDB_File; |
15 |
|
|
16 |
$|=1; |
$|=1; |
17 |
|
|
21 |
|
|
22 |
my $config; |
my $config; |
23 |
|
|
24 |
use index_DBI; # there is no other, right now ;-) |
#use index_DBI; # default DBI module for index |
25 |
|
use index_DBI_cache; # faster DBI module using memory cache |
26 |
my $index; |
my $index; |
27 |
|
|
28 |
my %opts; |
my %opts; |
37 |
|
|
38 |
my $path; # this is name of database |
my $path; # this is name of database |
39 |
|
|
40 |
Text::Iconv->raise_error(1); # Conversion errors raise exceptions |
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 |
# 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! |
# filter/*.pm files! It will be used to store strings in perl internally! |
59 |
# format in XML file |
# format in XML file |
60 |
my %type2tag = ( |
my %type2tag = ( |
61 |
'isis' => 'isis', |
'isis' => 'isis', |
62 |
'excel' => 'column' |
'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 |
sub data2xml { |
sub data2xml { |
76 |
|
|
77 |
use xmlify; |
use xmlify; |
79 |
my $type = shift @_; |
my $type = shift @_; |
80 |
my $row = shift @_; |
my $row = shift @_; |
81 |
my $add_xml = shift @_; |
my $add_xml = shift @_; |
82 |
|
# needed to read values from configuration file |
83 |
|
my $cfg = shift @_; |
84 |
|
my $database = shift @_; |
85 |
|
|
86 |
my $xml; |
my $xml; |
87 |
|
|
93 |
|
|
94 |
# sort subrouting using order="" attribute |
# sort subrouting using order="" attribute |
95 |
sub by_order { |
sub by_order { |
96 |
return 0 if (! $config->{indexer}->{$a}->{order}); |
my $va = $config->{indexer}->{$a}->{order} || |
97 |
return 0 if (! $config->{indexer}->{$b}->{order}); |
$config->{indexer}->{$a}; |
98 |
|
my $vb = $config->{indexer}->{$b}->{order} || |
99 |
|
$config->{indexer}->{$b}; |
100 |
|
|
101 |
|
return $va <=> $vb; |
102 |
|
} |
103 |
|
|
104 |
return $config->{indexer}->{$a}->{order} <=> |
my @sorted_tags; |
105 |
$config->{indexer}->{$b}->{order} ; |
if ($cache->{tags_by_order}->{$type}) { |
106 |
|
@sorted_tags = @{$cache->{tags_by_order}->{$type}}; |
107 |
|
} else { |
108 |
|
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
109 |
|
$cache->{tags_by_order}->{$type} = \@sorted_tags; |
110 |
} |
} |
111 |
|
|
112 |
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
# lookup key |
113 |
|
my $lookup_key; |
114 |
|
|
115 |
$field=x($field); |
foreach my $field (@sorted_tags) { |
116 |
|
|
117 |
|
$field=x($field); |
118 |
$field_usage{$field}++; |
$field_usage{$field}++; |
119 |
|
|
120 |
my $swish_data = ""; |
my $swish_data = ""; |
121 |
|
my $swish_exact_data = ""; |
122 |
my $display_data = ""; |
my $display_data = ""; |
123 |
my $line_delimiter; |
my $line_delimiter; |
124 |
|
|
132 |
|
|
133 |
my $repeat_off = 0; # repeatable offset |
my $repeat_off = 0; # repeatable offset |
134 |
|
|
135 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
# swish, swish_exact, display, index, index_lookup |
136 |
|
# swish and display defaults |
137 |
|
my ($s,$se,$d,$i,$il) = (1,0,1,0,0); |
138 |
$s = 0 if (lc($x->{type}) eq "display"); |
$s = 0 if (lc($x->{type}) eq "display"); |
139 |
$d = 0 if (lc($x->{type}) eq "swish"); |
$d = 0 if (lc($x->{type}) eq "swish"); |
140 |
|
$se = 1 if (lc($x->{type}) eq "swish_exact"); |
141 |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
142 |
|
$il = 1 if (lc($x->{type}) =~ /^lookup/); |
143 |
|
|
144 |
|
|
145 |
# what will separate last line from this one? |
# what will separate last line from this one? |
146 |
if ($display_data && $x->{append} && $x->{append} eq "1") { |
if ($display_data && $x->{append} && $x->{append} eq "1") { |
152 |
# init vars so that we go into while... |
# init vars so that we go into while... |
153 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
154 |
|
|
155 |
if ($swish || $display) { |
# placeholder for all repeatable entries for index |
156 |
|
my @index_data; |
157 |
|
|
158 |
|
sub mkformat { |
159 |
|
my $x = shift || die "mkformat needs tag reference"; |
160 |
|
my $data = shift || return; |
161 |
|
my $format_name = x($x->{format_name}) || return $data; |
162 |
|
my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!"; |
163 |
|
my $format_delimiter = x($x->{format_delimiter}); |
164 |
|
my @data; |
165 |
|
if ($format_delimiter) { |
166 |
|
@data = split(/$format_delimiter/,$data); |
167 |
|
} else { |
168 |
|
push @data,$data; |
169 |
|
} |
170 |
|
|
171 |
|
if ($fmt) { |
172 |
|
my $nr = scalar $fmt =~ s/%s/%s/g; |
173 |
|
if (($#data+1) == $nr) { |
174 |
|
return sprintf($fmt,@data); |
175 |
|
} else { |
176 |
|
print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
177 |
|
return $data; |
178 |
|
} |
179 |
|
} else { |
180 |
|
print STDERR "usage of link '$format_name' without defined format (<link> tag)\n"; |
181 |
|
} |
182 |
|
} |
183 |
|
|
184 |
|
# while because of repeatable fields |
185 |
|
while ($swish || $display) { |
186 |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
187 |
|
if ($repeat_off > 1000) { |
188 |
|
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
189 |
|
last; |
190 |
|
} |
191 |
|
|
192 |
|
# is this field is lookup? |
193 |
|
if ($display && $x->{lookup}) { |
194 |
|
if ($use_lhash_cache) { |
195 |
|
if (!defined($cache->{lhash}->{$display})) { |
196 |
|
my $new_display = $lhash{$display}; |
197 |
|
if ($new_display) { |
198 |
|
#print STDERR "lookup cache store '$display' = '$new_display'\n"; |
199 |
|
$display = $new_display; |
200 |
|
$cache->{lhash}->{$display} = $new_display; |
201 |
|
} else { |
202 |
|
print STDERR "WARNING: lookup for '$display' didn't find anything.\n"; |
203 |
|
$display = ""; |
204 |
|
$cache->{lhash}->{$display} = ""; |
205 |
|
} |
206 |
|
} else { |
207 |
|
$display = $cache->{lhash}->{$display}; |
208 |
|
} |
209 |
|
} else { |
210 |
|
$display = $lhash{$display} || ""; |
211 |
|
} |
212 |
|
} |
213 |
|
|
214 |
# filter="name" ; filter this field through |
# filter="name" ; filter this field through |
215 |
# filter/[name].pm |
# filter/[name].pm |
216 |
my $filter = $x->{filter}; |
my $filter = $x->{filter}; |
217 |
if ($filter) { |
if ($filter && !$cache->{filter_loaded}->{$filter}) { |
218 |
require "filter/".$filter.".pm"; |
require "filter/".$filter.".pm"; |
219 |
|
$cache->{filter_loaded}->{$filter}++; |
220 |
} |
} |
221 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
222 |
if ($s && $swish) { |
if ($swish) { |
223 |
if ($filter) { |
if ($filter && ($s || $se)) { |
224 |
no strict 'refs'; |
no strict 'refs'; |
225 |
$swish_data .= join(" ",&$filter($swish)); |
my $tmp = join(" ",&$filter($swish)) if ($s || $se); |
226 |
|
$swish_data .= $tmp if ($s); |
227 |
|
$swish_exact_data .= $tmp if ($se); |
228 |
|
|
229 |
} else { |
} else { |
230 |
$swish_data .= $swish; |
$swish_data .= $swish if ($s); |
231 |
|
$swish_exact_data .= $swish if ($se); |
232 |
} |
} |
233 |
} |
} |
234 |
|
|
240 |
} |
} |
241 |
if ($filter) { |
if ($filter) { |
242 |
no strict 'refs'; |
no strict 'refs'; |
243 |
$display_data .= join($delimiter,&$filter($display)); |
if ($display_data) { |
244 |
|
$display_data .= $delimiter.join($delimiter,mkformat($x,&$filter($display))); |
245 |
|
} else { |
246 |
|
$display_data = join($delimiter,mkformat($x,&$filter($display))); |
247 |
|
} |
248 |
} else { |
} else { |
249 |
if ($display_data) { |
if ($display_data) { |
250 |
$display_data .= $delimiter.$display; |
$display_data .= $delimiter.mkformat($x,$display); |
251 |
} else { |
} else { |
252 |
$display_data .= $display; |
$display_data = mkformat($x,$display); |
253 |
} |
} |
254 |
} |
} |
255 |
} |
} |
256 |
|
|
257 |
# type="index" ; insert into index |
# type="index" ; insert into index |
258 |
if ($i && $display) { |
if ($i && $display) { |
|
my $index_data = $display; |
|
259 |
if ($filter) { |
if ($filter) { |
260 |
no strict 'refs'; |
no strict 'refs'; |
261 |
foreach my $d (&$filter($index_data)) { |
$display = &$filter($display); |
262 |
$index->insert($field, $d, $path); |
} |
263 |
} |
if ($x->{append} && @index_data) { |
264 |
|
$index_data[$#index_data].=$display; |
265 |
} else { |
} else { |
266 |
$index->insert($field, $index_data, $path); |
push @index_data, $display; |
267 |
|
} |
268 |
|
} |
269 |
|
|
270 |
|
# store fields in lookup |
271 |
|
if ($il && $display) { |
272 |
|
if (lc($x->{type}) eq "lookup_key") { |
273 |
|
if ($lookup_key) { |
274 |
|
print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)"; |
275 |
|
} else { |
276 |
|
$lookup_key = $display; |
277 |
|
} |
278 |
|
} elsif (lc($x->{type}) eq "lookup_val") { |
279 |
|
if ($lookup_key) { |
280 |
|
$lhash{$lookup_key} = $display; |
281 |
|
} else { |
282 |
|
print STDERR "WARNING: no lookup_key defined for '$display'?"; |
283 |
|
} |
284 |
} |
} |
285 |
} |
} |
286 |
} |
} |
287 |
|
|
288 |
|
# fill data in index |
289 |
|
foreach my $d (@index_data) { |
290 |
|
$index->insert($field, $d, $path); |
291 |
|
} |
292 |
|
} |
293 |
|
|
294 |
|
# now try to parse variables from configuration file |
295 |
|
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
296 |
|
|
297 |
|
my $delimiter = x($x->{delimiter}) || ' '; |
298 |
|
my $val = $cfg->val($database, x($x->{content})); |
299 |
|
|
300 |
|
my ($s,$d,$i) = (1,1,0); # swish, display default |
301 |
|
$s = 0 if (lc($x->{type}) eq "display"); |
302 |
|
$d = 0 if (lc($x->{type}) eq "swish"); |
303 |
|
# no support for swish exact in config. |
304 |
|
# IMHO, it's useless |
305 |
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
306 |
|
|
307 |
|
if ($val) { |
308 |
|
$display_data .= $delimiter.$val if ($d); |
309 |
|
$swish_data .= $val if ($s); |
310 |
|
$index->insert($field, $val, $path) if ($i); |
311 |
|
} |
312 |
|
|
313 |
} |
} |
314 |
|
|
315 |
|
|
344 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
345 |
} |
} |
346 |
|
|
347 |
|
if ($swish_exact_data) { |
348 |
|
$swish_exact_data =~ s/ +/ /g; |
349 |
|
$swish_exact_data =~ s/ +$//g; |
350 |
|
|
351 |
|
# add delimiters before and after word. |
352 |
|
# That is required to produce exact match |
353 |
|
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
354 |
|
} |
355 |
|
|
356 |
|
|
357 |
} |
} |
358 |
|
|
359 |
# dump formatted output in <html> |
# dump formatted output in <html> |
360 |
if ($html) { |
if ($html) { |
361 |
$xml .= xmlify("html",$html); |
#$xml .= xmlify("html",$html); |
362 |
|
$xml .= "<html><![CDATA[ $html ]]></html>"; |
363 |
} |
} |
364 |
|
|
365 |
if ($xml) { |
if ($xml) { |
386 |
$cfg_global->val('global', 'dbi_passwd') || '', |
$cfg_global->val('global', 'dbi_passwd') || '', |
387 |
); |
); |
388 |
|
|
389 |
|
my $show_progress = $cfg_global->val('global', 'show_progress'); |
390 |
|
|
391 |
|
my $unac_filter = $cfg_global->val('global', 'unac_filter'); |
392 |
|
if ($unac_filter) { |
393 |
|
require $unac_filter; |
394 |
|
} |
395 |
|
|
396 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
397 |
|
|
398 |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
399 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
400 |
|
|
401 |
|
# create new lookup file |
402 |
|
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
403 |
|
if ($lookup_file) { |
404 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
405 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
406 |
|
print STDERR "creating lookup file '$lookup_file'\n"; |
407 |
|
} |
408 |
|
|
409 |
|
# open existing lookup file |
410 |
|
$lookup_file = $cfg -> val($database, 'lookup_open'); # optional |
411 |
|
if ($lookup_file) { |
412 |
|
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644; |
413 |
|
tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644; |
414 |
|
print STDERR "opening lookup file '$lookup_file'\n"; |
415 |
|
} |
416 |
|
|
417 |
print STDERR "reading ./import_xml/$type.xml\n"; |
print STDERR "reading ./import_xml/$type.xml\n"; |
418 |
|
|
419 |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type} ], forcecontent => 1); |
# extract just type basic |
420 |
|
my $type_base = $type; |
421 |
|
$type_base =~ s/_.+$//g; |
422 |
|
|
423 |
|
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1); |
424 |
|
|
425 |
# output current progress indicator |
# output current progress indicator |
426 |
my $last_p = 0; |
my $last_p = 0; |
427 |
sub progress { |
sub progress { |
428 |
#return if (! $opts{q}); # FIXME |
return if (! $show_progress); |
429 |
my $current = shift; |
my $current = shift; |
430 |
my $total = shift; |
my $total = shift || 1; |
431 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
432 |
if ($p != $last_p) { |
if ($p != $last_p) { |
433 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
435 |
} |
} |
436 |
} |
} |
437 |
|
|
438 |
|
my $fake_dir = 1; |
439 |
|
sub fakeprogress { |
440 |
|
return if (! $show_progress); |
441 |
|
my $current = shift @_; |
442 |
|
|
443 |
|
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
444 |
|
|
445 |
|
$last_p += $fake_dir; |
446 |
|
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
447 |
|
if ($last_p % 10 == 0) { |
448 |
|
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
449 |
|
} |
450 |
|
} |
451 |
|
|
452 |
# now read database |
# now read database |
453 |
print STDERR "using: $type...\n"; |
print STDERR "using: $type...\n"; |
454 |
|
|
455 |
if ($type eq "isis") { |
if ($type_base eq "isis") { |
456 |
|
|
457 |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
458 |
|
|
459 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
460 |
my $db = OpenIsis::open( $isis_db ); |
my $db = OpenIsis::open( $isis_db ); |
461 |
|
|
462 |
|
# check if .txt database for OpenIsis is zero length, |
463 |
|
# if so, erase it and re-open database |
464 |
|
sub check_txt_db { |
465 |
|
my $isis_db = shift || die "need isis database name"; |
466 |
|
my $reopen = 0; |
467 |
|
|
468 |
|
if (-e $isis_db.".TXT") { |
469 |
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
470 |
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
471 |
|
$reopen++; |
472 |
|
} |
473 |
|
if (-e $isis_db.".PTR") { |
474 |
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
475 |
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
476 |
|
$reopen++; |
477 |
|
} |
478 |
|
return OpenIsis::open( $isis_db ) if ($reopen); |
479 |
|
} |
480 |
|
|
481 |
|
# EOF error |
482 |
|
if ($db == -1) { |
483 |
|
$db = check_txt_db($isis_db); |
484 |
|
if ($db == -1) { |
485 |
|
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
486 |
|
next; |
487 |
|
} |
488 |
|
} |
489 |
|
|
490 |
|
# OpenIsis::ERR_BADF |
491 |
|
if ($db == -4) { |
492 |
|
print STDERR "FATAL: OpenIsis can't find file $isis_db\n"; |
493 |
|
next; |
494 |
|
# OpenIsis::ERR_IO |
495 |
|
} elsif ($db == -5) { |
496 |
|
print STDERR "FATAL: OpenIsis can't access file $isis_db\n"; |
497 |
|
next; |
498 |
|
} elsif ($db < 0) { |
499 |
|
print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n"; |
500 |
|
next; |
501 |
|
} |
502 |
|
|
503 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
my $max_rowid = OpenIsis::maxRowid( $db ); |
504 |
|
|
505 |
|
# if 0 records, try to rease isis .txt database |
506 |
|
if ($max_rowid == 0) { |
507 |
|
# force removal of database |
508 |
|
$db = check_txt_db($isis_db); |
509 |
|
$max_rowid = OpenIsis::maxRowid( $db ); |
510 |
|
} |
511 |
|
|
512 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
513 |
|
|
514 |
my $path = $database; |
my $path = $database; |
521 |
|
|
522 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
523 |
|
|
524 |
if (my $xml = data2xml($type,$row,$add_xml)) { |
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
525 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
526 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
527 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
530 |
} |
} |
531 |
} |
} |
532 |
} |
} |
533 |
|
# for this to work with current version of OpenIsis (0.9.0) |
534 |
|
# you might need my patch from |
535 |
|
# http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
536 |
|
OpenIsis::close($db); |
537 |
print STDERR "\n"; |
print STDERR "\n"; |
538 |
|
|
539 |
} elsif ($type eq "excel") { |
} elsif ($type_base eq "excel") { |
540 |
use Spreadsheet::ParseExcel; |
use Spreadsheet::ParseExcel; |
541 |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
use Spreadsheet::ParseExcel::Utility qw(int2col); |
542 |
|
|
582 |
|
|
583 |
next if (! $row); |
next if (! $row); |
584 |
|
|
585 |
if (my $xml = data2xml($type,$row,$add_xml)) { |
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
586 |
|
$xml = $cp2utf->convert($xml); |
587 |
|
use bytes; # as opposed to chars |
588 |
|
print "Path-Name: $swishpath\n"; |
589 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
590 |
|
print "Document-Type: XML\n\n$xml\n"; |
591 |
|
} |
592 |
|
} |
593 |
|
} elsif ($type_base eq "marc") { |
594 |
|
|
595 |
|
use MARC; |
596 |
|
|
597 |
|
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
598 |
|
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
599 |
|
|
600 |
|
# optional argument is format |
601 |
|
my $format = x($config->{format}) || 'usmarc'; |
602 |
|
|
603 |
|
print STDERR "Reading MARC file '$marc_file'\n"; |
604 |
|
|
605 |
|
my $marc = new MARC; |
606 |
|
my $nr = $marc->openmarc({ |
607 |
|
file=>$marc_file, format=>$format |
608 |
|
}) || die "Can't open MARC file '$marc_file'"; |
609 |
|
|
610 |
|
my $i=0; # record nr. |
611 |
|
|
612 |
|
my $rec; |
613 |
|
|
614 |
|
while ($marc->nextmarc(1)) { |
615 |
|
|
616 |
|
# XXX |
617 |
|
fakeprogress($i++); |
618 |
|
|
619 |
|
my $swishpath = $database."#".$i; |
620 |
|
|
621 |
|
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
622 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
623 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
624 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
626 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
627 |
} |
} |
628 |
} |
} |
629 |
|
} elsif ($type_base eq "feed") { |
630 |
|
|
631 |
|
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
632 |
|
my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!"; |
633 |
|
|
634 |
|
print STDERR "Reading feed from program '$prog'\n"; |
635 |
|
|
636 |
|
open(FEED,"feeds/$prog |") || die "can't start $prog: $!"; |
637 |
|
|
638 |
|
my $i=1; # record nr. |
639 |
|
|
640 |
|
my $data; |
641 |
|
my $line=1; |
642 |
|
|
643 |
|
while (<FEED>) { |
644 |
|
chomp; |
645 |
|
|
646 |
|
if (/^$/) { |
647 |
|
my $swishpath = $database."#".$i++; |
648 |
|
|
649 |
|
if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) { |
650 |
|
$xml = $cp2utf->convert($xml); |
651 |
|
use bytes; # as opposed to chars |
652 |
|
print "Path-Name: $swishpath\n"; |
653 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
654 |
|
print "Document-Type: XML\n\n$xml\n"; |
655 |
|
} |
656 |
|
$line = 1; |
657 |
|
$data = {}; |
658 |
|
next; |
659 |
|
} |
660 |
|
|
661 |
|
$line = $1 if (s/^(\d+):\s*//); |
662 |
|
$data->{$line++} = $_; |
663 |
|
|
664 |
|
fakeprogress($i); |
665 |
|
|
666 |
|
} |
667 |
|
# close lookup |
668 |
|
untie %lhash if (%lhash); |
669 |
} |
} |
670 |
} |
} |
671 |
|
|
678 |
|
|
679 |
=head1 NAME |
=head1 NAME |
680 |
|
|
681 |
isis2xml.pl - read isis file and dump XML |
all2xml.pl - read various file formats and dump XML for SWISH-E |
682 |
|
|
683 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
684 |
|
|
685 |
This command will read ISIS data file using OpenIsis perl module and |
This command will read ISIS data file using OpenIsis perl module, MARC |
686 |
create XML file for usage with I<SWISH-E> |
records using MARC module and optionally Micro$oft Excel files to |
687 |
indexer. Dispite it's name, this script B<isn't general xml generator> |
create one XML file for usage with I<SWISH-E> indexer. Dispite it's name, |
688 |
from isis files (isis allready has something like that). Output of this |
this script B<isn't general xml generator> from isis files (isis allready |
689 |
script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |
690 |
|
|
691 |
|
=head1 BUGS |
692 |
|
|
693 |
|
Documentation is really lacking. However, in true Open Source spirit, source |
694 |
|
is best documentation. I even made considerable effort to comment parts |
695 |
|
which are not intuitively clear, so... |
696 |
|
|
697 |
=head1 AUTHOR |
=head1 AUTHOR |
698 |
|
|