18 |
|
|
19 |
my $config; |
my $config; |
20 |
|
|
21 |
use index_DBI; # there is no other, right now ;-) |
#use index_DBI; # default DBI module for index |
22 |
|
use index_DBI_cache; # faster DBI module using memory cache |
23 |
my $index = new index_DBI(); # open index |
my $index; |
24 |
|
|
25 |
my %opts; |
my %opts; |
26 |
|
|
32 |
|
|
33 |
getopts('d:m:qs', \%opts); |
getopts('d:m:qs', \%opts); |
34 |
|
|
35 |
my $db_dir; |
my $path; # this is name of database |
36 |
|
|
37 |
Text::Iconv->raise_error(1); # Conversion errors raise exceptions |
Text::Iconv->raise_error(0); # Conversion errors don't raise exceptions |
38 |
|
|
39 |
# 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 |
40 |
# 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! |
46 |
return $utf2cp->convert($_[0]); |
return $utf2cp->convert($_[0]); |
47 |
} |
} |
48 |
|
|
49 |
# decode isis import codepage |
# decode isis/excel or other import codepage |
50 |
my $isis2cp; |
my $import2cp; |
51 |
|
|
52 |
# outgoing xml must be in UTF-8 |
# outgoing xml must be in UTF-8 |
53 |
my $cp2utf = Text::Iconv->new($codepage,'UTF-8'); |
my $cp2utf = Text::Iconv->new($codepage,'UTF-8'); |
54 |
|
|
55 |
sub isis2xml { |
# mapping between data type and tag which specify |
56 |
|
# format in XML file |
57 |
|
my %type2tag = ( |
58 |
|
'isis' => 'isis', |
59 |
|
'excel' => 'column', |
60 |
|
'marc' => 'marc', |
61 |
|
'feed' => 'feed' |
62 |
|
); |
63 |
|
|
64 |
|
my $cache; # for cacheing |
65 |
|
|
66 |
|
sub data2xml { |
67 |
|
|
68 |
use xmlify; |
use xmlify; |
69 |
|
|
70 |
|
my $type = shift @_; |
71 |
my $row = shift @_; |
my $row = shift @_; |
72 |
my $add_xml = shift @_; |
my $add_xml = shift @_; |
73 |
|
# needed to read values from configuration file |
74 |
|
my $cfg = shift @_; |
75 |
|
my $database = shift @_; |
76 |
|
|
77 |
my $xml; |
my $xml; |
78 |
|
|
84 |
|
|
85 |
# sort subrouting using order="" attribute |
# sort subrouting using order="" attribute |
86 |
sub by_order { |
sub by_order { |
87 |
return 0 if (! $config->{indexer}->{$a}->{order}); |
my $va = $config->{indexer}->{$a}->{order} || |
88 |
return 0 if (! $config->{indexer}->{$b}->{order}); |
$config->{indexer}->{$a}; |
89 |
|
my $vb = $config->{indexer}->{$b}->{order} || |
90 |
|
$config->{indexer}->{$b}; |
91 |
|
|
92 |
|
return $va <=> $vb; |
93 |
|
} |
94 |
|
|
95 |
return $config->{indexer}->{$a}->{order} <=> |
my @sorted_tags; |
96 |
$config->{indexer}->{$b}->{order} ; |
if ($cache->{tags_by_order}->{$type}) { |
97 |
|
@sorted_tags = @{$cache->{tags_by_order}->{$type}}; |
98 |
|
} else { |
99 |
|
@sorted_tags = sort by_order keys %{$config->{indexer}}; |
100 |
|
$cache->{tags_by_order}->{$type} = \@sorted_tags; |
101 |
} |
} |
102 |
|
|
103 |
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
foreach my $field (@sorted_tags) { |
104 |
|
|
105 |
$field=x($field); |
$field=x($field); |
|
|
|
106 |
$field_usage{$field}++; |
$field_usage{$field}++; |
107 |
|
|
108 |
my $swish_data = ""; |
my $swish_data = ""; |
109 |
|
my $swish_exact_data = ""; |
110 |
my $display_data = ""; |
my $display_data = ""; |
111 |
my $line_delimiter; |
my $line_delimiter; |
112 |
|
|
113 |
my ($swish,$display); |
my ($swish,$display); |
114 |
|
|
115 |
foreach my $x (@{$config->{indexer}->{$field}->{isis}}) { |
my $tag = $type2tag{$type} || die "can't find which tag to use for type $type"; |
116 |
|
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
117 |
|
|
118 |
my $format = x($x->{content}); |
my $format = x($x->{content}); |
119 |
my $delimiter = x($x->{delimiter}) || ' '; |
my $delimiter = x($x->{delimiter}) || ' '; |
120 |
|
|
121 |
my $isis_i = 0; # isis repeatable offset |
my $repeat_off = 0; # repeatable offset |
122 |
|
|
123 |
my ($s,$d,$i) = (1,1,0); # swish, display default |
my ($s,$se,$d,$i) = (1,0,1,0); # swish, display default |
124 |
$s = 0 if (lc($x->{type}) eq "display"); |
$s = 0 if (lc($x->{type}) eq "display"); |
125 |
$d = 0 if (lc($x->{type}) eq "swish"); |
$d = 0 if (lc($x->{type}) eq "swish"); |
126 |
|
$se = 1 if (lc($x->{type}) eq "swish_exact"); |
127 |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
128 |
|
|
129 |
# what will separate last line from this one? |
# what will separate last line from this one? |
136 |
# init vars so that we go into while... |
# init vars so that we go into while... |
137 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
138 |
|
|
139 |
while ($swish || $display) { |
# placeholder for all repeatable entries for index |
140 |
($swish,$display) = parse_format($format,$row,$isis_i++,$isis2cp); |
my @index_data; |
141 |
|
my $index_filter; |
142 |
|
|
143 |
|
sub mkformat { |
144 |
|
my $x = shift || die "mkformat needs tag reference"; |
145 |
|
my $data = shift || return; |
146 |
|
my $format_name = x($x->{format_name}) || return $data; |
147 |
|
my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!"; |
148 |
|
my $format_delimiter = x($x->{format_delimiter}); |
149 |
|
my @data; |
150 |
|
if ($format_delimiter) { |
151 |
|
@data = split(/$format_delimiter/,$data); |
152 |
|
} else { |
153 |
|
push @data,$data; |
154 |
|
} |
155 |
|
|
156 |
|
if ($fmt) { |
157 |
|
my $nr = scalar $fmt =~ s/%s/%s/g; |
158 |
|
if (($#data+1) == $nr) { |
159 |
|
return sprintf($fmt,@data); |
160 |
|
} else { |
161 |
|
print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
162 |
|
return $data; |
163 |
|
} |
164 |
|
} else { |
165 |
|
print STDERR "usage of link '$format_name' without defined format (<link> tag)\n"; |
166 |
|
} |
167 |
|
} |
168 |
|
|
169 |
|
# while because of repeatable fields |
170 |
|
while ($swish || $display) { |
171 |
|
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
172 |
|
if ($repeat_off > 1000) { |
173 |
|
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
174 |
|
last; |
175 |
|
} |
176 |
|
|
177 |
# filter="name" ; filter this field through |
# filter="name" ; filter this field through |
178 |
# filter/[name].pm |
# filter/[name].pm |
179 |
my $filter = $x->{filter}; |
my $filter = $x->{filter}; |
180 |
if ($filter) { |
if ($filter && !$cache->{filter_loaded}->{$filter}) { |
181 |
require "filter/".$filter.".pm"; |
require "filter/".$filter.".pm"; |
182 |
|
$cache->{filter_loaded}->{$filter}++; |
183 |
} |
} |
184 |
# type="swish" ; field for swish |
# type="swish" ; field for swish |
185 |
if ($s && $swish) { |
if ($swish) { |
186 |
if ($filter) { |
if ($filter && ($s || $se)) { |
187 |
no strict 'refs'; |
no strict 'refs'; |
188 |
$swish_data .= join(" ",&$filter($swish)); |
my $tmp = join(" ",&$filter($swish)) if ($s || $se); |
189 |
|
$swish_data .= $tmp if ($s); |
190 |
|
$swish_exact_data .= $tmp if ($se); |
191 |
|
|
192 |
} else { |
} else { |
193 |
$swish_data .= $swish; |
$swish_data .= $swish if ($s); |
194 |
|
$swish_exact_data .= $swish if ($se); |
195 |
} |
} |
196 |
} |
} |
197 |
|
|
203 |
} |
} |
204 |
if ($filter) { |
if ($filter) { |
205 |
no strict 'refs'; |
no strict 'refs'; |
206 |
$display_data .= join($delimiter,&$filter($display)); |
if ($display_data) { |
207 |
|
$display_data .= $delimiter.join($delimiter,mkformat($x,&$filter($display))); |
208 |
|
} else { |
209 |
|
$display_data = join($delimiter,mkformat($x,&$filter($display))); |
210 |
|
} |
211 |
} else { |
} else { |
212 |
if ($display_data) { |
if ($display_data) { |
213 |
$display_data .= $delimiter.$display; |
$display_data .= $delimiter.mkformat($x,$display); |
214 |
} else { |
} else { |
215 |
$display_data .= $display; |
$display_data = mkformat($x,$display); |
216 |
} |
} |
217 |
} |
} |
218 |
} |
} |
219 |
|
|
220 |
# type="index" ; insert into index |
# type="index" ; insert into index |
221 |
if ($i && $display) { |
if ($i && $display) { |
222 |
my $index_data = $display; |
push @index_data, $display; |
223 |
if ($filter) { |
$index_filter = $filter if ($filter); |
224 |
no strict 'refs'; |
} |
225 |
foreach my $d (&$filter($index_data)) { |
} |
226 |
$index->insert($field, $d, $db_dir); |
|
227 |
} |
# fill data in index |
228 |
} else { |
if (@index_data) { |
229 |
$index->insert($field, $index_data, $db_dir); |
if ($index_filter) { |
230 |
|
no strict 'refs'; |
231 |
|
foreach my $d (@index_data) { |
232 |
|
$index->insert($field, &$index_filter($d), $path); |
233 |
|
} |
234 |
|
} else { |
235 |
|
foreach my $d (@index_data) { |
236 |
|
$index->insert($field, $d, $path); |
237 |
} |
} |
238 |
} |
} |
239 |
} |
} |
240 |
} |
} |
241 |
|
|
242 |
|
# now try to parse variables from configuration file |
243 |
|
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
244 |
|
|
245 |
|
my $delimiter = x($x->{delimiter}) || ' '; |
246 |
|
my $val = $cfg->val($database, x($x->{content})); |
247 |
|
|
248 |
|
my ($s,$d,$i) = (1,1,0); # swish, display default |
249 |
|
$s = 0 if (lc($x->{type}) eq "display"); |
250 |
|
$d = 0 if (lc($x->{type}) eq "swish"); |
251 |
|
# no support for swish exact in config. |
252 |
|
# IMHO, it's useless |
253 |
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
254 |
|
|
255 |
|
if ($val) { |
256 |
|
$display_data .= $delimiter.$val if ($d); |
257 |
|
$swish_data .= $val if ($s); |
258 |
|
$index->insert($field, $val, $path) if ($i); |
259 |
|
} |
260 |
|
|
261 |
|
} |
262 |
|
|
263 |
|
|
264 |
if ($display_data) { |
if ($display_data) { |
265 |
|
|
292 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
293 |
} |
} |
294 |
|
|
295 |
|
if ($swish_exact_data) { |
296 |
|
$swish_exact_data =~ s/ +/ /g; |
297 |
|
$swish_exact_data =~ s/ +$//g; |
298 |
|
|
299 |
|
# add delimiters before and after word. |
300 |
|
# That is required to produce exact match |
301 |
|
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
302 |
|
} |
303 |
|
|
304 |
|
|
305 |
} |
} |
306 |
|
|
307 |
# dump formatted output in <html> |
# dump formatted output in <html> |
308 |
if ($html) { |
if ($html) { |
309 |
$xml .= xmlify("html",$html); |
#$xml .= xmlify("html",$html); |
310 |
|
$xml .= "<html><![CDATA[ $html ]]></html>"; |
311 |
} |
} |
312 |
|
|
313 |
if ($xml) { |
if ($xml) { |
320 |
|
|
321 |
########################################################################## |
########################################################################## |
322 |
|
|
323 |
|
# read configuration for this script |
324 |
my $cfg = new Config::IniFiles( -file => $config_file ); |
my $cfg = new Config::IniFiles( -file => $config_file ); |
325 |
|
|
326 |
|
# read global.conf configuration |
327 |
|
my $cfg_global = new Config::IniFiles( -file => 'global.conf' ); |
328 |
|
|
329 |
|
# open index |
330 |
|
$index = new index_DBI( |
331 |
|
$cfg_global->val('global', 'dbi_dbd'), |
332 |
|
$cfg_global->val('global', 'dbi_dsn'), |
333 |
|
$cfg_global->val('global', 'dbi_user'), |
334 |
|
$cfg_global->val('global', 'dbi_passwd') || '', |
335 |
|
); |
336 |
|
|
337 |
|
my $show_progress = $cfg_global->val('global', 'show_progress'); |
338 |
|
|
339 |
|
my $unac_filter = $cfg_global->val('global', 'unac_filter'); |
340 |
|
if ($unac_filter) { |
341 |
|
require $unac_filter; |
342 |
|
} |
343 |
|
|
344 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
345 |
|
|
346 |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
|
my $type = $cfg -> val($database, 'type') || die "$database doesn't have 'type' defined"; |
|
347 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
348 |
|
|
349 |
$config=XMLin("./import_xml/$type.xml", forcearray => [ 'isis' ], forcecontent => 1); |
print STDERR "reading ./import_xml/$type.xml\n"; |
350 |
|
|
351 |
$isis2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
# extract just type basic |
352 |
|
my $type_base = $type; |
353 |
|
$type_base =~ s/_.+$//g; |
354 |
|
|
355 |
my $db = OpenIsis::open( $isis_db ); |
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1); |
356 |
if (0) { |
|
357 |
# # FIX |
# output current progress indicator |
358 |
# if (! $db ) { |
my $last_p = 0; |
359 |
print STDERR "WARNING: can't open '$isis_db'\n"; |
sub progress { |
360 |
next ; |
return if (! $show_progress); |
361 |
|
my $current = shift; |
362 |
|
my $total = shift || 1; |
363 |
|
my $p = int($current * 100 / $total); |
364 |
|
if ($p != $last_p) { |
365 |
|
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
366 |
|
$last_p = $p; |
367 |
|
} |
368 |
} |
} |
369 |
|
|
370 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
my $fake_dir = 1; |
371 |
|
sub fakeprogress { |
372 |
|
return if (! $show_progress); |
373 |
|
my $current = shift @_; |
374 |
|
|
375 |
|
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
376 |
|
|
377 |
|
$last_p += $fake_dir; |
378 |
|
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
379 |
|
if ($last_p % 10 == 0) { |
380 |
|
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
381 |
|
} |
382 |
|
} |
383 |
|
|
384 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
# now read database |
385 |
|
print STDERR "using: $type...\n"; |
386 |
|
|
387 |
my $path = $database; # was $isis_db |
if ($type_base eq "isis") { |
388 |
|
|
389 |
my $last_p = 0; |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
390 |
|
|
391 |
# { my $row_id = 4514; |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
392 |
# FIX |
my $db = OpenIsis::open( $isis_db ); |
393 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
|
394 |
my $row = OpenIsis::read( $db, $row_id ); |
# check if .txt database for OpenIsis is zero length, |
395 |
if ($row && $row->{mfn}) { |
# if so, erase it and re-open database |
396 |
# output current process indicator |
sub check_txt_db { |
397 |
my $p = int($row->{mfn} * 100 / $max_rowid); |
my $isis_db = shift || die "need isis database name"; |
398 |
if ($p != $last_p) { |
my $reopen = 0; |
399 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$row->{mfn},$max_rowid,"=" x ($p/2).">", $p ) if (! $opts{q}); |
|
400 |
$last_p = $p; |
if (-e $isis_db.".TXT") { |
401 |
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
402 |
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
403 |
|
$reopen++; |
404 |
|
} |
405 |
|
if (-e $isis_db.".PTR") { |
406 |
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
407 |
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
408 |
|
$reopen++; |
409 |
} |
} |
410 |
|
return OpenIsis::open( $isis_db ) if ($reopen); |
411 |
|
} |
412 |
|
|
413 |
|
# EOF error |
414 |
|
if ($db == -1) { |
415 |
|
$db = check_txt_db($isis_db); |
416 |
|
if ($db == -1) { |
417 |
|
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
418 |
|
next; |
419 |
|
} |
420 |
|
} |
421 |
|
|
422 |
|
# OpenIsis::ERR_BADF |
423 |
|
if ($db == -4) { |
424 |
|
print STDERR "FATAL: OpenIsis can't find file $isis_db\n"; |
425 |
|
next; |
426 |
|
# OpenIsis::ERR_IO |
427 |
|
} elsif ($db == -5) { |
428 |
|
print STDERR "FATAL: OpenIsis can't access file $isis_db\n"; |
429 |
|
next; |
430 |
|
} elsif ($db < 0) { |
431 |
|
print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n"; |
432 |
|
next; |
433 |
|
} |
434 |
|
|
435 |
|
my $max_rowid = OpenIsis::maxRowid( $db ); |
436 |
|
|
437 |
|
# if 0 records, try to rease isis .txt database |
438 |
|
if ($max_rowid == 0) { |
439 |
|
# force removal of database |
440 |
|
$db = check_txt_db($isis_db); |
441 |
|
$max_rowid = OpenIsis::maxRowid( $db ); |
442 |
|
} |
443 |
|
|
444 |
|
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
445 |
|
|
446 |
|
my $path = $database; |
447 |
|
|
448 |
|
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
449 |
|
my $row = OpenIsis::read( $db, $row_id ); |
450 |
|
if ($row && $row->{mfn}) { |
451 |
|
|
452 |
|
progress($row->{mfn}, $max_rowid); |
453 |
|
|
454 |
|
my $swishpath = $path."#".int($row->{mfn}); |
455 |
|
|
456 |
|
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
457 |
|
$xml = $cp2utf->convert($xml); |
458 |
|
use bytes; # as opposed to chars |
459 |
|
print "Path-Name: $swishpath\n"; |
460 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
461 |
|
print "Document-Type: XML\n\n$xml\n"; |
462 |
|
} |
463 |
|
} |
464 |
|
} |
465 |
|
# for this to work with current version of OpenIsis (0.9.0) |
466 |
|
# you might need my patch from |
467 |
|
# http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
468 |
|
OpenIsis::close($db); |
469 |
|
print STDERR "\n"; |
470 |
|
|
471 |
|
} elsif ($type_base eq "excel") { |
472 |
|
use Spreadsheet::ParseExcel; |
473 |
|
use Spreadsheet::ParseExcel::Utility qw(int2col); |
474 |
|
|
475 |
|
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
476 |
|
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!"; |
477 |
|
|
478 |
|
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml"; |
479 |
|
my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml"; |
480 |
|
|
481 |
|
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'"; |
482 |
|
|
483 |
|
my $sheet_nr = 0; |
484 |
|
foreach my $oWks (@{$oBook->{Worksheet}}) { |
485 |
|
#print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n"; |
486 |
|
last if ($oWks->{Name} eq $sheet); |
487 |
|
$sheet_nr++; |
488 |
|
} |
489 |
|
|
490 |
|
my $oWorksheet = $oBook->{Worksheet}[$sheet_nr]; |
491 |
|
|
492 |
|
print STDERR "using sheet: ",$oWorksheet->{Name},"\n"; |
493 |
|
defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file"; |
494 |
|
my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow}; |
495 |
|
|
496 |
|
for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) { |
497 |
|
my $row; |
498 |
|
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
499 |
|
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
500 |
|
if ($cell) { |
501 |
|
$row->{int2col($iC)} = $cell->Value; |
502 |
|
} |
503 |
|
} |
504 |
|
|
505 |
|
progress($iR, $end_row); |
506 |
|
|
507 |
|
# print "row[$iR/$end_row] "; |
508 |
|
# foreach (keys %{$row}) { |
509 |
|
# print "$_: ",$row->{$_},"\t"; |
510 |
|
# } |
511 |
|
# print "\n"; |
512 |
|
|
513 |
|
my $swishpath = $database."#".$iR; |
514 |
|
|
515 |
|
next if (! $row); |
516 |
|
|
517 |
|
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
518 |
|
$xml = $cp2utf->convert($xml); |
519 |
|
use bytes; # as opposed to chars |
520 |
|
print "Path-Name: $swishpath\n"; |
521 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
522 |
|
print "Document-Type: XML\n\n$xml\n"; |
523 |
|
} |
524 |
|
} |
525 |
|
} elsif ($type_base eq "marc") { |
526 |
|
|
527 |
|
use MARC; |
528 |
|
|
529 |
|
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
530 |
|
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
531 |
|
|
532 |
|
# optional argument is format |
533 |
|
my $format = x($config->{format}) || 'usmarc'; |
534 |
|
|
535 |
|
print STDERR "Reading MARC file '$marc_file'\n"; |
536 |
|
|
537 |
|
my $marc = new MARC; |
538 |
|
my $nr = $marc->openmarc({ |
539 |
|
file=>$marc_file, format=>$format |
540 |
|
}) || die "Can't open MARC file '$marc_file'"; |
541 |
|
|
542 |
|
my $i=0; # record nr. |
543 |
|
|
544 |
|
my $rec; |
545 |
|
|
546 |
|
while ($marc->nextmarc(1)) { |
547 |
|
|
548 |
|
# XXX |
549 |
|
fakeprogress($i++); |
550 |
|
|
551 |
|
my $swishpath = $database."#".$i; |
552 |
|
|
553 |
if (my $xml = $cp2utf->convert(isis2xml($row,$add_xml))) { |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
554 |
|
$xml = $cp2utf->convert($xml); |
555 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
556 |
print "Path-Name: $path#".int($row->{mfn})."\n"; |
print "Path-Name: $swishpath\n"; |
557 |
print "Content-Length: ".(length($xml)+1)."\n"; |
print "Content-Length: ".(length($xml)+1)."\n"; |
558 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
559 |
} |
} |
560 |
} |
} |
561 |
|
} elsif ($type_base eq "feed") { |
562 |
|
|
563 |
|
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
564 |
|
my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!"; |
565 |
|
|
566 |
|
print STDERR "Reading feed from program '$prog'\n"; |
567 |
|
|
568 |
|
open(FEED,"feeds/$prog |") || die "can't start $prog: $!"; |
569 |
|
|
570 |
|
my $i=1; # record nr. |
571 |
|
|
572 |
|
my $data; |
573 |
|
my $line=1; |
574 |
|
|
575 |
|
while (<FEED>) { |
576 |
|
chomp; |
577 |
|
|
578 |
|
if (/^$/) { |
579 |
|
my $swishpath = $database."#".$i++; |
580 |
|
|
581 |
|
if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) { |
582 |
|
$xml = $cp2utf->convert($xml); |
583 |
|
use bytes; # as opposed to chars |
584 |
|
print "Path-Name: $swishpath\n"; |
585 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
586 |
|
print "Document-Type: XML\n\n$xml\n"; |
587 |
|
} |
588 |
|
$line = 1; |
589 |
|
$data = {}; |
590 |
|
next; |
591 |
|
} |
592 |
|
|
593 |
|
$line = $1 if (s/^(\d+):\s*//); |
594 |
|
$data->{$line++} = $_; |
595 |
|
|
596 |
|
fakeprogress($i); |
597 |
|
|
598 |
|
} |
599 |
} |
} |
|
print STDERR "\n"; |
|
600 |
} |
} |
601 |
|
|
602 |
# call this to commit index |
# call this to commit index |
608 |
|
|
609 |
=head1 NAME |
=head1 NAME |
610 |
|
|
611 |
isis2xml.pl - read isis file and dump XML |
all2xml.pl - read various file formats and dump XML for SWISH-E |
612 |
|
|
613 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
614 |
|
|
615 |
This command will read ISIS data file using OpenIsis perl module and |
This command will read ISIS data file using OpenIsis perl module, MARC |
616 |
create XML file for usage with I<SWISH-E> |
records using MARC module and optionally Micro$oft Excel files to |
617 |
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, |
618 |
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 |
619 |
script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |
620 |
|
|
621 |
|
=head1 BUGS |
622 |
|
|
623 |
|
Documentation is really lacking. However, in true Open Source spirit, source |
624 |
|
is best documentation. I even made considerable effort to comment parts |
625 |
|
which are not intuitively clear, so... |
626 |
|
|
627 |
=head1 AUTHOR |
=head1 AUTHOR |
628 |
|
|