5 |
use Getopt::Std; |
use Getopt::Std; |
6 |
use Data::Dumper; |
use Data::Dumper; |
7 |
use XML::Simple; |
use XML::Simple; |
8 |
use Text::Unaccent; |
use Text::Unaccent 1.02; # 1.01 won't compile on my platform, |
9 |
require Unicode::Map8; |
use Text::Iconv; |
10 |
|
use Config::IniFiles; |
11 |
my $config=XMLin(undef, forcearray => [ 'isis' ], forcecontent => 1); |
use Encode; |
12 |
|
|
13 |
|
$|=1; |
14 |
|
|
15 |
|
my $config_file = $0; |
16 |
|
$config_file =~ s/\.pl$/.conf/; |
17 |
|
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
18 |
|
|
19 |
|
my $config; |
20 |
|
|
21 |
|
#use index_DBI; # default DBI module for index |
22 |
|
use index_DBI_cache; # faster DBI module using memory cache |
23 |
|
my $index; |
24 |
|
|
25 |
my %opts; |
my %opts; |
26 |
|
|
27 |
getopts('d:m:q', \%opts); |
# usage: |
28 |
|
# -d directory name |
29 |
|
# -m multiple directories |
30 |
|
# -q quiet |
31 |
|
# -s run swish |
32 |
|
|
33 |
|
getopts('d:m:qs', \%opts); |
34 |
|
|
35 |
|
my $path; # this is name of database |
36 |
|
|
37 |
|
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 |
40 |
|
# filter/*.pm files! It will be used to store strings in perl internally! |
41 |
|
my $codepage = 'ISO-8859-2'; |
42 |
|
|
43 |
|
my $utf2cp = Text::Iconv->new('UTF-8',$codepage); |
44 |
|
# this function will convert data from XML files to local encoding |
45 |
|
sub x { |
46 |
|
return $utf2cp->convert($_[0]); |
47 |
|
} |
48 |
|
|
49 |
my $db_dir = $opts{d} || "ps"; # FIX |
# decode isis/excel or other import codepage |
50 |
|
my $import2cp; |
51 |
|
|
52 |
#die "usage: $0 -d [database_dir] -m [database1,database2] " if (! %opts); |
# outgoing xml must be in UTF-8 |
53 |
|
my $cp2utf = Text::Iconv->new($codepage,'UTF-8'); |
54 |
|
|
55 |
#print Dumper($config->{indexer}); |
# mapping between data type and tag which specify |
56 |
#print "-" x 70,"\n"; |
# format in XML file |
57 |
|
my %type2tag = ( |
58 |
|
'isis' => 'isis', |
59 |
|
'excel' => 'column', |
60 |
|
'marc' => 'marc', |
61 |
|
'feed' => 'feed' |
62 |
|
); |
63 |
|
|
64 |
# how to convert isis code page to UTF8? |
sub data2xml { |
|
my $isis_map = Unicode::Map8->new($config->{isis_codepage}) || die; |
|
65 |
|
|
66 |
sub isis2xml { |
use xmlify; |
67 |
|
|
68 |
|
my $type = shift @_; |
69 |
my $row = shift @_; |
my $row = shift @_; |
70 |
|
my $add_xml = shift @_; |
71 |
|
# needed to read values from configuration file |
72 |
|
my $cfg = shift @_; |
73 |
|
my $database = shift @_; |
74 |
|
|
75 |
my $xml; |
my $xml; |
76 |
|
|
77 |
sub isis_sf { |
use parse_format; |
78 |
my $row = shift @_; |
|
79 |
my $isis_id = shift @_; |
my $html = ""; # html formatted display output |
80 |
my $subfield = shift @_; |
|
81 |
if ($row->{$isis_id}->[0]) { |
my %field_usage; # counter for usage of each field |
82 |
my $sf = OpenIsis::subfields($row->{$isis_id}->[0]); |
|
83 |
if (! defined $subfield || length($subfield) == 0) { |
# sort subrouting using order="" attribute |
84 |
# subfield list undef, empty or no defined subfields for this record |
sub by_order { |
85 |
return $row->{$isis_id}->[0]; |
my $va = $config->{indexer}->{$a}->{order} || |
86 |
} elsif ($sf->{$subfield}) { |
$config->{indexer}->{$a}; |
87 |
return $sf->{$subfield}; |
my $vb = $config->{indexer}->{$b}->{order} || |
88 |
} |
$config->{indexer}->{$b}; |
89 |
} |
|
90 |
|
return $va <=> $vb; |
91 |
} |
} |
92 |
|
|
93 |
foreach my $field (keys %{$config->{indexer}}) { |
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
94 |
|
|
95 |
|
$field=x($field); |
96 |
|
$field_usage{$field}++; |
97 |
|
|
98 |
|
my $swish_data = ""; |
99 |
|
my $swish_exact_data = ""; |
100 |
my $display_data = ""; |
my $display_data = ""; |
101 |
my $index_data = ""; |
my $line_delimiter; |
102 |
|
|
103 |
|
my ($swish,$display); |
104 |
|
|
105 |
|
my $tag = $type2tag{$type} || die "can't find which tag to use for type $type"; |
106 |
|
foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) { |
107 |
|
|
108 |
|
my $format = x($x->{content}); |
109 |
|
my $delimiter = x($x->{delimiter}) || ' '; |
110 |
|
|
111 |
|
my $repeat_off = 0; # repeatable offset |
112 |
|
|
113 |
|
my ($s,$se,$d,$i) = (1,0,1,0); # swish, display default |
114 |
|
$s = 0 if (lc($x->{type}) eq "display"); |
115 |
|
$d = 0 if (lc($x->{type}) eq "swish"); |
116 |
|
$se = 1 if (lc($x->{type}) eq "swish_exact"); |
117 |
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
118 |
|
|
119 |
foreach my $x (@{$config->{indexer}->{$field}->{isis}}) { |
# what will separate last line from this one? |
120 |
|
if ($display_data && $x->{append} && $x->{append} eq "1") { |
121 |
|
$line_delimiter = ' '; |
122 |
|
} elsif ($display_data) { |
123 |
|
$line_delimiter = '<br/>'; |
124 |
|
} |
125 |
|
|
126 |
|
# init vars so that we go into while... |
127 |
|
($swish,$display) = (1,1); |
128 |
|
|
129 |
my $display_tmp = ""; |
# placeholder for all repeatable entries for index |
130 |
my $index_tmp = ""; |
my @index_data; |
131 |
|
my $index_filter; |
132 |
|
|
133 |
|
sub mkformat { |
134 |
|
my $x = shift || die "mkformat needs tag reference"; |
135 |
|
my $data = shift || return; |
136 |
|
my $format_name = x($x->{format_name}) || return $data; |
137 |
|
my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!"; |
138 |
|
my $format_delimiter = x($x->{format_delimiter}); |
139 |
|
my @data; |
140 |
|
if ($format_delimiter) { |
141 |
|
@data = split(/$format_delimiter/,$data); |
142 |
|
} else { |
143 |
|
push @data,$data; |
144 |
|
} |
145 |
|
|
146 |
my $format = $x->{content}; |
if ($fmt) { |
147 |
my $i = 1; # index only |
my $nr = scalar $fmt =~ s/%s/%s/g; |
148 |
my $d = 1; # display only |
if (($#data+1) == $nr) { |
149 |
$i = 0 if (lc($x->{type}) eq "display"); |
return sprintf($fmt,@data); |
150 |
$d = 0 if (lc($x->{type}) eq "index"); |
} else { |
151 |
#print "## i: $i d: $d ## $format ##"; |
print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n"; |
152 |
# parse format |
return $data; |
153 |
my $prefix = ""; |
} |
|
if ($format =~ s/^([^\d]+)//) { |
|
|
$prefix = $1; |
|
|
} |
|
|
while ($format) { |
|
|
if ($format =~ s/^(\d\d\d)(\w?)//) { |
|
|
my $isis_tmp = isis_sf($row,$1,$2); |
|
|
if ($isis_tmp) { |
|
|
# $display_tmp .= $prefix . "/$1/$2/".$isis_tmp if ($d); |
|
|
$display_tmp .= $prefix . $isis_tmp if ($d); |
|
|
$index_tmp .= $isis_tmp." " if ($i); |
|
|
#print " $isis_tmp <--\n"; |
|
|
} |
|
|
$prefix = ""; |
|
|
} elsif ($format =~ s/^([^\d]+)//) { |
|
|
$prefix = $1; |
|
154 |
} else { |
} else { |
155 |
print STDERR "WARNING: unparsed format '$format'\n"; |
print STDERR "usage of link '$format_name' without defined format (<link> tag)\n"; |
156 |
|
} |
157 |
|
} |
158 |
|
|
159 |
|
# while because of repeatable fields |
160 |
|
while ($swish || $display) { |
161 |
|
($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp); |
162 |
|
if ($repeat_off > 1000) { |
163 |
|
print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n"; |
164 |
last; |
last; |
165 |
}; |
} |
166 |
|
|
167 |
|
# filter="name" ; filter this field through |
168 |
|
# filter/[name].pm |
169 |
|
my $filter = $x->{filter}; |
170 |
|
if ($filter) { |
171 |
|
require "filter/".$filter.".pm"; |
172 |
|
} |
173 |
|
# type="swish" ; field for swish |
174 |
|
if ($swish) { |
175 |
|
if ($filter && ($s || $se)) { |
176 |
|
no strict 'refs'; |
177 |
|
my $tmp = join(" ",&$filter($swish)) if ($s || $se); |
178 |
|
$swish_data .= $tmp if ($s); |
179 |
|
$swish_exact_data .= $tmp if ($se); |
180 |
|
|
181 |
|
} else { |
182 |
|
$swish_data .= $swish if ($s); |
183 |
|
$swish_exact_data .= $swish if ($se); |
184 |
|
} |
185 |
|
} |
186 |
|
|
187 |
|
# type="display" ; field for display |
188 |
|
if ($d && $display) { |
189 |
|
if ($line_delimiter && $display_data) { |
190 |
|
$display_data .= $line_delimiter; |
191 |
|
undef $line_delimiter; |
192 |
|
} |
193 |
|
if ($filter) { |
194 |
|
no strict 'refs'; |
195 |
|
if ($display_data) { |
196 |
|
$display_data .= $delimiter.join($delimiter,mkformat($x,&$filter($display))); |
197 |
|
} else { |
198 |
|
$display_data = join($delimiter,mkformat($x,&$filter($display))); |
199 |
|
} |
200 |
|
} else { |
201 |
|
if ($display_data) { |
202 |
|
$display_data .= $delimiter.mkformat($x,$display); |
203 |
|
} else { |
204 |
|
$display_data = mkformat($x,$display); |
205 |
|
} |
206 |
|
} |
207 |
|
} |
208 |
|
|
209 |
|
# type="index" ; insert into index |
210 |
|
if ($i && $display) { |
211 |
|
push @index_data, $display; |
212 |
|
$index_filter = $filter if ($filter); |
213 |
|
} |
214 |
} |
} |
|
# add suffix |
|
|
$display_tmp .= $prefix if ($display_tmp); |
|
215 |
|
|
216 |
# $display_data .= $display_tmp if ($display_tmp ne ""); |
# fill data in index |
217 |
# $index_data .= $index_tmp if ($index_tmp ne ""); |
if (@index_data) { |
218 |
$display_data .= $display_tmp; |
if ($index_filter) { |
219 |
$index_data .= $index_tmp; |
no strict 'refs'; |
220 |
|
foreach my $d (@index_data) { |
221 |
|
$index->insert($field, &$index_filter($d), $path); |
222 |
|
} |
223 |
|
} else { |
224 |
|
foreach my $d (@index_data) { |
225 |
|
$index->insert($field, $d, $path); |
226 |
|
} |
227 |
|
} |
228 |
|
} |
229 |
|
} |
230 |
|
|
231 |
|
# now try to parse variables from configuration file |
232 |
|
foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) { |
233 |
|
|
234 |
|
my $delimiter = x($x->{delimiter}) || ' '; |
235 |
|
my $val = $cfg->val($database, x($x->{content})); |
236 |
|
|
237 |
|
my ($s,$d,$i) = (1,1,0); # swish, display default |
238 |
|
$s = 0 if (lc($x->{type}) eq "display"); |
239 |
|
$d = 0 if (lc($x->{type}) eq "swish"); |
240 |
|
# no support for swish exact in config. |
241 |
|
# IMHO, it's useless |
242 |
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
243 |
|
|
244 |
|
if ($val) { |
245 |
|
$display_data .= $delimiter.$val if ($d); |
246 |
|
$swish_data .= $val if ($s); |
247 |
|
$index->insert($field, $val, $path) if ($i); |
248 |
|
} |
249 |
|
|
250 |
} |
} |
251 |
#print "--display:$display_data\n--index:$index_data\n"; |
|
252 |
$xml->{$field}->{display} .= $isis_map->tou($display_data)->utf8 if ($display_data); |
|
253 |
$xml->{$field}->{index} .= unac_string($config->{isis_codepage},$index_data) if ($index_data); |
if ($display_data) { |
254 |
|
|
255 |
|
if ($field eq "headline") { |
256 |
|
$xml .= xmlify("headline", $display_data); |
257 |
|
} else { |
258 |
|
|
259 |
|
# find field name (signular, plural) |
260 |
|
my $field_name = ""; |
261 |
|
if ($config->{indexer}->{$field}->{name_singular} && $field_usage{$field} == 1) { |
262 |
|
$field_name = $config->{indexer}->{$field}->{name_singular}."#-#"; |
263 |
|
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
264 |
|
$field_name = $config->{indexer}->{$field}->{name_plural}."#-#"; |
265 |
|
} elsif ($config->{indexer}->{$field}->{name}) { |
266 |
|
$field_name = $config->{indexer}->{$field}->{name}."#-#"; |
267 |
|
} else { |
268 |
|
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
269 |
|
} |
270 |
|
if ($field_name) { |
271 |
|
$html .= x($field_name); |
272 |
|
} |
273 |
|
$html .= $display_data."###\n"; |
274 |
|
} |
275 |
|
} |
276 |
|
if ($swish_data) { |
277 |
|
# remove extra spaces |
278 |
|
$swish_data =~ s/ +/ /g; |
279 |
|
$swish_data =~ s/ +$//g; |
280 |
|
|
281 |
|
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
282 |
|
} |
283 |
|
|
284 |
|
if ($swish_exact_data) { |
285 |
|
$swish_exact_data =~ s/ +/ /g; |
286 |
|
$swish_exact_data =~ s/ +$//g; |
287 |
|
|
288 |
|
# add delimiters before and after word. |
289 |
|
# That is required to produce exact match |
290 |
|
$xml .= xmlify($field."_swish_exact", unac_string($codepage,'xxbxx '.$swish_exact_data.' xxexx')); |
291 |
|
} |
292 |
|
|
293 |
|
|
294 |
} |
} |
295 |
|
|
296 |
|
# dump formatted output in <html> |
297 |
|
if ($html) { |
298 |
|
#$xml .= xmlify("html",$html); |
299 |
|
$xml .= "<html><![CDATA[ $html ]]></html>"; |
300 |
|
} |
301 |
|
|
302 |
if ($xml) { |
if ($xml) { |
303 |
return XMLout($xml, rootname => 'xml', noattr => 1 ); |
$xml .= $add_xml if ($add_xml); |
304 |
|
return "<xml>\n$xml</xml>\n"; |
305 |
} else { |
} else { |
306 |
return; |
return; |
307 |
} |
} |
309 |
|
|
310 |
########################################################################## |
########################################################################## |
311 |
|
|
312 |
my $last_tell=0; |
# read configuration for this script |
313 |
|
my $cfg = new Config::IniFiles( -file => $config_file ); |
314 |
|
|
315 |
my @isis_dirs = ( '.' ); # use dirname as database name |
# read global.conf configuration |
316 |
|
my $cfg_global = new Config::IniFiles( -file => 'global.conf' ); |
317 |
|
|
318 |
if ($opts{m}) { |
# open index |
319 |
@isis_dirs = split(/,/,$opts{m}); |
$index = new index_DBI( |
320 |
} |
$cfg_global->val('global', 'dbi_dbd'), |
321 |
|
$cfg_global->val('global', 'dbi_dsn'), |
322 |
|
$cfg_global->val('global', 'dbi_user'), |
323 |
|
$cfg_global->val('global', 'dbi_passwd') || '', |
324 |
|
); |
325 |
|
|
326 |
my @isis_dbs; |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
327 |
|
|
328 |
foreach (@isis_dirs) { |
foreach my $database ($cfg->Sections) { |
329 |
if (-e $config->{isis_data}."/$db_dir/$_/LIBRI") { |
|
330 |
push @isis_dbs,$config->{isis_data}."/$db_dir/$_/LIBRI/LIBRI"; |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
331 |
} |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
332 |
if (-e $config->{isis_data}."/$db_dir/$_/PERI") { |
|
333 |
push @isis_dbs,$config->{isis_data}."/$db_dir/$_/PERI/PERI"; |
print STDERR "reading ./import_xml/$type.xml\n"; |
334 |
} |
|
335 |
if (-e $config->{isis_data}."/$db_dir/$_/AMS") { |
# extract just type basic |
336 |
push @isis_dbs,$config->{isis_data}."/$db_dir/$_/AMS/AMS"; |
my $type_base = $type; |
337 |
|
$type_base =~ s/_.+$//g; |
338 |
|
|
339 |
|
$config=XMLin("./import_xml/$type.xml", forcearray => [ $type2tag{$type_base}, 'config', 'format' ], forcecontent => 1); |
340 |
|
|
341 |
|
# output current progress indicator |
342 |
|
my $last_p = 0; |
343 |
|
sub progress { |
344 |
|
return if (! $show_progress); |
345 |
|
my $current = shift; |
346 |
|
my $total = shift || 1; |
347 |
|
my $p = int($current * 100 / $total); |
348 |
|
if ($p != $last_p) { |
349 |
|
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
350 |
|
$last_p = $p; |
351 |
|
} |
352 |
} |
} |
353 |
if (-e $config->{isis_data}."/$db_dir/$_/ARTI") { |
|
354 |
# push @isis_dbs,$config->{isis_data}."/$db_dir/$_/ARTI/ARTI"; |
my $fake_dir = 1; |
355 |
|
sub fakeprogress { |
356 |
|
return if (! $show_progress); |
357 |
|
my $current = shift @_; |
358 |
|
|
359 |
|
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
360 |
|
|
361 |
|
$last_p += $fake_dir; |
362 |
|
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
363 |
|
if ($last_p % 10 == 0) { |
364 |
|
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
365 |
|
} |
366 |
} |
} |
|
} |
|
367 |
|
|
368 |
print STDERR "FATAL: Can't find isis database.\nPerhaps isis_data (".$config->{isis_data}.") has wrong value?\n" if (! @isis_dbs); |
# now read database |
369 |
|
print STDERR "using: $type...\n"; |
370 |
|
|
371 |
my $db; |
if ($type_base eq "isis") { |
372 |
|
|
373 |
foreach my $isis_db (@isis_dbs) { |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
374 |
|
|
375 |
|
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
376 |
|
my $db = OpenIsis::open( $isis_db ); |
377 |
|
|
378 |
|
# check if .txt database for OpenIsis is zero length, |
379 |
|
# if so, erase it and re-open database |
380 |
|
sub check_txt_db { |
381 |
|
my $isis_db = shift || die "need isis database name"; |
382 |
|
my $reopen = 0; |
383 |
|
|
384 |
|
if (-e $isis_db.".TXT") { |
385 |
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
386 |
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
387 |
|
$reopen++; |
388 |
|
} |
389 |
|
if (-e $isis_db.".PTR") { |
390 |
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
391 |
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
392 |
|
$reopen++; |
393 |
|
} |
394 |
|
return OpenIsis::open( $isis_db ) if ($reopen); |
395 |
|
} |
396 |
|
|
397 |
my $db = OpenIsis::open( $isis_db ); |
# EOF error |
398 |
if (0) { |
if ($db == -1) { |
399 |
# # FIX |
$db = check_txt_db($isis_db); |
400 |
# if (! $db ) { |
if ($db == -1) { |
401 |
print STDERR "WARNING: can't open '$isis_db'\n"; |
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
402 |
next ; |
next; |
403 |
} |
} |
404 |
|
} |
405 |
|
|
406 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
# OpenIsis::ERR_BADF |
407 |
|
if ($db == -4) { |
408 |
|
print STDERR "FATAL: OpenIsis can't find file $isis_db\n"; |
409 |
|
next; |
410 |
|
# OpenIsis::ERR_IO |
411 |
|
} elsif ($db == -5) { |
412 |
|
print STDERR "FATAL: OpenIsis can't access file $isis_db\n"; |
413 |
|
next; |
414 |
|
} elsif ($db < 0) { |
415 |
|
print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n"; |
416 |
|
next; |
417 |
|
} |
418 |
|
|
419 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
my $max_rowid = OpenIsis::maxRowid( $db ); |
420 |
|
|
421 |
my $last_p = 0; |
# if 0 records, try to rease isis .txt database |
422 |
|
if ($max_rowid == 0) { |
423 |
|
# force removal of database |
424 |
|
$db = check_txt_db($isis_db); |
425 |
|
$max_rowid = OpenIsis::maxRowid( $db ); |
426 |
|
} |
427 |
|
|
428 |
|
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
429 |
|
|
430 |
|
my $path = $database; |
431 |
|
|
432 |
|
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
433 |
|
my $row = OpenIsis::read( $db, $row_id ); |
434 |
|
if ($row && $row->{mfn}) { |
435 |
|
|
436 |
|
progress($row->{mfn}, $max_rowid); |
437 |
|
|
438 |
|
my $swishpath = $path."#".int($row->{mfn}); |
439 |
|
|
440 |
|
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
441 |
|
$xml = $cp2utf->convert($xml); |
442 |
|
use bytes; # as opposed to chars |
443 |
|
print "Path-Name: $swishpath\n"; |
444 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
445 |
|
print "Document-Type: XML\n\n$xml\n"; |
446 |
|
} |
447 |
|
} |
448 |
|
} |
449 |
|
# for this to work with current version of OpenIsis (0.9.0) |
450 |
|
# you might need my patch from |
451 |
|
# http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
452 |
|
OpenIsis::close($db); |
453 |
|
print STDERR "\n"; |
454 |
|
|
455 |
|
} elsif ($type_base eq "excel") { |
456 |
|
use Spreadsheet::ParseExcel; |
457 |
|
use Spreadsheet::ParseExcel::Utility qw(int2col); |
458 |
|
|
459 |
|
$import2cp = Text::Iconv->new($config->{excel_codepage},$codepage); |
460 |
|
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!"; |
461 |
|
|
462 |
|
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml"; |
463 |
|
my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml"; |
464 |
|
|
465 |
|
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'"; |
466 |
|
|
467 |
|
my $sheet_nr = 0; |
468 |
|
foreach my $oWks (@{$oBook->{Worksheet}}) { |
469 |
|
#print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n"; |
470 |
|
last if ($oWks->{Name} eq $sheet); |
471 |
|
$sheet_nr++; |
472 |
|
} |
473 |
|
|
474 |
|
my $oWorksheet = $oBook->{Worksheet}[$sheet_nr]; |
475 |
|
|
476 |
|
print STDERR "using sheet: ",$oWorksheet->{Name},"\n"; |
477 |
|
defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file"; |
478 |
|
my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow}; |
479 |
|
|
480 |
|
for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) { |
481 |
|
my $row; |
482 |
|
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
483 |
|
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
484 |
|
if ($cell) { |
485 |
|
$row->{int2col($iC)} = $cell->Value; |
486 |
|
} |
487 |
|
} |
488 |
|
|
489 |
|
progress($iR, $end_row); |
490 |
|
|
491 |
# { my $row_id = 1; |
# print "row[$iR/$end_row] "; |
492 |
# FIX |
# foreach (keys %{$row}) { |
493 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
# print "$_: ",$row->{$_},"\t"; |
494 |
my $row = OpenIsis::read( $db, $row_id ); |
# } |
495 |
if ($row && $row->{mfn}) { |
# print "\n"; |
496 |
|
|
497 |
# output current process indicator |
my $swishpath = $database."#".$iR; |
498 |
my $p = int($row->{mfn} * 100 / $max_rowid); |
|
499 |
if ($p != $last_p) { |
next if (! $row); |
500 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$row->{mfn},$max_rowid,"=" x ($p/2).">", $p ) if (! $opts{q}); |
|
501 |
$last_p = $p; |
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
502 |
|
$xml = $cp2utf->convert($xml); |
503 |
|
use bytes; # as opposed to chars |
504 |
|
print "Path-Name: $swishpath\n"; |
505 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
506 |
|
print "Document-Type: XML\n\n$xml\n"; |
507 |
} |
} |
508 |
|
} |
509 |
|
} elsif ($type_base eq "marc") { |
510 |
|
|
511 |
|
use MARC; |
512 |
|
|
513 |
|
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
514 |
|
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
515 |
|
|
516 |
|
# optional argument is format |
517 |
|
my $format = x($config->{format}) || 'usmarc'; |
518 |
|
|
519 |
|
print STDERR "Reading MARC file '$marc_file'\n"; |
520 |
|
|
521 |
|
my $marc = new MARC; |
522 |
|
my $nr = $marc->openmarc({ |
523 |
|
file=>$marc_file, format=>$format |
524 |
|
}) || die "Can't open MARC file '$marc_file'"; |
525 |
|
|
526 |
|
my $i=0; # record nr. |
527 |
|
|
528 |
|
my $rec; |
529 |
|
|
530 |
|
while ($marc->nextmarc(1)) { |
531 |
|
|
532 |
|
# XXX |
533 |
|
fakeprogress($i++); |
534 |
|
|
535 |
|
my $swishpath = $database."#".$i; |
536 |
|
|
537 |
if (my $xml = isis2xml($row)) { |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
538 |
print "Path-Name: $isis_db#".$row->{mfn}."\n"; |
$xml = $cp2utf->convert($xml); |
539 |
|
use bytes; # as opposed to chars |
540 |
|
print "Path-Name: $swishpath\n"; |
541 |
print "Content-Length: ".(length($xml)+1)."\n"; |
print "Content-Length: ".(length($xml)+1)."\n"; |
542 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
543 |
} |
} |
544 |
} |
} |
545 |
|
} elsif ($type_base eq "feed") { |
546 |
|
|
547 |
|
$import2cp = Text::Iconv->new($config->{feed_codepage},$codepage); |
548 |
|
my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!"; |
549 |
|
|
550 |
|
print STDERR "Reading feed from program '$prog'\n"; |
551 |
|
|
552 |
|
open(FEED,"feeds/$prog |") || die "can't start $prog: $!"; |
553 |
|
|
554 |
|
my $i=1; # record nr. |
555 |
|
|
556 |
|
my $data; |
557 |
|
my $line=1; |
558 |
|
|
559 |
|
while (<FEED>) { |
560 |
|
chomp; |
561 |
|
|
562 |
|
if (/^$/) { |
563 |
|
my $swishpath = $database."#".$i++; |
564 |
|
|
565 |
|
if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) { |
566 |
|
$xml = $cp2utf->convert($xml); |
567 |
|
use bytes; # as opposed to chars |
568 |
|
print "Path-Name: $swishpath\n"; |
569 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
570 |
|
print "Document-Type: XML\n\n$xml\n"; |
571 |
|
} |
572 |
|
$line = 1; |
573 |
|
$data = {}; |
574 |
|
next; |
575 |
|
} |
576 |
|
|
577 |
|
$line = $1 if (s/^(\d+):\s*//); |
578 |
|
$data->{$line++} = $_; |
579 |
|
|
580 |
|
fakeprogress($i); |
581 |
|
|
582 |
|
} |
583 |
} |
} |
|
print STDERR "\n"; |
|
584 |
} |
} |
585 |
|
|
586 |
|
# call this to commit index |
587 |
|
$index->close; |
588 |
|
|
589 |
1; |
1; |
590 |
__END__ |
__END__ |
592 |
|
|
593 |
=head1 NAME |
=head1 NAME |
594 |
|
|
595 |
isis2xml.pl - read isis file and dump XML |
all2xml.pl - read various file formats and dump XML for SWISH-E |
596 |
|
|
597 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
598 |
|
|
599 |
This command will read ISIS data file using OpenIsis perl module and |
This command will read ISIS data file using OpenIsis perl module, MARC |
600 |
create XML file for usage with I<SWISH-E> |
records using MARC module and optionally Micro$oft Excel files to |
601 |
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, |
602 |
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 |
603 |
script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |
604 |
|
|
605 |
|
=head1 BUGS |
606 |
|
|
607 |
|
Documentation is really lacking. However, in true Open Source spirit, source |
608 |
|
is best documentation. I even made considerable effort to comment parts |
609 |
|
which are not intuitively clear, so... |
610 |
|
|
611 |
=head1 AUTHOR |
=head1 AUTHOR |
612 |
|
|