7 |
|
|
8 |
use base 'WebPAC::Common'; |
use base 'WebPAC::Common'; |
9 |
use File::Slurp; |
use File::Slurp; |
10 |
use Data::Dumper; |
use List::Util qw/first/; |
11 |
|
use Data::Dump qw/dump/; |
12 |
|
use WebPAC::Normalize qw/_pack_subfields_hash/; |
13 |
|
use Storable qw/dclone/; |
14 |
|
|
15 |
=head1 NAME |
=head1 NAME |
16 |
|
|
18 |
|
|
19 |
=head1 VERSION |
=head1 VERSION |
20 |
|
|
21 |
Version 0.01 |
Version 0.11 |
22 |
|
|
23 |
=cut |
=cut |
24 |
|
|
25 |
our $VERSION = '0.01'; |
our $VERSION = '0.11'; |
26 |
|
|
27 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
28 |
|
|
34 |
# same with 101 |
# same with 101 |
35 |
101 |
101 |
36 |
# field 200 have valid subfields a-g |
# field 200 have valid subfields a-g |
37 |
200 a b c d e f g |
# and field e is repeatable |
38 |
|
200 a b c d e* f g |
39 |
# field 205 can have only subfield a |
# field 205 can have only subfield a |
40 |
205 a |
# and must exists |
41 |
|
205! a |
42 |
# while 210 can have a c or d |
# while 210 can have a c or d |
43 |
210 a c d |
210 a c d |
44 |
|
# field which is ignored in validation |
45 |
|
999- |
46 |
|
|
47 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
48 |
|
|
51 |
Create new validation object |
Create new validation object |
52 |
|
|
53 |
my $validate = new WebPAC::Validate( |
my $validate = new WebPAC::Validate( |
54 |
path => '/path/to/input/validate_file', |
path => 'conf/validate/file', |
55 |
|
delimiters => [ ' : ', ' / ', ' ; ', ' , ' ], |
56 |
|
delimiters_path => 'conf/validate/delimiters/file', |
57 |
); |
); |
58 |
|
|
59 |
|
Optional parametar C<delimiters> will turn on validating of delimiters. Be |
60 |
|
careful here, those delimiters are just stuck into regex, so they can |
61 |
|
contain L<perlre> regexpes. |
62 |
|
|
63 |
=cut |
=cut |
64 |
|
|
65 |
sub new { |
sub new { |
67 |
my $self = {@_}; |
my $self = {@_}; |
68 |
bless($self, $class); |
bless($self, $class); |
69 |
|
|
70 |
|
warn dump( @_ ); |
71 |
|
|
72 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
73 |
|
|
74 |
foreach my $p (qw/path/) { |
if ( $self->{path} ) { |
75 |
$log->logconfess("need $p") unless ($self->{$p}); |
|
76 |
|
my $v_file = read_file( $self->{path} ) || |
77 |
|
$log->logdie("can't open validate path $self->{path}: $!"); |
78 |
|
|
79 |
|
my $v; |
80 |
|
my $curr_line = 1; |
81 |
|
|
82 |
|
foreach my $l (split(/[\n\r]+/, $v_file)) { |
83 |
|
$curr_line++; |
84 |
|
|
85 |
|
# skip comments and whitespaces |
86 |
|
next if ($l =~ /^#/ || $l =~ /^\s*$/); |
87 |
|
|
88 |
|
$l =~ s/^\s+//; |
89 |
|
$l =~ s/\s+$//; |
90 |
|
|
91 |
|
my @d = split(/\s+/, $l); |
92 |
|
|
93 |
|
my $fld = shift @d; |
94 |
|
|
95 |
|
if ($fld =~ s/!$//) { |
96 |
|
$self->{must_exist}->{$fld}++; |
97 |
|
} elsif ($fld =~ s/-$//) { |
98 |
|
$self->{dont_validate}->{$fld}++; |
99 |
|
} |
100 |
|
|
101 |
|
$log->logdie("need field name in line $curr_line: $l") unless (defined($fld)); |
102 |
|
|
103 |
|
if (@d) { |
104 |
|
$v->{$fld} = [ map { |
105 |
|
my $sf = $_; |
106 |
|
if ( $sf =~ s/!(\*)?$/$1/ ) { |
107 |
|
$self->{must_exist_sf}->{ $fld }->{ $sf }++; |
108 |
|
}; |
109 |
|
$sf; |
110 |
|
} @d ]; |
111 |
|
} else { |
112 |
|
$v->{$fld} = 1; |
113 |
|
} |
114 |
|
|
115 |
|
} |
116 |
|
|
117 |
|
$log->debug("current validation rules: ", dump($v)); |
118 |
|
|
119 |
|
$self->{rules} = $v; |
120 |
|
|
121 |
|
$log->info("validation uses rules from $self->{path}"); |
122 |
|
} |
123 |
|
|
124 |
|
if ( $self->{delimiters} ) { |
125 |
|
$self->{delimiters_regex} = '(\^[a-z0-9]|' . join('|', @{ $self->{delimiters} }) . ')'; |
126 |
|
$log->info("validation check delimiters with regex $self->{delimiters_regex}"); |
127 |
|
} |
128 |
|
|
129 |
|
if ( my $path = $self->{delimiters_path} ) { |
130 |
|
if ( -e $path ) { |
131 |
|
$log->info("using delimiter validation rules from $path"); |
132 |
|
open(my $d, $path) || $log->fatal("can't open $path: $!"); |
133 |
|
while(<$d>) { |
134 |
|
chomp($d); |
135 |
|
if (/^\s*(#*)\s*(\d+)\t+(\d+)\t+(.*)$/) { |
136 |
|
my ($comment,$field,$count,$template) = ($1,$2,$3,$4); |
137 |
|
$self->{_validate_delimiters_templates}->{$field}->{$template} = $count unless ($comment); |
138 |
|
} else { |
139 |
|
warn "## ignored $d\n"; |
140 |
|
} |
141 |
|
} |
142 |
|
close($d); |
143 |
|
warn "_validate_delimiters_templates = ",dump( $self->{_validate_delimiters_templates} ); |
144 |
|
} else { |
145 |
|
$log->warn("delimiters path $path doesn't exist, it will be created after this run"); |
146 |
|
} |
147 |
|
} |
148 |
|
|
149 |
|
$self ? return $self : return undef; |
150 |
|
} |
151 |
|
|
152 |
|
=head2 validate_rec |
153 |
|
|
154 |
|
Validate record and return errors |
155 |
|
|
156 |
|
my @errors = $validate->validate_rec( $rec, $rec_dump ); |
157 |
|
|
158 |
|
=cut |
159 |
|
|
160 |
|
sub validate_rec { |
161 |
|
my $self = shift; |
162 |
|
|
163 |
|
my $log = $self->_get_logger(); |
164 |
|
|
165 |
|
my $rec = shift || $log->logdie("validate_rec need record"); |
166 |
|
my $rec_dump = shift; |
167 |
|
|
168 |
|
$log->logdie("rec isn't HASH") unless (ref($rec) eq 'HASH'); |
169 |
|
# $log->logdie("can't find validation rules") unless (my $r = $self->{rules}); |
170 |
|
my $r = $self->{rules}; |
171 |
|
|
172 |
|
my $errors; |
173 |
|
|
174 |
|
$log->debug("rec = ", sub { dump($rec) }, "keys = ", keys %{ $rec }); |
175 |
|
|
176 |
|
my $fields; |
177 |
|
|
178 |
|
foreach my $f (keys %{ $rec }) { |
179 |
|
|
180 |
|
next if (!defined($f) || $f eq '' || $f eq '000'); |
181 |
|
|
182 |
|
# first check delimiters |
183 |
|
if ( my $regex = $self->{delimiters_regex} ) { |
184 |
|
|
185 |
|
foreach my $v (@{ $rec->{$f} }) { |
186 |
|
my $l = _pack_subfields_hash( $v, 1 ); |
187 |
|
my $subfield_dump = $l; |
188 |
|
my $template = ''; |
189 |
|
$l =~ s/$regex/$template.=$1/eg; |
190 |
|
#warn "## template: $template\n"; |
191 |
|
|
192 |
|
if ( $template ) { |
193 |
|
$self->{_delimiters_templates}->{$f}->{$template}++; |
194 |
|
|
195 |
|
if ( my $v = $self->{_validate_delimiters_templates} ) { |
196 |
|
if ( ! defined( $v->{$f}->{$template} ) ) { |
197 |
|
$errors->{$f}->{invalid_delimiters_combination} = $template; |
198 |
|
$errors->{$f}->{dump} = $subfield_dump; |
199 |
|
#} else { |
200 |
|
# warn "## $f $template ok\n"; |
201 |
|
} |
202 |
|
} |
203 |
|
} |
204 |
|
} |
205 |
|
} |
206 |
|
|
207 |
|
next unless ( $r ); # skip validation of no rules are specified |
208 |
|
|
209 |
|
next if (defined( $self->{dont_validate}->{$f} )); |
210 |
|
|
211 |
|
# track field usage |
212 |
|
$fields->{$f}++; |
213 |
|
|
214 |
|
if ( ! defined($r->{$f}) ) { |
215 |
|
$errors->{ $f }->{unexpected} = "this field is not expected"; |
216 |
|
next; |
217 |
|
} |
218 |
|
|
219 |
|
|
220 |
|
if (ref($rec->{$f}) ne 'ARRAY') { |
221 |
|
$errors->{ $f }->{not_repeatable} = "probably bug in parsing input data"; |
222 |
|
next; |
223 |
|
} |
224 |
|
|
225 |
|
foreach my $v (@{ $rec->{$f} }) { |
226 |
|
# can we have subfields? |
227 |
|
if (ref($r->{$f}) eq 'ARRAY') { |
228 |
|
# are values hashes? (has subfields) |
229 |
|
if (! defined($v)) { |
230 |
|
# $errors->{$f}->{empty} = undef; |
231 |
|
# $errors->{dump} = $rec_dump if ($rec_dump); |
232 |
|
} elsif (ref($v) ne 'HASH') { |
233 |
|
$errors->{$f}->{missing_subfield} = join(",", @{ $r->{$f} }) . " required"; |
234 |
|
next; |
235 |
|
} else { |
236 |
|
|
237 |
|
my $h = dclone( $v ); |
238 |
|
|
239 |
|
my $sf_repeatable; |
240 |
|
|
241 |
|
delete($v->{subfields}) if (defined($v->{subfields})); |
242 |
|
|
243 |
|
my $subfields; |
244 |
|
|
245 |
|
foreach my $sf (keys %{ $v }) { |
246 |
|
|
247 |
|
$subfields->{ $sf }++; |
248 |
|
|
249 |
|
# is non-repeatable but with multiple values? |
250 |
|
if ( ! first { $_ eq $sf.'*' } @{$r->{$f}} ) { |
251 |
|
if ( ref($v->{$sf}) eq 'ARRAY' ) { |
252 |
|
$sf_repeatable->{$sf}++; |
253 |
|
}; |
254 |
|
if (! first { $_ eq $sf } @{ $r->{$f} }) { |
255 |
|
$errors->{ $f }->{subfield}->{extra}->{$sf}++; |
256 |
|
} |
257 |
|
} |
258 |
|
|
259 |
|
} |
260 |
|
if (my @r_sf = sort keys( %$sf_repeatable )) { |
261 |
|
|
262 |
|
foreach my $sf (@r_sf) { |
263 |
|
$errors->{$f}->{subfield}->{extra_repeatable}->{$sf}++; |
264 |
|
$errors->{$f}->{dump} = _pack_subfields_hash( $h, 1 ); |
265 |
|
} |
266 |
|
|
267 |
|
} |
268 |
|
|
269 |
|
if ( defined( $self->{must_exist_sf}->{$f} ) ) { |
270 |
|
foreach my $sf (sort keys %{ $self->{must_exist_sf}->{$f} }) { |
271 |
|
#warn "====> $f $sf must exist\n"; |
272 |
|
$errors->{$f}->{subfield}->{missing}->{$sf}++ |
273 |
|
unless defined( $subfields->{$sf} ); |
274 |
|
} |
275 |
|
} |
276 |
|
|
277 |
|
} |
278 |
|
} elsif (ref($v) eq 'HASH') { |
279 |
|
$errors->{$f}->{unexpected_subfields}++; |
280 |
|
$errors->{$f}->{dump} = _pack_subfields_hash( $v, 1 ); |
281 |
|
} |
282 |
|
} |
283 |
} |
} |
284 |
|
|
285 |
my $v_file = read_file( $self->{path} ) || |
$log->debug("_delimiters_templates = ", dump( $self->{_delimiters_templates} ) ); |
286 |
$log->logdie("can't open validate path $self->{path}: $!"); |
|
287 |
|
foreach my $must (sort keys %{ $self->{must_exist} }) { |
288 |
|
next if ($fields->{$must}); |
289 |
|
$errors->{$must}->{missing}++; |
290 |
|
$errors->{dump} = $rec_dump if ($rec_dump); |
291 |
|
} |
292 |
|
|
293 |
|
if ($errors) { |
294 |
|
$log->debug("errors: ", $self->report_error( $errors ) ); |
295 |
|
|
296 |
|
my $mfn = $rec->{'000'}->[0] || $log->logconfess("record ", dump( $rec ), " doesn't have MFN"); |
297 |
|
$self->{errors}->{$mfn} = $errors; |
298 |
|
} |
299 |
|
|
300 |
|
#$log->logcluck("return from this function is ARRAY") unless wantarray; |
301 |
|
|
302 |
|
return $errors; |
303 |
|
} |
304 |
|
|
305 |
|
=head2 reset_errors |
306 |
|
|
307 |
my $v; |
Clean all accumulated errors for this input |
|
my $curr_line = 1; |
|
308 |
|
|
309 |
foreach my $l (split(/[\n\r]+/, $v_file)) { |
$validate->reset_errors; |
310 |
$curr_line++; |
|
311 |
# skip comments |
=cut |
312 |
next if ($l =~ m/^#/); |
|
313 |
|
sub reset_errors { |
314 |
|
my $self = shift; |
315 |
|
delete ($self->{errors}); |
316 |
|
} |
317 |
|
|
318 |
|
=head2 all_errors |
319 |
|
|
320 |
|
Return hash with all errors |
321 |
|
|
322 |
|
print dump( $validate->all_errors ); |
323 |
|
|
324 |
|
=cut |
325 |
|
|
326 |
|
sub all_errors { |
327 |
|
my $self = shift; |
328 |
|
return $self->{errors}; |
329 |
|
} |
330 |
|
|
331 |
|
=head2 report_error |
332 |
|
|
333 |
|
Produce nice humanly readable report of single error |
334 |
|
|
335 |
|
print $validate->report_error( $error_hash ); |
336 |
|
|
337 |
|
=cut |
338 |
|
|
339 |
$l =~ s/^\s+//; |
sub report_error { |
340 |
$l =~ s/\s+$//; |
my $self = shift; |
341 |
|
|
342 |
my @d = split(/\s+/, $l); |
my $h = shift || die "no hash?"; |
343 |
|
|
344 |
my $fld = shift @d || $log->logdie("need field name in line $curr_line: $l"); |
sub _unroll { |
345 |
|
my ($self, $tree, $accumulated) = @_; |
346 |
|
|
347 |
if (@d) { |
my $log = $self->_get_logger(); |
348 |
$v->{$fld}->{ref} = 'ARRAY'; |
|
349 |
$v->{$fld}->{sf} = \@d; |
$log->debug("# ", |
350 |
|
( $tree ? "tree: $tree " : '' ), |
351 |
|
( $accumulated ? "accumulated: $accumulated " : '' ), |
352 |
|
); |
353 |
|
|
354 |
|
my $results; |
355 |
|
|
356 |
|
if (ref($tree) ne 'HASH') { |
357 |
|
return ("$accumulated\t($tree)", undef); |
358 |
|
} |
359 |
|
|
360 |
|
my $dump; |
361 |
|
|
362 |
|
foreach my $k (sort keys %{ $tree }) { |
363 |
|
|
364 |
|
if ($k eq 'dump') { |
365 |
|
$dump = $tree->{dump}; |
366 |
|
#warn "## dump ",dump($dump),"\n"; |
367 |
|
next; |
368 |
|
} |
369 |
|
|
370 |
|
$log->debug("current: $k"); |
371 |
|
|
372 |
|
my ($new_results, $new_dump) = $self->_unroll($tree->{$k}, |
373 |
|
$accumulated ? "$accumulated\t$k" : $k |
374 |
|
); |
375 |
|
|
376 |
|
$log->debug( |
377 |
|
( $new_results ? "new_results: " . dump($new_results) ." " : '' ), |
378 |
|
); |
379 |
|
|
380 |
|
push @$results, $new_results if ($new_results); |
381 |
|
$dump = $new_dump if ($new_dump); |
382 |
|
|
383 |
|
} |
384 |
|
|
385 |
|
$log->debug( |
386 |
|
( $results ? "results: " . dump($results) ." " : '' ), |
387 |
|
); |
388 |
|
|
389 |
|
if ($#$results == 0) { |
390 |
|
return ($results->[0], $dump); |
391 |
} else { |
} else { |
392 |
$v->{$fld}->{ref} = ''; |
return ($results, $dump); |
393 |
} |
} |
394 |
|
} |
395 |
|
|
396 |
|
|
397 |
|
sub _reformat { |
398 |
|
my $l = shift; |
399 |
|
$l =~ s/\t/ /g; |
400 |
|
$l =~ s/_/ /g; |
401 |
|
return $l; |
402 |
} |
} |
403 |
|
|
404 |
$log->debug("current validation rules: ", Dumper($v)); |
my $out = ''; |
405 |
|
|
406 |
$self->{rules} = $v; |
for my $f (sort keys %{ $h }) { |
407 |
|
$out .= "$f: "; |
408 |
|
|
409 |
|
my ($r, $d) = $self->_unroll( $h->{$f} ); |
410 |
|
my $e; |
411 |
|
if (ref($r) eq 'ARRAY') { |
412 |
|
$e .= join(", ", map { _reformat( $_ ) } @$r); |
413 |
|
} else { |
414 |
|
$e .= _reformat( $r ); |
415 |
|
} |
416 |
|
$e .= "\n\t$d" if ($d); |
417 |
|
|
418 |
$self ? return $self : return undef; |
$out .= $e . "\n"; |
419 |
|
} |
420 |
|
return $out; |
421 |
|
} |
422 |
|
|
423 |
|
|
424 |
|
=head2 report |
425 |
|
|
426 |
|
Produce nice humanly readable report of errors |
427 |
|
|
428 |
|
print $validate->report; |
429 |
|
|
430 |
|
=cut |
431 |
|
|
432 |
|
sub report { |
433 |
|
my $self = shift; |
434 |
|
my $e = $self->{errors} || return; |
435 |
|
|
436 |
|
my $out; |
437 |
|
foreach my $mfn (sort { $a <=> $b } keys %$e) { |
438 |
|
$out .= "MFN $mfn\n" . $self->report_error( $e->{$mfn} ) . "\n"; |
439 |
|
} |
440 |
|
|
441 |
|
return $out; |
442 |
|
|
443 |
|
} |
444 |
|
|
445 |
|
=head2 delimiters_templates |
446 |
|
|
447 |
|
Generate report of delimiter tamplates |
448 |
|
|
449 |
|
my $report = $validate->delimiter_teplates( |
450 |
|
report => 1, |
451 |
|
); |
452 |
|
|
453 |
|
Options: |
454 |
|
|
455 |
|
=over 4 |
456 |
|
|
457 |
|
=item report |
458 |
|
|
459 |
|
Generate humanly readable report with single fields |
460 |
|
|
461 |
|
=back |
462 |
|
|
463 |
|
=cut |
464 |
|
|
465 |
|
sub delimiters_templates { |
466 |
|
my $self = shift; |
467 |
|
|
468 |
|
my $args = {@_}; |
469 |
|
|
470 |
|
my $t = $self->{_delimiters_templates}; |
471 |
|
|
472 |
|
my $log = $self->_get_logger; |
473 |
|
|
474 |
|
unless ($t) { |
475 |
|
$log->error("called without delimiters"); |
476 |
|
return; |
477 |
|
} |
478 |
|
|
479 |
|
my $out; |
480 |
|
|
481 |
|
foreach my $f (sort { $a <=> $b } keys %$t) { |
482 |
|
$out .= "$f\n" if ( $args->{report} ); |
483 |
|
foreach my $template (sort { $a cmp $b } keys %{ $t->{$f} }) { |
484 |
|
my $count = $t->{$f}->{$template}; |
485 |
|
$out .= |
486 |
|
( $count ? "" : "# " ) . |
487 |
|
( $args->{report} ? "" : "$f" ) . |
488 |
|
"\t$count\t$template\n"; |
489 |
|
} |
490 |
|
} |
491 |
|
|
492 |
|
return $out; |
493 |
|
} |
494 |
|
|
495 |
|
=head2 save_delimiters_templates |
496 |
|
|
497 |
|
=cut |
498 |
|
|
499 |
|
sub save_delimiters_templates { |
500 |
|
my $self = shift; |
501 |
|
|
502 |
|
my $path = $self->{delimiters_path}; |
503 |
|
|
504 |
|
return unless ( $path ); |
505 |
|
|
506 |
|
my $log = $self->_get_logger; |
507 |
|
|
508 |
|
open(my $d, '>', $path) || $log->fatal("can't open $path: $!"); |
509 |
|
print $d $self->delimiters_templates; |
510 |
|
close($d); |
511 |
|
|
512 |
|
$log->info("new delimiters templates saved to $path"); |
513 |
} |
} |
514 |
|
|
515 |
=head1 AUTHOR |
=head1 AUTHOR |