script/fb3_2_json.pl untested

filter: Statementstmt 0.0% 0 / 604 Branchbran 0.0% 0 / 236 Conditioncond 0.0% 0 / 152 MC/DCmcdc 0.0% 0 / 81 Subroutinesub 0.0% 0 / 41 Podpod 0.0% 0 / 20 totaltotal 0.0% 0 / 1134 CC 105 SCAR 56.1
CC
105
Cov
0%
DumpTree
870.0
ProceedNode
306.0
DumpRootLevelTOC
306.0
CRAP 273.2
1 #!/usr/local/bin/perl
2 use strict;
3  
4 use XML::LibXML;
5 use OPC::Node;
6 use FB3;
7 use utf8;
8 use Encode;
9 use open qw(:std :utf8);
10 use Image::Info;
11 use JSON::Path;
12 use JSON::PP;
13 use File::Copy;
14 use File::Basename;
15 use MIME::Base64;
16 use TeX::Hyphen;
17  
18 use Getopt::Long;
19  
20 my $FB3 = '';
21 my $Out = '';
22 my $Version = '1.0';
23 my $Lang = 'ru';
24 my $ArtID = undef;
25 my $Dictionary = undef;
26 my $Hyp = undef;
27  
28 GetOptions ('in|from|src|fb3=s' => \$FB3,
29 'out|to|dst|json=s' => \$Out,
30 'lang=s' => \$Lang,
31 'art|art-id=s' => \$ArtID,
32 'dict=s' => \$Dictionary,
33 'version=s' => \$Version) or print join('', <DATA>) and die("Error in command line arguments\n");
34  
35 print join('', <DATA>) and die "ERROR: source directory not specified, use --fb3 parameter\n" unless $FB3;
36 print join('', <DATA>) and die "ERROR: destination directory not specified, use --json parameter\n" unless $Out;
37  
38 die "\nERROR: source directory `$FB3' not found\n" unless -d $FB3;
39 die "\nERROR: destination directory `$Out' not found\n" unless -d $Out;
40  
41 $Out = $Out.'/' unless $Out =~ /\/$/;
42  
43 unless ($Version =~ /^\d+\.\d+$/) {
44 $Version = ($Version =~ /^\d+$/) ? "1.$Version" : "1.0"
45 }
46  
47 my %HyphenMin = (
48 'ru' => [2,2],
49 'uk' => [2,2],
50 'pl' => [2,2],
51 'en' => [2,3],
52 'de' => [2,2],
53 'fr' => [2,3],
54 'es' => [2,2],
55 'et' => [2,3],
56 'ka' => [2,2],
57 'lt' => [2,2],
58 'lv' => [2,2],
59 );
60  
61 my $CannotHyph;
62 if ($Dictionary) {
63 if (-e $Dictionary) {
64 $Hyp = new TeX::Hyphen 'file' => $Dictionary,
65 'style' => 'utf8', leftmin => ($HyphenMin{$Lang}[0] ? $HyphenMin{$Lang}[0] : 2),
66 rightmin => ($HyphenMin{$Lang}[1] ? $HyphenMin{$Lang}[1] : 2);
67 } else {
68 die "\nERROR: dictionary file `$Dictionary' not found\n"
69 }
70 } elsif ($Lang eq 'pl') {
71 $CannotHyph = 1;
72 }
73  
74 my $PartLimit = 20000;
75 my $IsTrial = 0;
76  
77 use constant {
78 RELATION_TYPE_FB3_BOOK =>
79 'http://www.fictionbook.org/FictionBook3/relationships/Book',
80 RELATION_TYPE_FB3_BODY =>
81 'http://www.fictionbook.org/FictionBook3/relationships/body',
82 RELATION_TYPE_OPC_THUMBNAIL =>
83 'http://schemas.openxmlformats.org/package/2006/relationships/metadata/thumbnail',
84 RELATION_TYPE_CORE_PROP =>
85 'http://schemas.openxmlformats.org/package/2006/relationships/metadata/core-properties',
86 RELATION_TYPE_FB3_IMAGES =>
87 'http://www.fictionbook.org/FictionBook3/relationships/image',
88 };
89  
90 use constant {
91 NS_XLINK => 'http://www.w3.org/1999/xlink',
92 NS_FB3_DESCRIPTION => 'http://www.fictionbook.org/FictionBook3/description',
93 NS_FB3_BODY => 'http://www.fictionbook.org/FictionBook3/body'
94 };
95  
96 my %AttrHash = (
97 'article' => 'artc',
98 'float' => 'fl',
99 'align' => 'aln',
100 'valign' => 'valn',
101 'bindto' => 'bnd',
102 'border' => 'brd',
103 'on-one-page' => 'op',
104 'colspan' => 'csp',
105 'rowspan' => 'rsp',
106 'class' => 'nc',
107 'autotext' => 'att',
108 'page-before' => 'pb',
109 'page-after' => 'pa',
110 'width' => 'wth',
111 'min-width' => 'minw',
112 'max-width' => 'maxw',
113 'clipped' => 'cl',
114 'first-char-pos' => 'fcp',
115 );
116  
117 my %LangDependentStr = (
118 'ru' => ['Конец ознакомительного фрагмента'],
119 'uk' => ['Кінець ознайомчого фрагмента'],
120 'en' => ['The end of the free preview'],
121 'de' => ['Ende des Fragments'],
122 'fr' => ['Fin du fragment'],
123 'az' => ['Fragmanın sonu'],
124 'be' => ['Канец азнаямленчага фрагмента'],
125 'bn' => ['ফ্র্যাগমেন্ট সম্পূর্ণ হয়েছে'],
126 'bg' => ['Край на фрагмента'],
127 'el' => ['Τέλος του θραύσματος'],
128 'ka' => ['ფრაგმენტის დასასრული'],
129 'da' => ['Fragmentet afsluttet'],
130 'es' => ['Fin del fragmento'],
131 'it' => ['Fine del frammento'],
132 'kk' => ['Фрагменттің соңы'],
133 'ca' => ['Fi del fragment'],
134 'ky' => ['Сот жазманын'],
135 'zn' => ['片段結束'],
136 'ko' => ['프래그먼트의 끝'],
137 'la' => ['Finis iudicii fragmentum'],
138 'lv' => ['Fragments beigas'],
139 'lt' => ['Fragmento pabaiga'],
140 'mk' => ['Крајот на фрагментот'],
141 'no' => ['Slutten av fragmentet'],
142 'pl' => ['Koniec fragmentu'],
143 #'pt' => ['Koniec fragmentu'],
144 'ro' => ['Sfârșitul fragmentului'],
145 'sr' => ['Фрагмент завршен'],
146 'sk' => ['Konec fragmenta'],
147 'fi' => ['Fragmentin loppu'],
148 'hbs' => ['Kraj fragmenta'],
149 'cs' => ['Konec fragmentu'],
150 'sv' => ['Fragmentet slutfört'],
151 'eo' => ['Fino de fragmento'],
152 'et' => ['Fragment lõpp'],
153 'ja' => ['断片の終わり'],
154 'nl' => ['Einde fragment'],
155 );
156  
157 $Lang = 'en' unless $LangDependentStr{$Lang}[0];
158  
159 my @AuthorsPriority = qw(author co_author dubious_author lecturer compiler screenwriter translator contributing_editor managing_editor editor editorial_board_member adapter conceptor rendering associated commentator consultant scientific_advisor recipient_of_letters corrector composer);
160  
161 my $NoHyphRe = qr/^(poem|epigraph|subtitle|title)$/; # Ноды, текст которых(и их потомков) нельзя переносить
162 my $Semiblocks = qr/^(epigraph|annotation|poem|stanza)$/;
163 my $BlockPParents = qr/^(section|annotation|epigraph|notebody)$/;
164 my $LineBreakChars = qr/[\-\/…\?\!\}\|–—]/;
165 my $SubtitleParents = qr/^(section)$/;
166  
167 # ------------------------ Hyphenation settings ------------------------------
168 use constant HYPHEN => "\x{AD}"; #visible in, e.g., Komodo Edit.
169  
170 my ($hyphenPatterns, $hyphenRegexPattern, $soglasnie, $glasnie, $znaki, $RgxSoglasnie, $RgxGlasnie, $RgxZnaki, $RgxNonChar);
171  
172 $hyphenPatterns = {
173 GSS => 'GS' . &HYPHEN . 'S',
174 SGSG => 'SG' . &HYPHEN . 'SG',
175 SQS => 'SQ' . &HYPHEN . 'S',
176 GG => 'G' . &HYPHEN . 'G',
177 SS => 'S' . &HYPHEN . 'S'
178 };
179 $hyphenRegexPattern = join "|",keys %{$hyphenPatterns};
180 $hyphenRegexPattern = qr/(.*)($hyphenRegexPattern){1}(.*)/o;
181  
182 $soglasnie = "bcdfghjklmnpqrstvwxzбвгджзйклмнпрстфхцчшщłćżźśńżçßґñšžčģķļņბგდვზთკლმნპჟრსტჳფქღყშჩცძწჭხჴჯჰჶ";
183 $glasnie = "aeiouyАОУЮИЫЕЭЯЁєіїўóąęéàèùâêîôûëïüÿæœäöõāēīūėįųაეჱიჲოუჷჵ";
184 $znaki = "ъьყ";
185  
186 $RgxSoglasnie = qr/[$soglasnie]/oi;
187 $RgxGlasnie = qr/[$glasnie]/oi;
188 $RgxZnaki = qr/[$znaki]/oi;
189 $RgxNonChar = qr/([^$soglasnie$glasnie$znaki]+)/oi; #в скобках, чтобы оно возвращалось при сплите.
190 # /----------------------- Hyphenation settings ------------------------------
191  
192 my $jsonC = JSON::PP->new->pretty->allow_barekey;
193  
194 my $FB3Package = FB3->new( from_dir => $FB3 );
195 my $FB3Body = $FB3Package->Body;
196  
197 my $Parser = XML::LibXML->new();
198 my $BodyDoc = $Parser->load_xml( string => $FB3Body->Content, huge => 1 );
199  
200 my $TOCHeader = ProceedDescr($FB3Package->Meta->Content);
201 my @Img = $FB3Body->Relations( type => RELATION_TYPE_FB3_IMAGES );
202 my @ImgFiles;
203 for (@Img) {
204 push @ImgFiles, basename($_->{'TargetFullName'});
205 }
206 my $ImgFileNum = 0; # номер файла для замены
207 my $ImgHash;
208 for my $Image (@Img) {
209 $ImgHash->{$Image->{Id}}->{name} = basename($Image->{'TargetFullName'});
210 unless ($ImgHash->{$Image->{Id}}->{name} =~ /^[A-Za-z0-9_\-]+\..+$/) { # Если имя файла странное - переименуем
211 $ImgHash->{$Image->{Id}}->{name} =~ /(\.[^.]+)$/;
212 while (grep('renamed_img_'.$ImgFileNum.$1 eq $_, @ImgFiles)) { # проверка, что такого имени файла у нас еще нет(если есть - следующий номер и т.д.)
213 $ImgFileNum++;
214 }
215 $ImgHash->{$Image->{Id}}->{name} = 'renamed_img_'.$ImgFileNum.$1;
216 $ImgFileNum++;
217 }
218 my $FN = $Out.$ImgHash->{$Image->{Id}}->{name};
219 my $PhysicalName = $FB3Package->{opc}->PhysicalNameByPartName($Image->{'TargetFullName'});
220 File::Copy::copy( $PhysicalName, $FN );
221 ($ImgHash->{$Image->{Id}}->{height}, $ImgHash->{$Image->{Id}}->{width}) = GetImgSize($FN);
222 }
223  
224 ProceedBody($FB3Body->Content);
225  
226 my $BlockN;
227 my @Parts;
228 sub ProceedBody {
229 my $BodyXML = shift;
230  
231 $BodyXML = DecodeUtf8($BodyXML);
232 $BodyXML =~ s/\r?\n\r?/ /g;
233 $BodyXML =~ s/([\s>])([^\s<>]+)(<note\s+[^>]*?role="(foot|end)note"[^>]*?>[^<]{1,10}<\/note>[,\.\?"'“”«»‘’;:\)…\/]?)/$1.HypheNOBR($2,$3)/ges;
234 my $JsonStr;
235 my $TOCStr;
236 my $TotalBlocks;
237 # Начинаем обход всех узлов дерева
238 my $Node = $BodyDoc->getDocumentElement();
239 $Node = PrepareBodyXML($Node, $BodyDoc);
240 my $NodeTree = ProceedNode($Node);
241 my $RootLevelTOC = DumpRootLevelTOC($NodeTree);
242 $JsonStr = DumpTree($NodeTree);
243 $TOCStr = '{'.$TOCHeader.'Body: ['.$RootLevelTOC.'],Parts:['.(join ",\n", @Parts).']}';
244 ProceedJsonBodyPart($TOCStr, 'toc.js');
245 return;
246 }
247  
248 my %InnerRefsHash;
249 sub ProceedNode {
250 my $Node = shift || return;
251 my $ParentName = shift || undef;
252 my $NoHyph = shift || 0;
253 my $Xp = shift || undef;
254 my $NoCut = shift;
255 my $NoteType = shift;
256 my $InEmptySection = shift || 0;
257 my $NodeName = $Node->nodeName;
258  
259 my $NodeHash;
260  
261 if ($NoHyph || $NodeName =~ /$NoHyphRe/) {
262 $NodeHash->{no_hyph} = 1;
263 }
264  
265 $Xp = 1 if $NodeName eq 'fb3-body';
266  
267 if ($NodeName eq '#text') {
268 my $Text = $Node->nodeValue;
269 return if $Text =~ /^\s+$/;
270 return unless defined $Text;
271 $Text = '['.$Text.']' if $NoteType eq 'footnote';
272 $Text = '{'.$Text.'}' if $NoteType eq 'endnote';
273 $NodeHash->{text} = $Text;
274 $NodeHash->{xp} = $Xp;
275 return $NodeHash;
276 }
277  
278 for my $Attr ($Node->attributes) {
279 my $AttrName = $Attr->getName;
280 $AttrName =~ s/^.+://; # Откусим ns
281 if ($AttrName eq 'empty') {
282 $NodeHash->{in_empty_section} = 1;
283 } else {
284 $NodeHash->{attr}->{$AttrName} = EscString($Attr->getValue);
285 }
286 }
287  
288 $NodeHash->{in_empty_section} = 1 if $InEmptySection;
289 $NoteType = $NodeHash->{attr}->{role};
290  
291 $NodeHash->{xp} = $Xp;
292  
293 $NodeHash->{name} = $NodeName;
294  
295 my ($IsBlock, $IsRootBlock, $Printable) = AnalyseNode($Node, $NodeName, $ParentName);
296 $Printable = 0 if $NodeHash->{in_empty_section};
297  
298 $NodeHash->{no_cut} = $NoCut;
299 if ($Printable && !$NodeHash->{no_cut}) {
300 $NodeHash->{b_id} = $BlockN || '0';
301 $NoCut = 1;
302 $BlockN++;
303 }
304  
305 $NodeHash->{pr} = $Printable;
306 $NodeHash->{rb} = $IsRootBlock;
307  
308 my @Childs;
309 my $i = 1;
310 for my $Child ($Node->childNodes) {
311 my $ChildHash = ProceedNode($Child, $NodeName, $NodeHash->{no_hyph}, $Xp.','.$i, $NoCut, $NoteType, $NodeHash->{in_empty_section});
312 next unless defined $ChildHash;
313 push @Childs, $ChildHash;
314 $i++;
315 }
316 $NodeHash->{c} = [@Childs] if scalar @Childs;
317  
318 my $Id = $Node->getAttribute('id');
319 if ($Id) {
320 if ($NodeHash->{pr}) {
321 $InnerRefsHash{$Id} = $NodeHash->{b_id};
322 } else {
323 MoveRefToPrintableChild($NodeHash, $Id); # Если этот блок не отражается в json, то его первому потомку, который отражается
324 }
325 }
326  
327 my $Href = $NodeHash->{attr}->{href};
328  
329 if (defined $Href) {
330 $Href =~ s/^\#//g;
331 $NodeHash->{href} = $Href;
332 delete $NodeHash->{attr}->{href};
333 }
334  
335 return $NodeHash;
336 }
337  
338 sub MoveRefToPrintableChild {
339 my $NodeHash = shift;
340 my $Id = shift;
341  
342 foreach (@{$NodeHash->{c}}) {
343 if ($_->{pr} && $_->{b_id}) { # Потомок нам подходит
344 $InnerRefsHash{$Id} = $_->{b_id};
345 return 1;
346 } else { # Проходим по его потомкам
347 if (MoveRefToPrintableChild($_, $Id)) {
348 return 1;
349 }
350 }
351 }
352  
353 }
354  
355 my @ResultArr;
356 my ($Length, $FirstBlockN, $LastBlockN, $FirstXP, $LastXP);
357 my $FileN = 0;
358 sub DumpTree {
359 my $NodeHash = shift || return;
360 my $JsonStr;
361  
362 if (defined $NodeHash->{text} || $NodeHash->{pr}) {
363 $FirstXP = $NodeHash->{xp} if !$FirstXP && $NodeHash->{b_id};
364 $LastXP = $NodeHash->{xp} if $NodeHash->{b_id};
365 }
366  
367 if (defined $NodeHash->{text}) {
368 $Length += length($NodeHash->{text});
369 $JsonStr = ProceedTextNode($NodeHash->{text}, $NodeHash->{no_hyph} == 1 ? 0 : 1);
370 return $JsonStr;
371 } elsif ($NodeHash->{pr} == 1) {
372 my $AttrStr = join ( ',', map { ($AttrHash{$_} ? $AttrHash{$_} : $_).':"'.$NodeHash->{attr}->{$_}.'"' } keys %{$NodeHash->{attr}}); # Дампим все атрибуты в строку
373 $AttrStr = ','.$AttrStr if $AttrStr;
374  
375 if ($NodeHash->{name} eq 'br') {
376 $JsonStr = '{t:"'.$NodeHash->{name}.'"'.$AttrStr.',xp:['.$NodeHash->{xp};
377 } elsif ($NodeHash->{name} eq 'img') {
378 $JsonStr = '{t:"'.$NodeHash->{name}.'"'.$AttrStr.',xp:['.$NodeHash->{xp}.'],s:"'.$ImgHash->{$NodeHash->{attr}->{src}}->{name}.'"';
379 $JsonStr .= ',w:' . ($ImgHash->{$NodeHash->{attr}->{src}}->{width}) if ($ImgHash->{$NodeHash->{attr}->{src}}->{width});
380 $JsonStr .= ',h:' . ($ImgHash->{$NodeHash->{attr}->{src}}->{height}) if ($ImgHash->{$NodeHash->{attr}->{src}}->{height});
381 $JsonStr .= '}';
382 } else {
383 $JsonStr = '{t:"'.$NodeHash->{name}.'"'.$AttrStr.',xp:['.$NodeHash->{xp}.'],c:[';
384 }
385  
386 $JsonStr .= "\n" if $NodeHash->{name} =~ /$Semiblocks/;
387 } else {
388 $JsonStr = '';
389 }
390 my $ChildsCount = scalar @{$NodeHash->{c}} if $NodeHash->{c};
391  
392 $FirstBlockN = ($NodeHash->{b_id}||0) unless defined $FirstBlockN;
393 $LastBlockN = defined $NodeHash->{b_id} ? $NodeHash->{b_id} : $LastBlockN ? $LastBlockN : 0;
394  
395 for my $ChildHash (@{$NodeHash->{c}}) {
396 $ChildsCount--;
397 $JsonStr =~ s/,$/],f:/g if $ChildHash->{name} eq 'footnote'; # Лучше бы что другое придумать, но работает
398 my $ChildStr .= DumpTree($ChildHash);
399 if ($ChildStr) {
400 $JsonStr .= $ChildStr;
401 $JsonStr .= ',' if $ChildsCount && $ChildStr;
402 }
403 }
404 $JsonStr .= "\n" if $NodeHash->{name} =~ /$Semiblocks/;
405  
406 if ( defined $NodeHash->{href}) {
407 if ($InnerRefsHash{$NodeHash->{href}}) {
408 $JsonStr .= '],hr:['.$InnerRefsHash{$NodeHash->{href}}.']';
409 } else {
410 $JsonStr .= '],href:["'.$NodeHash->{href}.'"]';
411 }
412 }
413  
414 $JsonStr .= ']}' if ($NodeHash->{pr} == 1 && $NodeHash->{name} ne 'img' && $NodeHash->{name} ne 'note' && $NodeHash->{name} ne 'a');
415 $JsonStr .= '}' if ($NodeHash->{name} eq 'note' || $NodeHash->{name} eq 'a');
416  
417 if ($NodeHash->{pr} == 1 && !$NodeHash->{no_cut}) {
418 push @ResultArr, $JsonStr;
419 }
420  
421 if (($NodeHash->{b_id} && $Length > $PartLimit) || ($NodeHash->{name} eq 'fb3-body' && $Length > 0)) {
422 my $ResultStr = join ",\n",@ResultArr;
423 $ResultStr =~ s/,\n$//g;
424 if (trim($ResultStr)) {
425 $ResultStr = '['.$ResultStr.']';
426 my $FileName = sprintf("%03i.js",$FileN);
427 ProceedJsonBodyPart($ResultStr, $FileName);
428 $FileN++;
429 @ResultArr = ();
430 $Length = 0;
431 push @Parts, '{s:'.$FirstBlockN.',e:'.$LastBlockN.',xps:['.$FirstXP.'],xpe:['.$LastXP.'],url:"'.$FileName.'"}';
432 $FirstBlockN = undef;
433 $FirstXP = undef;
434 }
435 }
436  
437 return $JsonStr;
438 }
439  
440 my $LastBlock;
441 sub DumpRootLevelTOC {
442 my $NodeHash = shift || return;
443 my $TOCStr;
444 my @TocStrArr;
445  
446 $LastBlock = $NodeHash->{b_id} if $NodeHash->{b_id};
447 if ($NodeHash->{rb}) {
448 $NodeHash->{b_id} ||= $NodeHash->{c}[0]->{b_id};
449 $LastBlock = $NodeHash->{b_id} if $NodeHash->{b_id};
450 push @TocStrArr, 's:'.($LastBlock ? $LastBlock : '0') unless $NodeHash->{in_empty_section};
451  
452 my @Clilds;
453 my $ChildsSection;
454 my $Title;
455 my $TotalClipped = 0;
456 if ($NodeHash->{c}[0]->{name} eq 'title') {
457 $NodeHash->{c}[0]->{'intitle'} = 1;
458 $Title = 't:"'.ExtractText($NodeHash->{c}[0]).'"';
459 }
460 for my $ChildHash (@{$NodeHash->{c}}) {
461 if ($ChildHash->{name} eq 'clipped') {
462 $TotalClipped = 1;
463 }
464 my $ChildStr = DumpRootLevelTOC($ChildHash);
465 push @Clilds, $ChildStr if $ChildStr;
466 }
467 if (scalar @Clilds) {
468 $ChildsSection = 'c:[';
469 $ChildsSection .= join ",",@Clilds;
470 $ChildsSection .= ']';
471 }
472 push @TocStrArr, 'e:'.($LastBlock ? $LastBlock : '0') unless $NodeHash->{in_empty_section};
473 push @TocStrArr, $Title if $Title;
474 push @TocStrArr, $AttrHash{'first-char-pos'}.':'.$NodeHash->{attr}->{'first-char-pos'} if $NodeHash->{attr}->{'first-char-pos'};
475 push @TocStrArr, $AttrHash{clipped}.':"true"' if $NodeHash->{attr}->{clipped} eq 'true';
476 push @TocStrArr, 'tcl:"true"' if $TotalClipped;
477 push @TocStrArr, $ChildsSection if $ChildsSection;
478 $TOCStr = '{';
479 $TOCStr .= join ",",@TocStrArr;
480 $TOCStr .= '}';
481 }
482 return $TOCStr;
483 }
484  
485 sub ExtractText {
486 my $NodeHash = shift || return;
487 my @TextArr;
488  
489 for my $ChildHash (@{$NodeHash->{c}}) {
490  
491 $ChildHash->{'intitle'} = 1 if exists $NodeHash->{'intitle'} && $NodeHash->{'intitle'};
492 #note внутри title нам не нужны в оглавлении
493 next if $ChildHash->{'name'} eq 'note' && exists $ChildHash->{'intitle'} && $ChildHash->{'intitle'};
494  
495 push @TextArr, $ChildHash->{text} if $ChildHash->{text};
496 my $Text = ExtractText($ChildHash);
497 $Text = EscString($Text);
498 push @TextArr, $Text if $Text;
499 }
500  
501 return join ' ', @TextArr;
502 }
503  
504 sub PrepareBodyXML {
505 my $Node = shift || undef;
506 my $BodyDoc = shift;
507  
508 my $FootNoteHash;
509 my $xpc = XML::LibXML::XPathContext->new($Node);
510 $xpc->registerNs('fbb', &NS_FB3_BODY);
511 for my $Note ($xpc->findnodes('/fbb:fb3-body/fbb:notes', $Node)) {
512 if ($Note->getAttribute('show') == 0) {
513 for my $NoteBody ($xpc->findnodes('./fbb:notebody', $Note)) {
514 $FootNoteHash->{$NoteBody->getAttribute('id')} = ($xpc->findnodes('./*', $NoteBody));
515 }
516 $Note->unbindNode();
517 }
518 }
519  
520 for my $NoteRef ($xpc->findnodes('//fbb:note')) {
521 if ($NoteRef->getAttributeNS(NS_XLINK, 'role') eq 'footnote') {
522 my $Id = $NoteRef->getAttribute('href');
523 $NoteRef->removeAttribute('href');
524 my $FootNoteNode = $NoteRef->appendChild($BodyDoc->createElement('footnote'));
525 for my $N (@{$FootNoteHash->{$Id}}) {
526 $FootNoteNode->appendChild($N);
527 }
528 }
529 }
530  
531 if ($IsTrial) {
532 for my $SectionNode ($xpc->findnodes('//fbb:section', $Node)) {
533 if (IsEmptySection($SectionNode)) {
534 $SectionNode->setAttribute('empty', 1)
535 }
536 }
537 my $NewSectionNode = $BodyDoc->createElementNS('fbb', 'section');
538 my $NewPNode = $BodyDoc->createElementNS('fbb', 'p');
539 $NewPNode->appendChild($BodyDoc->createTextNode($LangDependentStr{$Lang}[0]));
540 $NewSectionNode->appendChild($NewPNode);
541 my $FB3BodyNode = $xpc->findnodes('fbb:fb3-body', $BodyDoc)->[0];
542 $FB3BodyNode->appendChild($NewSectionNode);
543 }
544  
545 return $Node;
546 }
547  
548 sub IsEmptySection {
549 my $SectionNode = shift || return;
550  
551 my $IsEmpty;
552 for my $Child ($SectionNode->nonBlankChildNodes) {
553 my $ChildName = $Child->nodeName;
554 if ($ChildName ne 'clipped' && $ChildName ne 'title' && $ChildName ne 'section') {
555 return 0;
556 } elsif ($ChildName eq 'clipped') {
557 return 1;
558 } elsif ($ChildName eq 'section') {
559 $IsEmpty = IsEmptySection($Child);
560 return 0 unless $IsEmpty;
561 }
562 }
563 return 1 if $IsEmpty;
564 }
565  
566 sub ProceedJsonBodyPart {
567 my $JsonStr = shift;
568 my $FileName = shift;
569  
570 my $JData;
571 eval { $JData = $jsonC->decode($JsonStr); };
572 if ($@) {
573 # Такой json нам не нужен
574 die "$JsonStr\n===============\n$@";
575 #warn "\n$FileName: $@"; # Для отладки и теста. В боевых условиях падать, если фигня получается.
576 }
577  
578 open TMPOUT, ">", $Out.$FileName or die "Cannot open tmp file: `$Out.$FileName'";
579 print TMPOUT $JsonStr;
580 close TMPOUT;
581  
582 return $FileName;
583 }
584  
585 sub ProceedDescr {
586  
587 my $DescrXML = shift;
588  
589 $DescrXML = DecodeUtf8($DescrXML);
590 $DescrXML =~ s/\r?\n\r?/ /g;
591  
592 my @description_data = ();
593  
594 my $Parser = XML::LibXML->new();
595 my $xpc = XML::LibXML::XPathContext->new($Parser->load_xml( string => $DescrXML, huge => 1 ));
596 $xpc->registerNs('fbd', &NS_FB3_DESCRIPTION);
597  
598 my $UUID = ($xpc->findnodes('/fbd:fb3-description')->[0])->getAttribute('id');
599  
600 my $SimpleFields = {
601 'Title' => 'fbd:title/fbd:main',
602 'Subtitle' => 'fbd:title/fbd:sub',
603 'Lang' => 'fbd:lang',
604 'Annotation' => 'fbd:annotation',
605 'Preamble' => 'fbd:preamble',
606 'Translated' => 'fbd:translated'
607 };
608  
609 for my $field ( sort keys %$SimpleFields ) {
610  
611 my $node = $xpc->findnodes('/fbd:fb3-description/' . $SimpleFields->{$field})->[0] || next;
612 my $value = EscString($node->string_value) || next;
613  
614 push @description_data, sprintf('%s:"%s"', $field, $value);
615 }
616  
617 my $description = join ',', @description_data;
618  
619 my @Sequences = ();
620 my $getSequence; $getSequence = sub {
621  
622 my @sequenceNodes = @_;
623  
624 for my $sequenceNode ( @sequenceNodes ) {
625  
626 push @Sequences, '"' . EscString($xpc->findnodes('./fbd:title/fbd:main', $sequenceNode)->[0]->string_value) . '"';
627  
628 $getSequence->( $xpc->findnodes('./fbd:sequence', $sequenceNode) );
629 }
630 };
631 $getSequence->( $xpc->findnodes('/fbd:fb3-description/fbd:sequence') );
632 $description .= ',Sequences:[' . join(',', @Sequences) . ']' if scalar @Sequences;
633  
634 my $getParts = sub {
635  
636 my $node = shift;
637 my $struct = shift;
638  
639 my @values = ();
640  
641 for my $part ( keys %$struct ) {
642  
643 my $subNode = $xpc->findnodes('./fbd:' . $struct->{$part}, $node)->[0] || next;
644 my $value = EscString($subNode->string_value) || next;
645  
646 push @values, sprintf('%s:"%s"', $part, $value);
647 }
648  
649 return join(',', @values);
650 };
651  
652 my $getAuthorNamePart = sub {
653  
654 my $node = shift;
655  
656 my $NameParts = {'First' => 'first-name', 'Last' => 'last-name', 'Middle' => 'middle-name'};
657  
658 return $getParts->($node, $NameParts);
659 };
660  
661 my $Created = $xpc->findnodes('/fbd:fb3-description/fbd:document-info/@created');
662 $description .= ',Created:"'.$Created.'"' if $Created;
663 my $Updated = $xpc->findnodes('/fbd:fb3-description/fbd:document-info/@updated');
664 $description .= ',Updated:"'.$Updated.'"' if $Updated;
665  
666 if ( my $WrittenNode = $xpc->findnodes('/fbd:fb3-description/fbd:written')->[0] ) {
667  
668 my $WrittenParts = {'Date' => 'date', 'DatePublic' => 'date-public', 'Lang' => 'lang'};
669 my $Written = $getParts->($WrittenNode, $WrittenParts);
670 $description .= ',Written:{' . $Written . '}' if scalar $Written;
671 }
672  
673 my @Authors;
674 foreach (@AuthorsPriority) {
675 for my $Author ($xpc->findnodes('/fbd:fb3-description/fbd:fb3-relations/fbd:subject[@link="'.$_.'"]')) {
676 push @Authors, '{Role:"' . $_ . '",' . $getAuthorNamePart->($Author) . '}';
677 }
678 }
679 $description .= ',Authors:[' . (join ",", @Authors) . ']' if scalar @Authors;
680  
681 my @Translators;
682 for my $Translator ($xpc->findnodes('/fbd:fb3-description/fbd:fb3-relations/fbd:subject[@link="translator"]')) {
683 push @Translators, '{Role:"translator",' . $getAuthorNamePart->($Translator) . '}';
684 }
685 $description .= ',Translators:[' . (join ",", @Translators) . ']' if scalar @Translators;
686  
687 my @Relations;
688 for my $ObjectNode ($xpc->findnodes('/fbd:fb3-description/fbd:fb3-relations/fbd:object')) {
689  
690 my $ObjectId = EscString( $ObjectNode->getAttribute('id') );
691 my $ObjectType = EscString( $ObjectNode->getAttribute('link') );
692 my $ObjectTitle = EscString($xpc->findnodes('./fbd:title/fbd:main', $ObjectNode)->[0]->string_value);
693  
694 push @Relations, sprintf('{id:"%s",type:"%s",title:"%s"}', $ObjectId, $ObjectType, $ObjectTitle);
695 }
696 $description .= ',Relations:[' . (join ",", @Relations) . ']' if scalar @Relations;
697  
698 $description .= ',ArtID:"' . EscString($ArtID) . '"' if $ArtID;
699  
700 my $DraftStr = '';
701 if ( my $FragmentNode = $xpc->findnodes('/fbd:fb3-description/fbd:draft-status')->[0]) {
702 $DraftStr = ',DraftStatus:{expected_chars:"'.$FragmentNode->getAttribute('expected-chars').'"'; #required
703 if (my $ExpFreq = $FragmentNode->getAttribute('expected-frequency')) {
704 $DraftStr .= ',expected_frequency:"'.$ExpFreq.'"';
705 }
706 if (my $ExpRelease = $FragmentNode->getAttribute('expected-release')) {
707 $DraftStr .= ',expected_release:"'.$ExpRelease.'"';
708 }
709 $DraftStr .= '}';
710 $description .= $DraftStr;
711 }
712  
713 my $LengthStr = '';
714 if ( my $FragmentNode = $xpc->findnodes('/fbd:fb3-description/fbd:fb3-fragment')->[0] ) {
715 $IsTrial = 1;
716 $LengthStr = ",\n".'"fb3-fragment":{"full_length":'.$FragmentNode->getAttribute('full_length').',"fragment_length":'.$FragmentNode->getAttribute('fragment_length').'}';
717 } else {
718 my $RootNode = $BodyDoc->getDocumentElement();
719 my $CharsFull = length($RootNode->textContent);
720 $xpc->registerNs('fbb', &NS_FB3_BODY);
721 foreach my $TrialOnlyNode ($xpc->findnodes('/fbb:fb3-body/fbb:section[@output="trial-only"]',$RootNode)) {
722 $CharsFull -= (length($TrialOnlyNode->textContent) || 0);
723 }
724 $LengthStr = ",\n".'"full_length":'.$CharsFull;
725 }
726  
727 return 'Meta:{' . $description . ',UUID:"' . $UUID . '",version:"' . $Version . '"}' . $LengthStr . ",\n";
728 }
729  
730 sub DecodeUtf8 {
731 my $Out = shift;
732 unless (Encode::is_utf8($Out)) {
733 $Out = Encode::decode_utf8($Out);
734 }
735 return $Out;
736 }
737  
738 sub EscString{
739 my $Esc=shift;
740 return unless defined $Esc;
741 $Esc = DecodeUtf8($Esc." ");
742 $Esc =~ s/(["\\])/\\$1/g;
743 $Esc =~ s/(?:\r?\n\r?)|(?:\t)/ /g;
744 $Esc =~ s/ $//;
745 return $Esc;
746 }
747  
748 my %HyphCache;
749 sub ProceedTextNode{
750 my $Str=shift;
751 my $NeedHyph = shift;
752 return unless defined $Str;
753 $Str = EscString($Str);
754 return if $Str =~ /^\s+$/;
755 my $SRC = $Str;
756 if ($NeedHyph){
757 $Str = $HyphCache{$SRC} || HyphString($Str);
758 }
759 $Str =~ s/[ \t]+/ ","/g;
760 $Str =~ s/($LineBreakChars)(?![ "])/$1","/g;
761 $Str =~ s/"",|,""|","$//g;
762 if ($NeedHyph){
763 $HyphCache{$SRC} = $Str;
764 $Str =~ s/\x{AD}/","\x{AD}/g; # Full version
765 }
766 $Str = '"'.$Str.'"';
767 return $Str;
768 }
769  
770 sub AnalyseNode {
771 my $Node = shift || return;
772 my $NodeName = shift;
773 my $ParentName = shift;
774  
775 my $IsBlock = 0;
776 my $IsRootBlock = 0;
777 my $Printable = 1;
778  
779 if ($NodeName eq 'section' || $NodeName eq 'fb3-body' || $NodeName eq 'notes' || $NodeName eq 'notebody') {
780 $IsBlock = 1;
781 $IsRootBlock = 1;
782 $Printable = 0;
783 } elsif (($NodeName eq 'p' && $ParentName =~ /$BlockPParents/) || $NodeName eq 'title' || $NodeName eq 'subtitle' || $NodeName eq 'epigraph' ||
784 $NodeName eq 'annotation' || $NodeName eq 'div' || $NodeName eq 'blockquote' || $NodeName eq 'subscription') {
785 $IsBlock = 1;
786 }
787  
788 return ($IsBlock, $IsRootBlock, $Printable);
789 }
790  
791 sub HypheNOBR {
792 my ($Word, $NOBRCharSeq) = @_;
793  
794 # $Word = EscString($Word);
795 my $Esc = $HyphCache{$Word} || HyphString($Word);
796  
797 unless ($Esc =~ s/\xAD?([^\xAD]+)$/<nobr>$1/s) {
798 $Esc = '<nobr>'.$Esc;
799 }
800 $Esc =~ s/\xAD//gis;
801  
802 return $Esc . $NOBRCharSeq . '</nobr>';
803 }
804  
805 sub GetImgSize {
806 my $File = shift;
807  
808 open(my $TempStdErr, ">&STDERR"); # Image::Info много и не по делу говорит, воткнём кляп.
809 close STDERR;
810 my $ImgInfo = Image::Info::image_info($File);
811 open(STDERR, ">&", $TempStdErr); # А вот остальных послушаем.
812  
813 $ImgInfo->{height} =~ s/\D//g;
814 my $Height = $ImgInfo->{height};
815 $ImgInfo->{width} =~ s/\D//g;
816 my $Width = $ImgInfo->{width};
817  
818 return ($Height, $Width);
819 }
820  
821 sub trim {
822 my $str = shift;
823 $str =~ s/^\s+//s;
824 $str =~ s/\s+$//s;
825 return $str;
826 }
827  
828 # ------------------------ Hyphenation functions ------------------------------
829  
830 sub HyphString {
831 use utf8;
832 my $word = shift;
833 return $word if $CannotHyph;
834 my @wordArrayWithUnknownSymbols = split $RgxNonChar , $word; #собрали все слова и неизвестные символы. Для слова "пример!№?;слова" будет содержать "пример", "!№?;", "слова".
835  
836 for my $word (@wordArrayWithUnknownSymbols) {
837 next if $word =~ $RgxNonChar;
838 if (-e $Dictionary) {
839 $word = $Hyp->visualize($word);
840 $word =~ s/-/\x{AD}/g;
841 } else {
842 next if $Lang eq 'pl';
843 $word = HyphParticularWord($word);
844 }
845 }
846 return join "", @wordArrayWithUnknownSymbols;
847 }
848  
849 sub HyphParticularWord {
850 use utf8;
851 my $word = shift;
852  
853 my $softHyphMinPart = 2;
854  
855 return $word if ( length($word) < 2 * $softHyphMinPart + 1 || $word eq uc($word));
856 my $wordCopy = $word; #чтобы сохранить оригинальное слово. А $word заменим структурным эквивалентном
857 $word =~ s/$RgxSoglasnie/S/g;
858 $word =~ s/$RgxGlasnie/G/g;
859 $word =~ s/$RgxZnaki/Q/g;
860 while ($word =~ s/$hyphenRegexPattern/Hyphenate($1,$2,$3,\$wordCopy,$softHyphMinPart)/ge) {}
861 return $wordCopy;
862 }
863  
864 sub Hyphenate {
865 use utf8;
866 my ($leftFromPattern,$pattern,$rightFromPattern,$wordCopyRef,$softHyphMinPart) = @_;
867 my $leftOffsetOfCurrentHyphen = length($leftFromPattern) + index($hyphenPatterns->{$pattern},&HYPHEN);
868 my $rightOffsetOfCurrentHyphen = length(${$wordCopyRef}) - $leftOffsetOfCurrentHyphen; #слева дефисы не добавляются. Они добавляются справа
869  
870 substr(${$wordCopyRef}, 0, $leftOffsetOfCurrentHyphen) .= &HYPHEN
871 if ($leftOffsetOfCurrentHyphen >= $softHyphMinPart && $rightOffsetOfCurrentHyphen >= $softHyphMinPart);
872 #переносы ставим только если остается у нас в конце и в начале по softHyphMinPart символов
873  
874 return $leftFromPattern . $hyphenPatterns->{$pattern} . $rightFromPattern; #новую структуру кидаем в структурный эквивалент
875 }
876  
877 1;
878  
879 __DATA__