gfl-indexer.xslt 32.9 KB
Newer Older
Dennis Neumann's avatar
Dennis Neumann committed
1
2
<?xml version="1.0" encoding="utf-8"?>

Dennis Neumann's avatar
Dennis Neumann committed
3
4
<!-- 

Dennis Neumann's avatar
Dennis Neumann committed
5
This script produces two types of Solr XML documents, representing letters and individual pages of those letters.
Dennis Neumann's avatar
Dennis Neumann committed
6

Dennis Neumann's avatar
Dennis Neumann committed
7
8
9
10
11
12
13
The letter documents contain many metadata fields and two text fields. The metadata fields are
values that are copied from the TEI files. The text fields are described next.

Field 'fulltext'

This is just plain text copied from the complete TEI file. Line breaks and page beginnings are
transformed to whitespace. This field can be useful for searching or for generating highlighting snippets. 
Dennis Neumann's avatar
Dennis Neumann committed
14
15
16

Field 'fulltext_html'

Dennis Neumann's avatar
Dennis Neumann committed
17
18
19
20
21
22
This field  contains the HTML representation of the text of a TEI file (e. g. a Goethe letter).
A letter that consists of several pages is split on-the-fly into those pages.
However, the pages are all kept in this one field. The templates that are used here are the same
as the ones for the individual page documents (see below).

The letters are composed of different parts, for example 'opener', 'closer', 'salute'.
Dennis Neumann's avatar
Dennis Neumann committed
23
24
25
All those parts are represented here as <div>'s with the corresponding CSS classes.
The frontend viewer must decide how to format those parts and present them to the user.

Dennis Neumann's avatar
Dennis Neumann committed
26
The original TEI files contain mark-up for many in-text parts, like dates, names, underlined words, etc.
Dennis Neumann's avatar
Dennis Neumann committed
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Most of these are also transformed to <div>'s with their own CSS classes.
Although the in-text parts are by nature inline elements, we use here <div>'s and not <span>'s.
The reason is that Solr seems to have problems when highlighting fields that contain <span>'s
by sometimes producing corrupt HTML.
By using <div>'s, we avoid this problem.
In the frontend, these <div>'s must be set to 'display: inline'.

Some other in-text parts are transformed to special HTML elements.
For example, superscripted text is marked as <sup>, because HTML offers the appropriate element.

The project is still continuing and new TEI files are being produced.
That's why there might be new elements in the future that cannot be handled yet in this script.
The text of such TEI elements is enclosed in HTML elements of class 'unknown-element'.
Furthermore, a warning message is generated that contains data of the first occurrence of such a new element.


Dennis Neumann's avatar
Dennis Neumann committed
43
44
45
46
The second kind of documents that are produced are page documents.
The resulting pages are in the HTML format.
As the TEI file is processed, the TEI XML structure is split into pages using 
the page beginning elements (<pb/>).
Dennis Neumann's avatar
Dennis Neumann committed
47
Refer to comments in the code to understand the used algorithm.
Dennis Neumann's avatar
Dennis Neumann committed
48

49
-->
Dennis Neumann's avatar
Dennis Neumann committed
50

Dennis Neumann's avatar
Dennis Neumann committed
51
<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
52
   xpath-default-namespace="http://www.tei-c.org/ns/1.0" xmlns:gfl="http://sub.gfl.de"
bettin5's avatar
bettin5 committed
53
54
   xmlns:xs="http://www.w3.org/2001/XMLSchema" 
   xmlns:saxon="http://saxon.sf.net/" exclude-result-prefixes="gfl saxon xs"
bettin5's avatar
bettin5 committed
55
   xmlns:fn="http://www.w3.org/2005/02/xpath-functions"
bettin5's avatar
bettin5 committed
56
   xmlns:marc="http://www.loc.gov/MARC21/slim">
57

Dennis Neumann's avatar
Dennis Neumann committed
58
   <xsl:output method="xml" indent="yes" saxon:suppress-indentation="div" />
Dennis Neumann's avatar
Dennis Neumann committed
59
   <xsl:strip-space elements="*" />
60
   <xsl:preserve-space elements="msIdentifier bibl p" />
61
62

   <xsl:template match="/">
63
64
65
66
      <xsl:apply-templates select="TEI" />
   </xsl:template>

   <xsl:template match="TEI">
67
68
      <add>
         <doc>
69
            <xsl:apply-templates select="teiHeader| text" />
Dennis Neumann's avatar
Dennis Neumann committed
70
71
72
73
74
            <xsl:for-each select="distinct-values(//@xml:lang)">
               <field name="language">
                  <xsl:value-of select="." />
               </field>
            </xsl:for-each>
75
            <xsl:apply-templates select="facsimile" />
bettin5's avatar
bettin5 committed
76
            <xsl:apply-templates select="//body/listBibl" />       
bettin5's avatar
bettin5 committed
77
         </doc>  
78
         <xsl:apply-templates select="text" mode="page_splitting" />
bettin5's avatar
bettin5 committed
79
         <xsl:apply-templates select="//@ref" />
bettin5's avatar
bettin5 committed
80
      </add>
81
   </xsl:template>
bettin5's avatar
bettin5 committed
82
83
   
   <xsl:template match="@ref">
bettin5's avatar
bettin5 committed
84
      <xsl:if test=". != 'gnd:_' and ./parent::name/text() and ./parent::name/@type">
bettin5's avatar
bettin5 committed
85
         <doc>
bettin5's avatar
bettin5 committed
86
87
            <field name="entity_name">
               <xsl:value-of select="./parent::name/normalize-space()"/> 
88
            </field>    
bettin5's avatar
bettin5 committed
89
90
91
92
93
            <field name ="doctype">
               <xsl:value-of select="./parent::name/@type"/>
            </field>
            <field name="id">  
               <xsl:value-of select="."/>
bettin5's avatar
bettin5 committed
94
95
            </field>
            <xsl:variable name="gnd_number">
96
               <xsl:value-of select="normalize-space(substring-after(., 'gnd:'))"/>
bettin5's avatar
bettin5 committed
97
98
            </xsl:variable>
            <xsl:variable name="uri_path">
99
               <xsl:value-of select="concat('https://d-nb.info/gnd/', $gnd_number, '/about/marcxml')"/>
bettin5's avatar
bettin5 committed
100
            </xsl:variable>
bettin5's avatar
bettin5 committed
101
102
103
104
105
            <xsl:choose>
               <xsl:when test="doc-available($uri_path)">
                  <xsl:apply-templates select="doc($uri_path)/marc:record" />
               </xsl:when>
            </xsl:choose>
bettin5's avatar
bettin5 committed
106
107
108
109
         </doc>
      </xsl:if>
   </xsl:template>
   
bettin5's avatar
bettin5 committed
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
   <xsl:template match="marc:record">
      <xsl:choose>
         <xsl:when test="marc:datafield[@tag='100']">
            <field name="mostly_use_name">  
               <xsl:value-of  select="marc:datafield[@tag='100']/marc:subfield[@code='a']"/>
            </field>
         </xsl:when>
         <xsl:when test="marc:datafield[@tag='110']">
            <field name="mostly_use_name">  
               <xsl:value-of  select="marc:datafield[@tag='110']/marc:subfield[@code='a']"/>
            </field>
         </xsl:when>
         <xsl:when test="marc:datafield[@tag='111']">
            <field name="mostly_use_name">  
               <xsl:value-of  select="marc:datafield[@tag='111']/marc:subfield[@code='a']"/>
            </field>
         </xsl:when>
         <xsl:when test="marc:datafield[@tag='130']">
            <field name="mostly_use_name">  
               <xsl:value-of  select="marc:datafield[@tag='130']/marc:subfield[@code='a']"/>
            </field>
         </xsl:when>
         <xsl:when test="marc:datafield[@tag='150']">
            <field name="mostly_use_name">  
               <xsl:value-of  select="marc:datafield[@tag='150']"/>
            </field>
         </xsl:when>
         <xsl:when test="marc:datafield[@tag='151']">
            <field name="mostly_use_name">  
               <xsl:value-of  select="marc:datafield[@tag='151']"/>
            </field>
         </xsl:when>
      </xsl:choose> 
      <xsl:for-each select="marc:datafield[@tag='400']">
      <field name="alternatively_name">  
         <xsl:value-of select="marc:subfield[@code='a']"/>
      </field>
      </xsl:for-each>
      <xsl:for-each select="marc:datafield[@tag='410']">
         <field name="alternatively_name">  
            <xsl:value-of select="marc:subfield[@code='a']"/>
151
152
153
154
155
156
157
158
159
160
            <xsl:choose>
               <xsl:when test="marc:subfield[@code='b']">
                  <xsl:text> </xsl:text>   
                  <xsl:value-of select="marc:subfield[@code='b']"/>
               </xsl:when>
               <xsl:when test="marc:subfield[@code='g']">
                  <xsl:text> </xsl:text>   
                  <xsl:value-of select="marc:subfield[@code='g']"/>
               </xsl:when>
            </xsl:choose>
bettin5's avatar
bettin5 committed
161
162
163
164
165
166
167
168
169
170
171
172
173
174
         </field>
      </xsl:for-each>
      <xsl:for-each select="marc:datafield[@tag='430']">
         <field name="alternatively_name">  
            <xsl:value-of select="marc:subfield[@code='a']"/><xsl:text></xsl:text>   
            <xsl:value-of select="marc:subfield[@code='g']"/><xsl:text></xsl:text>   
            <xsl:value-of select="marc:subfield[@code='n']"/>
         </field>
      </xsl:for-each>
      <xsl:for-each select="marc:datafield[@tag='451']">
         <field name="alternatively_name">  
            <xsl:value-of select="marc:subfield[@code='a']"/>
         </field>
      </xsl:for-each>
175
      <xsl:for-each select="marc:datafield[@tag='034']"> 
bettin5's avatar
bettin5 committed
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
         <field name="west_cordinate">  
            <xsl:value-of select="marc:subfield[@code='e']"/>
         </field>
         <field name="east_cordinate">  
            <xsl:value-of select="marc:subfield[@code='e']"/>
         </field>
         <field name="north_cordinate">  
            <xsl:value-of select="marc:subfield[@code='f']"/>
         </field>
         <field name="south_cordinate">  
            <xsl:value-of select="marc:subfield[@code='g']"/>
         </field>
      </xsl:for-each>
   </xsl:template>

bettin5's avatar
bettin5 committed
191
192
   <xsl:template match="listBibl">
         <xsl:for-each select="bibl">
bettin5's avatar
bettin5 committed
193
            <doc>
bettin5's avatar
bettin5 committed
194
               <field name="id"> 
195
                  <xsl:value-of select="replace(./@xml:id , '_', ' ')"/>
bettin5's avatar
bettin5 committed
196
               </field>
197
               <field name="doctype">literature</field>
bettin5's avatar
bettin5 committed
198
                  <xsl:for-each select="relatedItem">
199
200
201
202
203
204
205
206
207
208
209
                    <xsl:choose>
                       <xsl:when test="ref/@target = 'https://_' or ref/@target = '' or ref/@target = '_'">
                          <field name="uri"/>
                       </xsl:when>
                       <xsl:otherwise>
                          <field name="uri">
                             <xsl:value-of select="ref/@target"/>
                          </field>
                       </xsl:otherwise>
                    </xsl:choose>         
                  </xsl:for-each>     
bettin5's avatar
bettin5 committed
210
               <xsl:for-each select="author">
211
                  <field name="literature_author"> 
bettin5's avatar
bettin5 committed
212
213
214
215
216
217
218
219
                     <xsl:value-of select="."/>
                  </field>
               </xsl:for-each>
               <xsl:for-each select="editor">
                  <field name="editor"> 
                     <xsl:value-of select="./normalize-space()"/>
                  </field>
               </xsl:for-each>
bettin5's avatar
bettin5 committed
220
221
               <field name="title"> 
                     <xsl:for-each select="title">
222
                        <xsl:value-of select="./normalize-space()"/><xsl:text></xsl:text>
bettin5's avatar
bettin5 committed
223
224
225
226
227
228
229
                     </xsl:for-each>
                  </field>
               <xsl:if test="edition!=''">
                  <field name="edition"> 
                     <xsl:value-of select="edition/normalize-space()"/>
                  </field>
               </xsl:if> 
bettin5's avatar
bettin5 committed
230
231
232
233
234
235
236
237
238
239
240
241
242
243
               <field name="pub_place"> 
                  <xsl:value-of select="pubPlace/name"/>
               </field>
               <field name="publisher">
                  <xsl:value-of select="publisher/normalize-space()"/>
               </field>               
               <xsl:if test="date/@type='pub'">
                  <field name="pub_date">
                     <xsl:value-of select="date"/>
                  </field>
               </xsl:if>
               <field name="idno">
                  <xsl:value-of select="idno"/> 
               </field>
bettin5's avatar
bettin5 committed
244
           </doc>
bettin5's avatar
bettin5 committed
245
246
247
           </xsl:for-each>
   </xsl:template>
   
Dennis Neumann's avatar
Dennis Neumann committed
248

249
250
251
252
   <xsl:template match="text()" mode="text_only">
      <xsl:variable name="currentText" select="replace(., '\s+', ' ')" />
      <xsl:choose>
         <xsl:when test="ends-with(., '-')">
Dennis Neumann's avatar
Dennis Neumann committed
253
254
            <!-- These are cases where one word is divided between two lines. 
            The minus sign is removed here, the line break (<lb/>) is removed in its own template. -->
255
256
257
            <xsl:value-of select="substring($currentText, 1, string-length($currentText)-1)" />
         </xsl:when>
         <xsl:when test="ends-with(., '&#0173;')">
Dennis Neumann's avatar
Dennis Neumann committed
258
259
260
            <!-- A soft hyphen is a convention to mark a hyphen that belongs to the word or is a hyphen on its own.
            For now, it is just replaced by a minus sign. 
            Later, it might be useful to differentiate between word divisions and hyphens. -->
261
262
263
264
265
266
            <xsl:value-of select="replace($currentText, '&#0173;', '-')" />
         </xsl:when>
         <xsl:otherwise>
            <xsl:value-of select="$currentText" />
         </xsl:otherwise>
      </xsl:choose>
Dennis Neumann's avatar
Dennis Neumann committed
267
268
269
270
271
   </xsl:template>
   
   <!--###########   Header   #######################-->
   
   <xsl:template match="teiHeader">
Dennis Neumann's avatar
Dennis Neumann committed
272
      <xsl:apply-templates select="fileDesc | profileDesc/textClass" />
Dennis Neumann's avatar
Dennis Neumann committed
273
274
275
276
277
278
279
   </xsl:template>
   
   <xsl:template match="fileDesc">
      <xsl:apply-templates select="titleStmt/title" />
      <xsl:apply-templates select="titleStmt/title[@type='desc']/name" />
      <xsl:apply-templates select="titleStmt/title[@type='desc']/date[@type='orn']" />
      <xsl:apply-templates select="titleStmt/author/name" />
280
      <xsl:apply-templates select="sourceDesc" />
Dennis Neumann's avatar
Dennis Neumann committed
281
      <xsl:apply-templates select="publicationStmt/availability/licence" />
Dennis Neumann's avatar
Dennis Neumann committed
282
283
   </xsl:template>
   
Dennis Neumann's avatar
Dennis Neumann committed
284
285
286
287
   <xsl:template match="profileDesc/textClass">
      <xsl:apply-templates select="keywords/term" />
   </xsl:template>
   
Dennis Neumann's avatar
Dennis Neumann committed
288
   <xsl:template match="title[@type='short']">
Dennis Neumann's avatar
Dennis Neumann committed
289
      <field name="short_title">
Dennis Neumann's avatar
Dennis Neumann committed
290
         <xsl:apply-templates mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
291
292
293
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
294
   <xsl:template match="title[@type='desc']">
Dennis Neumann's avatar
Dennis Neumann committed
295
      <field name="title">
Dennis Neumann's avatar
Dennis Neumann committed
296
         <xsl:apply-templates mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
297
298
299
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
300
   <xsl:template match="title/name[@type='place' and @subtype='orn']">
Dennis Neumann's avatar
Dennis Neumann committed
301
      <field name="origin_place">
Dennis Neumann's avatar
Dennis Neumann committed
302
         <xsl:apply-templates mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
303
304
305
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
306
   <xsl:template match="title/name[@type='place' and @subtype='dtn']">
Dennis Neumann's avatar
Dennis Neumann committed
307
      <field name="destination_place">
Dennis Neumann's avatar
Dennis Neumann committed
308
         <xsl:apply-templates mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
309
310
311
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
312
   <xsl:template match="title/name[@type='person' and @subtype='rcp']">
Dennis Neumann's avatar
Dennis Neumann committed
313
      <field name="recipient">
Dennis Neumann's avatar
Dennis Neumann committed
314
         <xsl:apply-templates mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
315
316
317
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
318
   <xsl:template match="title/date[@type='orn']">
Dennis Neumann's avatar
Dennis Neumann committed
319
320
321
322
323
      <field name="origin_date">
         <xsl:value-of select="@when" />
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
324
   <xsl:template match="author/name[@type='person' and @subtype='aut']">
Dennis Neumann's avatar
Dennis Neumann committed
325
      <field name="author">
Dennis Neumann's avatar
Dennis Neumann committed
326
         <xsl:apply-templates mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
327
328
      </field>
   </xsl:template>
329
330
331
332
333
334
   
   <xsl:template match="sourceDesc">
      <field name="source_description">
         <xsl:apply-templates mode="text_only" />
      </field>
   </xsl:template>
Dennis Neumann's avatar
Dennis Neumann committed
335

Dennis Neumann's avatar
Dennis Neumann committed
336
   <xsl:template match="licence">
Dennis Neumann's avatar
Dennis Neumann committed
337
      <field name="license">
Dennis Neumann's avatar
Dennis Neumann committed
338
339
340
341
         <xsl:value-of select="text()" />
      </field>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
342
343
344
345
346
347
348
349
350
351
352
353
   <xsl:template match="textClass/keywords[@scheme='#gnd']/term">
      <field name="gnd_keyword">
         <xsl:apply-templates mode="text_only" />
      </field>
   </xsl:template>

   <xsl:template match="textClass/keywords[@scheme='free']/term">
      <field name="free_keyword">
         <xsl:apply-templates mode="text_only" />
      </field>
   </xsl:template>

bettin5's avatar
bettin5 committed
354
   <!--###################   text/body   ##########################-->  
355
   
bettin5's avatar
bettin5 committed
356
357
358
359
360
361
362
363
364
365
366
367

   <xsl:template match="facsimile">
      <xsl:for-each select="./graphic/@xml:id">
         <field name="image_ids"> 
            <xsl:value-of select="."/>
         </field>
      </xsl:for-each>    
      <xsl:for-each select="./graphic/@url">
         <field name="image_urls">
            <xsl:value-of select="."/>
         </field> 
      </xsl:for-each>    
368
369
   </xsl:template> 

bettin5's avatar
bettin5 committed
370
371


Dennis Neumann's avatar
Dennis Neumann committed
372
   <xsl:template match="text">
bettin5's avatar
bettin5 committed
373
374
375
376
377
378
379
380
381
382
383
384
      <xsl:choose>
         <xsl:when test="./@xml:id"> 
            <field name="id">
               <xsl:value-of select="./@xml:id"/>
            </field> 
         </xsl:when>
         <xsl:otherwise>
            <field name="id">
               <xsl:value-of select="substring-after(substring-before(document-uri(/), '.xml'), 'Lit_Bearbeitung/')" /> <!--dirty solution not safe with other dirs-->
            </field> 
         </xsl:otherwise>
      </xsl:choose>
385
386
387
388
      <field name="doctype">article</field>
      <field name="number_of_pages">
         <xsl:value-of select="count(//pb)" />
      </field>
Dennis Neumann's avatar
Dennis Neumann committed
389
      <field name="fulltext">
Dennis Neumann's avatar
Dennis Neumann committed
390
         <xsl:apply-templates select="body" mode="text_only" />
Dennis Neumann's avatar
Dennis Neumann committed
391
      </field>
Dennis Neumann's avatar
Dennis Neumann committed
392
393
394
      <field name="fulltext_html">
         <xsl:text disable-output-escaping="yes">&lt;![CDATA[</xsl:text>
            <div class="article">
395
396
397
398
               <xsl:variable name="context" select="." />
               <xsl:for-each-group select="descendant::node()[not(node())]" group-starting-with="pb">
                  <xsl:if test="self::pb">
                     <div class="page">
Dennis Neumann's avatar
Dennis Neumann committed
399
400
401
                        <xsl:call-template name="page-beginning-with-possible-link">
                           <xsl:with-param name="current-pb" select="." />
                        </xsl:call-template>
402
403
404
405
406
407
                        <xsl:apply-templates select="$context/*" mode="page_splitting">
                           <xsl:with-param name="restricted-to" select="current-group()/ancestor-or-self::node()" tunnel="yes" />
                        </xsl:apply-templates>
                     </div>
                  </xsl:if>
               </xsl:for-each-group>
Dennis Neumann's avatar
Dennis Neumann committed
408
409
410
            </div>
         <xsl:text disable-output-escaping="yes">]]&gt;</xsl:text>
      </field>
Dennis Neumann's avatar
Dennis Neumann committed
411
      <xsl:apply-templates select=".//note[@type='com']" />
412
413
      <xsl:for-each select="//listBibl/bibl">
         <xsl:if test="relatedItem/@type='print'">
bettin5's avatar
bettin5 committed
414
            <field name="type_refs">
415
               <xsl:value-of select="./relatedItem/@type"/>
bettin5's avatar
bettin5 committed
416
            </field> 
bettin5's avatar
bettin5 committed
417
            <xsl:choose>
418
419
420
421
422
423
               <xsl:when test="./relatedItem/ref/@target">
                  <xsl:for-each select="./relatedItem/bibl/ref/@target">
                     <field name="target">
                        <xsl:value-of select="substring-after(./relatedItem/ref/@target, 'http://fbl_bln/Lit/')" />
                     </field>
                  </xsl:for-each>
bettin5's avatar
bettin5 committed
424
425
            </xsl:when>
            <xsl:when test="./relatedItem/bibl/ref/@target">
426
               <xsl:for-each select="./relatedItem/bibl/ref/@target">
bettin5's avatar
bettin5 committed
427
               <field name="target">
428
                  <xsl:value-of select="substring-after(., 'http://fbl_bln/Lit/')" />
bettin5's avatar
bettin5 committed
429
               </field>
430
               </xsl:for-each>
bettin5's avatar
bettin5 committed
431
            </xsl:when> 
bettin5's avatar
bettin5 committed
432
            </xsl:choose>                     
bettin5's avatar
bettin5 committed
433
            <field name="ref_refs">
434
435
436
437
438
439
440
               <xsl:value-of select="./relatedItem//ref/normalize-space()"/>
            </field>
            <xsl:if test="./relatedItem/bibl/text() != ''">
               <field name="bibl_refs">
                  <xsl:value-of select="./relatedItem/bibl/text()/normalize-space()"/>
               </field>
            </xsl:if>
bettin5's avatar
bettin5 committed
441
442
443
444
            <field name="literature_refs">
               <xsl:value-of select="./normalize-space()"/>
            </field> 
         </xsl:if>
445
446
447
448
449
450
451
452
      </xsl:for-each>
      <xsl:for-each select="//name/@ref ">
         <xsl:if test=". != 'gnd:_' and ./parent::name/text() and ./parent::name/@type">
            <field name="entitys_in_documents">
               <xsl:value-of select="./normalize-space()"/> 
            </field>    
         </xsl:if>
      </xsl:for-each>
453
   </xsl:template>
Dennis Neumann's avatar
Dennis Neumann committed
454
455
   
   <xsl:template match="body | div" mode="text_only">
456
      <xsl:apply-templates mode="text_only"/>
Dennis Neumann's avatar
Dennis Neumann committed
457
458
   </xsl:template>

459
   <xsl:template match="p | salute | signed | addrLine | dateline" mode="text_only">
Dennis Neumann's avatar
Dennis Neumann committed
460
461
462
      <xsl:apply-templates mode="text_only" />
      <xsl:text> </xsl:text>
   </xsl:template>
463

Dennis Neumann's avatar
Dennis Neumann committed
464
   <xsl:template match="note[@place='end' and not(@type='com')]" mode="text_only">
465
466
467
468
      <xsl:text> </xsl:text>
      <xsl:apply-templates mode="text_only" />
   </xsl:template>

469
   <xsl:template match="lb" mode="text_only">
470
471
472
473
474
475
476
477
478
479
480
481
482
483
      <xsl:variable name="precedingText" select="preceding-sibling::text()[1]" />
      <xsl:choose>
         <xsl:when test="ends-with($precedingText, '-')">
            <!-- Cases where a word is divided between two lines -->
            <!-- no output -->
         </xsl:when>
         <xsl:when test="ends-with($precedingText, '&#0173;') and not(ends-with($precedingText, ' &#0173;'))">
            <!-- Cases where the hyphen belongs to the word (Anna-<lb/>Lena) -->
            <!-- no output -->
         </xsl:when>
         <xsl:otherwise>
            <xsl:text> </xsl:text>
         </xsl:otherwise>
      </xsl:choose>
484
   </xsl:template>
Dennis Neumann's avatar
Dennis Neumann committed
485

486
487
488
489
490
491
492
493
   <xsl:template match="pb" mode="text_only">
      <xsl:text> </xsl:text>
   </xsl:template>

   <xsl:template match="space" mode="text_only">
      <xsl:text> </xsl:text>
   </xsl:template>

494
   <xsl:template match="note[@type='com']" mode="text_only">
495
   </xsl:template>
Dennis Neumann's avatar
Dennis Neumann committed
496
   
Dennis Neumann's avatar
Dennis Neumann committed
497
498
499
500
501
502
   <xsl:template match="note[@type='com']">
      <field name="note_comment">
         <xsl:apply-templates mode="text_only" />
      </field>
   </xsl:template>
   
Dennis Neumann's avatar
Dennis Neumann committed
503
   
504
   <!-- %%%%%%%%%%%%% page splitting and HTML generating %%%%%%%%%%%%%%%%%%%%%%% -->
505

Dennis Neumann's avatar
Dennis Neumann committed
506
507
508
509
   <!-- Here we start the page splitting algorithm. 
      In general, it groups all elements between two <pb/>'s and creates one page for each such group.
      Note: We use a trick with a tunnel parameter. The templates that are run from here must check 
      if they really should execute or not using that tunnel parameter. -->
510
   <xsl:template match="text" mode="page_splitting">
Dennis Neumann's avatar
Dennis Neumann committed
511
      <!-- We keep the parent of all pages (-> super-parent). -->
512
      <xsl:variable name="context" select="." />
Dennis Neumann's avatar
Dennis Neumann committed
513
514
      <!-- We make groups of only 'small' nodes, like text and <pb/>.
         We need this granularity, because a <pb/> can separate two text nodes to different pages. -->
515
516
517
      <xsl:for-each-group select="descendant::node()[not(node())]" group-starting-with="pb">
         <xsl:if test="self::pb">
            <doc>
Dennis Neumann's avatar
Dennis Neumann committed
518
               <xsl:variable name="pageNumber" select="count(preceding::pb) + 1" />
519
               <field name="id">
Dennis Neumann's avatar
Dennis Neumann committed
520
                  <xsl:value-of select="$context/@xml:id" />
521
522
523
                  <xsl:text>_page</xsl:text>
                  <xsl:value-of select="$pageNumber" />
               </field>
Dennis Neumann's avatar
Dennis Neumann committed
524
               <field name="article_id">
Dennis Neumann's avatar
Dennis Neumann committed
525
                  <xsl:value-of select="$context/@xml:id" />
526
               </field>
527
               <field name="doctype">page</field>
528
529
530
531
532
533
               <field name="page_number">
                  <xsl:value-of select="$pageNumber" />
               </field>
               <field name="html_page">
                  <xsl:text disable-output-escaping="yes">&lt;![CDATA[</xsl:text>
                  <div class="page">
Dennis Neumann's avatar
Dennis Neumann committed
534
535
536
                     <xsl:call-template name="page-beginning-with-possible-link">
                        <xsl:with-param name="current-pb" select="." />
                     </xsl:call-template>
Dennis Neumann's avatar
Dennis Neumann committed
537
                     <!-- We run the templates of all the super-parent's children... -->
538
                     <xsl:apply-templates select="$context/*" mode="page_splitting">
Dennis Neumann's avatar
Dennis Neumann committed
539
540
541
542
543
                        <!-- ..., but we use a trick with a tunnel parameter to restrict the actually executed templates 
                           to the ones relevant for the current page. 
                           We choose the grouped nodes, as well as all their ancestors.
                           We need all the ancestors to construct correctly opened and closed elements 
                           that otherwise would be cut in two by the <pb/>. -->
544
545
546
547
548
                        <xsl:with-param name="restricted-to" select="current-group()/ancestor-or-self::node()" tunnel="yes" />
                     </xsl:apply-templates>
                  </div>
                  <xsl:text disable-output-escaping="yes">]]&gt;</xsl:text>
               </field>
Dennis Neumann's avatar
Dennis Neumann committed
549
550
551
552
553
               <xsl:for-each select="distinct-values(//@xml:lang)">
                  <field name="language">
                     <xsl:value-of select="." />
                  </field>
               </xsl:for-each>
554
               <xsl:if test="./@facs">
bettin5's avatar
bettin5 committed
555
                  <field name="image_id"> 
556
557
558
                     <xsl:value-of select="./@facs"/>
                  </field>
               </xsl:if>
559
560
561
562
563
            </doc>
         </xsl:if>
      </xsl:for-each-group>
   </xsl:template>

Dennis Neumann's avatar
Dennis Neumann committed
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
   <xsl:template name="page-beginning-with-possible-link">
      <xsl:param name="current-pb" />
      <div class="page-beginning">
         <xsl:variable name="facsId" select="substring($current-pb/@facs, 2, string-length($current-pb/@facs))" />
         <xsl:variable name="graphicUrl" select="id($facsId)/@url" />
         <xsl:variable name="graphicUrlWithoutJpg">
            <xsl:value-of select="if (ends-with($graphicUrl, '.jpg')) then substring($graphicUrl, 1, string-length($graphicUrl)-4) else $graphicUrl" />
         </xsl:variable>
         <xsl:choose>
            <xsl:when test="$current-pb/@n ne '' and $graphicUrlWithoutJpg">
               <a href="{concat('/', $graphicUrlWithoutJpg)}" target="_blank">
                  <xsl:value-of select="$current-pb/@n" />
               </a>
            </xsl:when>
            <xsl:when test="$current-pb/@n ne ''">
               <xsl:value-of select="$current-pb/@n" />
            </xsl:when>
            <xsl:otherwise>
               <xsl:text> </xsl:text>
            </xsl:otherwise>
         </xsl:choose>
      </div>
   </xsl:template>

588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
   <xsl:template match="pb" mode="page_splitting" />

   <xsl:template match="*" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <xsl:message>
            <xsl:text>Unknown element &lt;</xsl:text>
            <xsl:value-of select="local-name()" />
            <xsl:if test="@rendition">
               <xsl:text> rendition="</xsl:text>
               <xsl:value-of select="@rendition" />
               <xsl:text>"</xsl:text>
            </xsl:if>
            <xsl:if test="@type">
               <xsl:text> type="</xsl:text>
               <xsl:value-of select="@type" />
               <xsl:text>"</xsl:text>
            </xsl:if>
            <xsl:text>&gt; - first occurrence: </xsl:text>
         </xsl:message>
         <span class="unknown-element">
            <xsl:apply-templates mode="page_splitting" />
         </span>
      </xsl:if>
   </xsl:template>

   <xsl:template match="body | div | front" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <xsl:apply-templates mode="page_splitting" />
      </xsl:if>
   </xsl:template>

   <xsl:template match="p | opener | salute | seg | bibl | closer | signed | dateline | date 
   | label[not(@rendition)] | choice | abbr | expan | postscript" mode="page_splitting">
623
624
625
626
627
628
629
630
631
632
633
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="{local-name(.)}">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="text()" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
         <xsl:variable name="currentText" select="replace(., '\s+', ' ')" />
         <xsl:choose>
            <xsl:when test="ends-with(., '&#0173;')">
               <xsl:value-of select="replace($currentText, '&#0173;', '-')" />
            </xsl:when>
            <xsl:otherwise>
               <xsl:value-of select="$currentText" />
            </xsl:otherwise>
         </xsl:choose>
      </xsl:if>
   </xsl:template>

   <xsl:template match="lb" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <br />
650
651
652
      </xsl:if>
   </xsl:template>

653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
   <xsl:template match="space[@unit='lines']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <xsl:variable name="emptyLines" select="@quantity" />
         <xsl:choose>
            <xsl:when test="$emptyLines castable as xs:integer">
               <xsl:for-each select="1 to $emptyLines">
                  <br />
               </xsl:for-each>
            </xsl:when>
            <xsl:otherwise>
               <br />
            </xsl:otherwise>
         </xsl:choose>
      </xsl:if>
   </xsl:template>

   <xsl:template match="name[@type='place']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="place">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="name[@type='org']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="org">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="name[@type='person']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="person">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="name[@type='object']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="object">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="rs[@type='person']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="rs-person">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="rs[@type='place']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="rs-place">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="note[@type='com']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="note-comment">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="note[@type='footnote']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="note-footnote">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="note[not(@type)]" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="note">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:underline']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="underline">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:doubleunderline']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="doubleunderline">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:superscript']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <sup>
            <xsl:apply-templates mode="page_splitting" />
         </sup>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:subscript']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <sub>
            <xsl:apply-templates mode="page_splitting" />
         </sub>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:italic']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="italic">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:letterspace']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="letterspace">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="hi[@rendition='simple:right']" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <div class="right">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>

   <xsl:template match="ref[@target]" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <a href="{@target}">
            <xsl:apply-templates mode="page_splitting" />
         </a>
      </xsl:if>
   </xsl:template>

   <xsl:template match="head[@rendition] | label[@rendition]" mode="page_splitting">
      <xsl:param name="restricted-to" tunnel="yes" />
      <xsl:if test="exists(. intersect $restricted-to)">
         <xsl:variable name="classNames" select="local-name(), substring-after(@rendition, 'simple:')" />
         <div class="{$classNames}">
            <xsl:apply-templates mode="page_splitting" />
         </div>
      </xsl:if>
   </xsl:template>
832

Dennis Neumann's avatar
Dennis Neumann committed
833
</xsl:stylesheet>